From e1ed0b987e68b8306cded58348e948fcb000537a Mon Sep 17 00:00:00 2001 From: smilesun Date: Fri, 27 Sep 2024 16:12:54 +0200 Subject: [PATCH] fix irl doc --- docs/build/html/.buildinfo | 2 +- docs/build/html/_modules/domainlab.html | 12 +- .../domainlab/algos/a_algo_builder.html | 12 +- .../domainlab/algos/builder_api_model.html | 12 +- .../domainlab/algos/builder_custom.html | 12 +- .../domainlab/algos/builder_dann.html | 12 +- .../domainlab/algos/builder_diva.html | 12 +- .../_modules/domainlab/algos/builder_erm.html | 12 +- .../domainlab/algos/builder_hduva.html | 12 +- .../domainlab/algos/builder_jigen1.html | 12 +- .../domainlab/algos/msels/a_model_sel.html | 12 +- .../domainlab/algos/msels/c_msel_oracle.html | 12 +- .../domainlab/algos/msels/c_msel_tr_loss.html | 12 +- .../domainlab/algos/msels/c_msel_val.html | 12 +- .../domainlab/algos/observers/a_observer.html | 12 +- .../algos/observers/b_obvisitor.html | 12 +- .../algos/observers/c_obvisitor_cleanup.html | 12 +- .../algos/observers/c_obvisitor_gen.html | 12 +- .../domainlab/algos/trainers/a_trainer.html | 12 +- .../domainlab/algos/trainers/args_dial.html | 12 +- .../algos/trainers/compos/matchdg_args.html | 12 +- .../algos/trainers/compos/matchdg_match.html | 12 +- .../algos/trainers/compos/matchdg_utils.html | 12 +- .../algos/trainers/hyper_scheduler.html | 12 +- .../domainlab/algos/trainers/mmd_base.html | 12 +- .../domainlab/algos/trainers/train_basic.html | 12 +- .../algos/trainers/train_causIRL.html | 12 +- .../domainlab/algos/trainers/train_coral.html | 12 +- .../domainlab/algos/trainers/train_dial.html | 12 +- .../domainlab/algos/trainers/train_ema.html | 12 +- .../domainlab/algos/trainers/train_fishr.html | 12 +- .../algos/trainers/train_hyper_scheduler.html | 12 +- .../domainlab/algos/trainers/train_irm.html | 12 +- .../algos/trainers/train_matchdg.html | 12 +- .../domainlab/algos/trainers/train_mldg.html | 12 +- .../domainlab/algos/trainers/zoo_trainer.html | 12 +- .../_modules/domainlab/algos/zoo_algos.html | 12 +- .../html/_modules/domainlab/arg_parser.html | 12 +- .../domainlab/compos/a_nn_builder.html | 12 +- .../domainlab/compos/builder_nn_alex.html | 12 +- .../compos/builder_nn_conv_bn_relu_2.html | 12 +- .../compos/builder_nn_external_from_file.html | 12 +- .../compos/nn_zoo/net_adversarial.html | 12 +- .../domainlab/compos/nn_zoo/net_classif.html | 12 +- .../nn_zoo/net_conv_conv_bn_pool_2.html | 12 +- .../domainlab/compos/nn_zoo/net_gated.html | 12 +- .../_modules/domainlab/compos/nn_zoo/nn.html | 12 +- .../domainlab/compos/nn_zoo/nn_alex.html | 12 +- .../compos/nn_zoo/nn_torchvision.html | 12 +- .../domainlab/compos/pcr/p_chain_handler.html | 12 +- .../domainlab/compos/pcr/request.html | 12 +- .../compos/utils_conv_get_flat_dim.html | 12 +- .../domainlab/compos/vae/a_model_builder.html | 12 +- .../domainlab/compos/vae/a_vae_builder.html | 12 +- .../compos/vae/c_vae_adaptor_model_recon.html | 12 +- .../compos/vae/c_vae_builder_classif.html | 12 +- .../domainlab/compos/vae/c_vae_recon.html | 12 +- .../decoder_concat_vec_reshape_conv.html | 12 +- ...er_concat_vec_reshape_conv_gated_conv.html | 12 +- .../compos/vae/compos/decoder_cond_prior.html | 12 +- .../compos/vae/compos/decoder_losses.html | 12 +- .../domainlab/compos/vae/compos/encoder.html | 12 +- .../compos/vae/compos/encoder_dirichlet.html | 12 +- .../vae/compos/encoder_domain_topic.html | 12 +- .../encoder_domain_topic_img2topic.html | 12 +- .../encoder_domain_topic_img_topic2zd.html | 12 +- .../vae/compos/encoder_xyd_parallel.html | 12 +- .../vae/compos/encoder_xydt_elevator.html | 12 +- .../compos/vae/compos/encoder_zy.html | 12 +- .../vae/utils_request_chain_builder.html | 12 +- .../compos/vae/zoo_vae_builders_classif.html | 12 +- .../vae/zoo_vae_builders_classif_topic.html | 12 +- .../_modules/domainlab/compos/zoo_nn.html | 12 +- .../dsets/a_dset_mnist_color_rgb_solo.html | 12 +- .../domainlab/dsets/dset_img_path_list.html | 12 +- .../dsets/dset_mnist_color_solo_default.html | 12 +- ...dset_poly_domains_mnist_color_default.html | 12 +- .../domainlab/dsets/dset_subfolder.html | 12 +- .../_modules/domainlab/dsets/utils_data.html | 12 +- .../dsets/utils_wrapdset_patches.html | 12 +- .../_modules/domainlab/exp/exp_cuda_seed.html | 12 +- .../html/_modules/domainlab/exp/exp_main.html | 513 ------------ .../_modules/domainlab/exp/exp_utils.html | 668 --------------- .../exp_protocol/aggregate_results.html | 12 +- .../exp_protocol/run_experiment.html | 565 ------------- .../build/html/_modules/domainlab/mk_exp.html | 411 --------- .../_modules/domainlab/models/a_model.html | 12 +- .../domainlab/models/a_model_classif.html | 12 +- .../_modules/domainlab/models/args_jigen.html | 12 +- .../_modules/domainlab/models/args_vae.html | 12 +- .../domainlab/models/interface_vae_xyd.html | 12 +- .../domainlab/models/model_custom.html | 12 +- .../_modules/domainlab/models/model_dann.html | 12 +- .../_modules/domainlab/models/model_diva.html | 12 +- .../domainlab/models/model_hduva.html | 12 +- .../domainlab/models/model_jigen.html | 12 +- .../models/model_vae_xyd_classif.html | 12 +- .../html/_modules/domainlab/tasks/a_task.html | 12 +- .../domainlab/tasks/a_task_classif.html | 12 +- .../html/_modules/domainlab/tasks/b_task.html | 12 +- .../domainlab/tasks/b_task_classif.html | 12 +- .../_modules/domainlab/tasks/task_dset.html | 12 +- .../_modules/domainlab/tasks/task_folder.html | 12 +- .../domainlab/tasks/task_folder_mk.html | 12 +- .../domainlab/tasks/task_mini_vlcs.html | 12 +- .../domainlab/tasks/task_mnist_color.html | 12 +- .../domainlab/tasks/task_pathlist.html | 12 +- .../_modules/domainlab/tasks/task_utils.html | 12 +- .../_modules/domainlab/tasks/utils_task.html | 12 +- .../domainlab/tasks/utils_task_dset.html | 12 +- .../_modules/domainlab/tasks/zoo_tasks.html | 12 +- .../domainlab/utils/flows_gen_img_model.html | 12 +- .../utils/generate_benchmark_plots.html | 12 +- .../_modules/domainlab/utils/get_git_tag.html | 12 +- .../utils/hyperparameter_gridsearch.html | 12 +- .../utils/hyperparameter_sampling.html | 12 +- .../html/_modules/domainlab/utils/logger.html | 12 +- .../domainlab/utils/override_interface.html | 12 +- .../html/_modules/domainlab/utils/perf.html | 12 +- .../domainlab/utils/perf_metrics.html | 12 +- .../domainlab/utils/sanity_check.html | 12 +- .../_modules/domainlab/utils/test_img.html | 12 +- .../_modules/domainlab/utils/u_import.html | 12 +- .../domainlab/utils/u_import_net_module.html | 12 +- .../_modules/domainlab/utils/utils_class.html | 12 +- .../domainlab/utils/utils_classif.html | 12 +- .../_modules/domainlab/utils/utils_cuda.html | 12 +- .../domainlab/utils/utils_img_sav.html | 12 +- docs/build/html/_modules/index.html | 17 +- docs/build/html/_static/basic.css | 54 +- docs/build/html/_static/doctools.js | 450 +++++----- .../html/_static/documentation_options.js | 6 +- docs/build/html/_static/language_data.js | 100 ++- docs/build/html/_static/pygments.css | 1 - docs/build/html/_static/searchtools.js | 788 +++++++++--------- docs/build/html/docDIAL.html | 30 +- docs/build/html/docFishr.html | 26 +- docs/build/html/docHDUVA.html | 44 +- docs/build/html/docIRM.html | 20 +- docs/build/html/docJiGen.html | 24 +- docs/build/html/docMA.html | 16 +- docs/build/html/docMatchDG.html | 34 +- docs/build/html/doc_MNIST_classification.html | 28 +- docs/build/html/doc_benchmark.html | 42 +- .../doc_benchmark_further_explanation.html | 20 +- docs/build/html/doc_benchmark_pacs.html | 16 +- docs/build/html/doc_benchmark_yaml.html | 32 +- docs/build/html/doc_coral.html | 20 +- docs/build/html/doc_custom_nn.html | 30 +- docs/build/html/doc_dann.html | 18 +- docs/build/html/doc_diva.html | 34 +- docs/build/html/doc_extend_contribute.html | 18 +- docs/build/html/doc_install.html | 24 +- docs/build/html/doc_irl.html | 37 +- docs/build/html/doc_mldg.html | 20 +- docs/build/html/doc_model.html | 26 +- docs/build/html/doc_output.html | 16 +- docs/build/html/doc_tasks.html | 48 +- docs/build/html/doc_trainer.html | 20 +- docs/build/html/doc_usage_cmd.html | 40 +- docs/build/html/domainlab.algos.html | 60 +- docs/build/html/domainlab.algos.msels.html | 34 +- .../build/html/domainlab.algos.observers.html | 34 +- .../html/domainlab.algos.trainers.compos.html | 30 +- docs/build/html/domainlab.algos.trainers.html | 98 +-- docs/build/html/domainlab.compos.html | 38 +- docs/build/html/domainlab.compos.nn_zoo.html | 74 +- docs/build/html/domainlab.compos.pcr.html | 28 +- .../html/domainlab.compos.vae.compos.html | 88 +- docs/build/html/domainlab.compos.vae.html | 76 +- docs/build/html/domainlab.dsets.html | 56 +- docs/build/html/domainlab.exp.html | 174 +--- docs/build/html/domainlab.exp_protocol.html | 71 +- docs/build/html/domainlab.html | 75 +- docs/build/html/domainlab.models.html | 98 +-- docs/build/html/domainlab.tasks.html | 94 +-- docs/build/html/domainlab.utils.html | 94 +-- docs/build/html/genindex.html | 141 +--- docs/build/html/index.html | 26 +- docs/build/html/modules.html | 26 +- docs/build/html/objects.inv | Bin 28469 -> 27836 bytes docs/build/html/py-modindex.html | 37 +- docs/build/html/readme_link.html | 40 +- docs/build/html/requirements.html | 16 +- docs/build/html/search.html | 12 +- docs/build/html/searchindex.js | 2 +- docs/build/html/tips.html | 14 +- 187 files changed, 1757 insertions(+), 5513 deletions(-) delete mode 100644 docs/build/html/_modules/domainlab/exp/exp_main.html delete mode 100644 docs/build/html/_modules/domainlab/exp/exp_utils.html delete mode 100644 docs/build/html/_modules/domainlab/exp_protocol/run_experiment.html delete mode 100644 docs/build/html/_modules/domainlab/mk_exp.html diff --git a/docs/build/html/.buildinfo b/docs/build/html/.buildinfo index ae7a184ef..7ce915873 100644 --- a/docs/build/html/.buildinfo +++ b/docs/build/html/.buildinfo @@ -1,4 +1,4 @@ # Sphinx build info version 1 # This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done. -config: 6b1f707754558cc1c68f56bd9c949897 +config: fe17894255414d0052da08104b737402 tags: 645f666f9bcd5a90fca523b33c5a78b7 diff --git a/docs/build/html/_modules/domainlab.html b/docs/build/html/_modules/domainlab.html index 33aada708..b1a9ed042 100644 --- a/docs/build/html/_modules/domainlab.html +++ b/docs/build/html/_modules/domainlab.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -386,7 +378,7 @@

    Source code for domainlab

    Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/a_algo_builder.html b/docs/build/html/_modules/domainlab/algos/a_algo_builder.html index 77488d34a..ffbd152cc 100644 --- a/docs/build/html/_modules/domainlab/algos/a_algo_builder.html +++ b/docs/build/html/_modules/domainlab/algos/a_algo_builder.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -428,7 +420,7 @@

    Source code for domai

    Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/builder_api_model.html b/docs/build/html/_modules/domainlab/algos/builder_api_model.html index cd1b90c84..068339bed 100644 --- a/docs/build/html/_modules/domainlab/algos/builder_api_model.html +++ b/docs/build/html/_modules/domainlab/algos/builder_api_model.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -390,7 +382,7 @@

    Source code for do Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/builder_custom.html b/docs/build/html/_modules/domainlab/algos/builder_custom.html index dfc01a364..814a2d518 100644 --- a/docs/build/html/_modules/domainlab/algos/builder_custom.html +++ b/docs/build/html/_modules/domainlab/algos/builder_custom.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -481,7 +473,7 @@

    Source code for domai Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/builder_dann.html b/docs/build/html/_modules/domainlab/algos/builder_dann.html index 397cd2282..036d0cdda 100644 --- a/docs/build/html/_modules/domainlab/algos/builder_dann.html +++ b/docs/build/html/_modules/domainlab/algos/builder_dann.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -454,7 +446,7 @@

    Source code for domainl Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/builder_diva.html b/docs/build/html/_modules/domainlab/algos/builder_diva.html index 1c8dcd68c..7b6a204e5 100644 --- a/docs/build/html/_modules/domainlab/algos/builder_diva.html +++ b/docs/build/html/_modules/domainlab/algos/builder_diva.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -424,7 +416,7 @@

    Source code for domainl Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/builder_erm.html b/docs/build/html/_modules/domainlab/algos/builder_erm.html index cd03f5314..723ccc306 100644 --- a/docs/build/html/_modules/domainlab/algos/builder_erm.html +++ b/docs/build/html/_modules/domainlab/algos/builder_erm.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -415,7 +407,7 @@

    Source code for domainla Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/builder_hduva.html b/docs/build/html/_modules/domainlab/algos/builder_hduva.html index 4b61b668c..0b6d76e5a 100644 --- a/docs/build/html/_modules/domainlab/algos/builder_hduva.html +++ b/docs/build/html/_modules/domainlab/algos/builder_hduva.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -414,7 +406,7 @@

    Source code for domain Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/builder_jigen1.html b/docs/build/html/_modules/domainlab/algos/builder_jigen1.html index 2b065b192..39b43c590 100644 --- a/docs/build/html/_modules/domainlab/algos/builder_jigen1.html +++ b/docs/build/html/_modules/domainlab/algos/builder_jigen1.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -440,7 +432,7 @@

    Source code for domai Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/msels/a_model_sel.html b/docs/build/html/_modules/domainlab/algos/msels/a_model_sel.html index b208d1dcb..7e4ec7155 100644 --- a/docs/build/html/_modules/domainlab/algos/msels/a_model_sel.html +++ b/docs/build/html/_modules/domainlab/algos/msels/a_model_sel.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -505,7 +497,7 @@

    Source code for do Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/msels/c_msel_oracle.html b/docs/build/html/_modules/domainlab/algos/msels/c_msel_oracle.html index eb94ee946..f91f16dac 100644 --- a/docs/build/html/_modules/domainlab/algos/msels/c_msel_oracle.html +++ b/docs/build/html/_modules/domainlab/algos/msels/c_msel_oracle.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -432,7 +424,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/msels/c_msel_tr_loss.html b/docs/build/html/_modules/domainlab/algos/msels/c_msel_tr_loss.html index 6599b9cf0..d5fa1996b 100644 --- a/docs/build/html/_modules/domainlab/algos/msels/c_msel_tr_loss.html +++ b/docs/build/html/_modules/domainlab/algos/msels/c_msel_tr_loss.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -421,7 +413,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/msels/c_msel_val.html b/docs/build/html/_modules/domainlab/algos/msels/c_msel_val.html index 8c92037ce..81b479922 100644 --- a/docs/build/html/_modules/domainlab/algos/msels/c_msel_val.html +++ b/docs/build/html/_modules/domainlab/algos/msels/c_msel_val.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -441,7 +433,7 @@

    Source code for dom Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/observers/a_observer.html b/docs/build/html/_modules/domainlab/algos/observers/a_observer.html index d50b8b345..1ec2267c8 100644 --- a/docs/build/html/_modules/domainlab/algos/observers/a_observer.html +++ b/docs/build/html/_modules/domainlab/algos/observers/a_observer.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -426,7 +418,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/observers/b_obvisitor.html b/docs/build/html/_modules/domainlab/algos/observers/b_obvisitor.html index b08de9f19..f00ceb504 100644 --- a/docs/build/html/_modules/domainlab/algos/observers/b_obvisitor.html +++ b/docs/build/html/_modules/domainlab/algos/observers/b_obvisitor.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -524,7 +516,7 @@

    Source code fo Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/observers/c_obvisitor_cleanup.html b/docs/build/html/_modules/domainlab/algos/observers/c_obvisitor_cleanup.html index e20e84305..5e8c20c15 100644 --- a/docs/build/html/_modules/domainlab/algos/observers/c_obvisitor_cleanup.html +++ b/docs/build/html/_modules/domainlab/algos/observers/c_obvisitor_cleanup.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -392,7 +384,7 @@

    Source Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/observers/c_obvisitor_gen.html b/docs/build/html/_modules/domainlab/algos/observers/c_obvisitor_gen.html index 7a736be81..8c86e6398 100644 --- a/docs/build/html/_modules/domainlab/algos/observers/c_obvisitor_gen.html +++ b/docs/build/html/_modules/domainlab/algos/observers/c_obvisitor_gen.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -410,7 +402,7 @@

    Source cod Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/a_trainer.html b/docs/build/html/_modules/domainlab/algos/trainers/a_trainer.html index 8e1ebc764..ef3958703 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/a_trainer.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/a_trainer.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -671,7 +663,7 @@

    Source code for d Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/args_dial.html b/docs/build/html/_modules/domainlab/algos/trainers/args_dial.html index c11a3e045..ffd0f78c2 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/args_dial.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/args_dial.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -398,7 +390,7 @@

    Source code for d Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_args.html b/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_args.html index 136056849..b386f74a5 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_args.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_args.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -389,7 +381,7 @@

    Source Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_match.html b/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_match.html index ecdb23f6b..ee71ef5bc 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_match.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_match.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -765,7 +757,7 @@

    Source Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_utils.html b/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_utils.html index e9c9fc729..24bf5b185 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_utils.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/compos/matchdg_utils.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -516,7 +508,7 @@

    Source Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/hyper_scheduler.html b/docs/build/html/_modules/domainlab/algos/trainers/hyper_scheduler.html index 08081bc3e..4736003fa 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/hyper_scheduler.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/hyper_scheduler.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -427,7 +419,7 @@

    Source code Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/mmd_base.html b/docs/build/html/_modules/domainlab/algos/trainers/mmd_base.html index 2156f28ef..7ead11e39 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/mmd_base.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/mmd_base.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -412,7 +404,7 @@

    Source code for do Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_basic.html b/docs/build/html/_modules/domainlab/algos/trainers/train_basic.html index 1d417dceb..1b492cc5f 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_basic.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_basic.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -479,7 +471,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_causIRL.html b/docs/build/html/_modules/domainlab/algos/trainers/train_causIRL.html index 1f484ff7f..3f72504cd 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_causIRL.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_causIRL.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -441,7 +433,7 @@

    Source code f Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_coral.html b/docs/build/html/_modules/domainlab/algos/trainers/train_coral.html index 773bba964..3e8ffb96a 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_coral.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_coral.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -413,7 +405,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_dial.html b/docs/build/html/_modules/domainlab/algos/trainers/train_dial.html index 3fc64d665..9982e47ba 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_dial.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_dial.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -417,7 +409,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_ema.html b/docs/build/html/_modules/domainlab/algos/trainers/train_ema.html index dc5ef16f2..2bfbe0487 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_ema.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_ema.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -430,7 +422,7 @@

    Source code for d Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_fishr.html b/docs/build/html/_modules/domainlab/algos/trainers/train_fishr.html index dffea521f..22e34e3ef 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_fishr.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_fishr.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -536,7 +528,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_hyper_scheduler.html b/docs/build/html/_modules/domainlab/algos/trainers/train_hyper_scheduler.html index d0b42c797..f2922aea6 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_hyper_scheduler.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_hyper_scheduler.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -429,7 +421,7 @@

    Sourc Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_irm.html b/docs/build/html/_modules/domainlab/algos/trainers/train_irm.html index 11dacb741..395cd1509 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_irm.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_irm.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -435,7 +427,7 @@

    Source code for d Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_matchdg.html b/docs/build/html/_modules/domainlab/algos/trainers/train_matchdg.html index 06ffaf9c2..19a94feb3 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_matchdg.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_matchdg.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -736,7 +728,7 @@

    Source code f Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/train_mldg.html b/docs/build/html/_modules/domainlab/algos/trainers/train_mldg.html index c38403aad..66464de6c 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/train_mldg.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/train_mldg.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -486,7 +478,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/trainers/zoo_trainer.html b/docs/build/html/_modules/domainlab/algos/trainers/zoo_trainer.html index 704cb75fd..e40af6c02 100644 --- a/docs/build/html/_modules/domainlab/algos/trainers/zoo_trainer.html +++ b/docs/build/html/_modules/domainlab/algos/trainers/zoo_trainer.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -433,7 +425,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/algos/zoo_algos.html b/docs/build/html/_modules/domainlab/algos/zoo_algos.html index 08ac2a5d3..0f66b7428 100644 --- a/docs/build/html/_modules/domainlab/algos/zoo_algos.html +++ b/docs/build/html/_modules/domainlab/algos/zoo_algos.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -420,7 +412,7 @@

    Source code for domainlab. Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/arg_parser.html b/docs/build/html/_modules/domainlab/arg_parser.html index 6bb029e73..842b82700 100644 --- a/docs/build/html/_modules/domainlab/arg_parser.html +++ b/docs/build/html/_modules/domainlab/arg_parser.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -777,7 +769,7 @@

    Source code for domainlab.arg_p Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/a_nn_builder.html b/docs/build/html/_modules/domainlab/compos/a_nn_builder.html index e51840ffc..299694ad2 100644 --- a/docs/build/html/_modules/domainlab/compos/a_nn_builder.html +++ b/docs/build/html/_modules/domainlab/compos/a_nn_builder.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -416,7 +408,7 @@

    Source code for domain Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/builder_nn_alex.html b/docs/build/html/_modules/domainlab/compos/builder_nn_alex.html index 8aa2eda69..82d4f487c 100644 --- a/docs/build/html/_modules/domainlab/compos/builder_nn_alex.html +++ b/docs/build/html/_modules/domainlab/compos/builder_nn_alex.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -400,7 +392,7 @@

    Source code for dom Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/builder_nn_conv_bn_relu_2.html b/docs/build/html/_modules/domainlab/compos/builder_nn_conv_bn_relu_2.html index 4afe9cc36..8cb7ffe8b 100644 --- a/docs/build/html/_modules/domainlab/compos/builder_nn_conv_bn_relu_2.html +++ b/docs/build/html/_modules/domainlab/compos/builder_nn_conv_bn_relu_2.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -403,7 +395,7 @@

    Source co Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/builder_nn_external_from_file.html b/docs/build/html/_modules/domainlab/compos/builder_nn_external_from_file.html index 3acdc690b..ca8fc8d57 100644 --- a/docs/build/html/_modules/domainlab/compos/builder_nn_external_from_file.html +++ b/docs/build/html/_modules/domainlab/compos/builder_nn_external_from_file.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -406,7 +398,7 @@

    Sourc Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/nn_zoo/net_adversarial.html b/docs/build/html/_modules/domainlab/compos/nn_zoo/net_adversarial.html index 901793af9..066352d07 100644 --- a/docs/build/html/_modules/domainlab/compos/nn_zoo/net_adversarial.html +++ b/docs/build/html/_modules/domainlab/compos/nn_zoo/net_adversarial.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -401,7 +393,7 @@

    Source code Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/nn_zoo/net_classif.html b/docs/build/html/_modules/domainlab/compos/nn_zoo/net_classif.html index a1402492b..c788cd208 100644 --- a/docs/build/html/_modules/domainlab/compos/nn_zoo/net_classif.html +++ b/docs/build/html/_modules/domainlab/compos/nn_zoo/net_classif.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -391,7 +383,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/nn_zoo/net_conv_conv_bn_pool_2.html b/docs/build/html/_modules/domainlab/compos/nn_zoo/net_conv_conv_bn_pool_2.html index 7c85c61ed..dd5641b0f 100644 --- a/docs/build/html/_modules/domainlab/compos/nn_zoo/net_conv_conv_bn_pool_2.html +++ b/docs/build/html/_modules/domainlab/compos/nn_zoo/net_conv_conv_bn_pool_2.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -452,7 +444,7 @@

    Sour Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/nn_zoo/net_gated.html b/docs/build/html/_modules/domainlab/compos/nn_zoo/net_gated.html index d99b70cea..b5542778d 100644 --- a/docs/build/html/_modules/domainlab/compos/nn_zoo/net_gated.html +++ b/docs/build/html/_modules/domainlab/compos/nn_zoo/net_gated.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -454,7 +446,7 @@

    Source code for do Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/nn_zoo/nn.html b/docs/build/html/_modules/domainlab/compos/nn_zoo/nn.html index aa8d1aa93..0873d5bed 100644 --- a/docs/build/html/_modules/domainlab/compos/nn_zoo/nn.html +++ b/docs/build/html/_modules/domainlab/compos/nn_zoo/nn.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -412,7 +404,7 @@

    Source code for domainlab Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/nn_zoo/nn_alex.html b/docs/build/html/_modules/domainlab/compos/nn_zoo/nn_alex.html index f705bf18b..396381278 100644 --- a/docs/build/html/_modules/domainlab/compos/nn_zoo/nn_alex.html +++ b/docs/build/html/_modules/domainlab/compos/nn_zoo/nn_alex.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -437,7 +429,7 @@

    Source code for doma Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/nn_zoo/nn_torchvision.html b/docs/build/html/_modules/domainlab/compos/nn_zoo/nn_torchvision.html index 441313003..1bafed913 100644 --- a/docs/build/html/_modules/domainlab/compos/nn_zoo/nn_torchvision.html +++ b/docs/build/html/_modules/domainlab/compos/nn_zoo/nn_torchvision.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -397,7 +389,7 @@

    Source code f Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/pcr/p_chain_handler.html b/docs/build/html/_modules/domainlab/compos/pcr/p_chain_handler.html index 9c9bc9d81..4db38fdb0 100644 --- a/docs/build/html/_modules/domainlab/compos/pcr/p_chain_handler.html +++ b/docs/build/html/_modules/domainlab/compos/pcr/p_chain_handler.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -497,7 +489,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/pcr/request.html b/docs/build/html/_modules/domainlab/compos/pcr/request.html index a0bb2bddf..952fb7f6c 100644 --- a/docs/build/html/_modules/domainlab/compos/pcr/request.html +++ b/docs/build/html/_modules/domainlab/compos/pcr/request.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -406,7 +398,7 @@

    Source code for domainl Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/utils_conv_get_flat_dim.html b/docs/build/html/_modules/domainlab/compos/utils_conv_get_flat_dim.html index 4369490fb..30c3c4d88 100644 --- a/docs/build/html/_modules/domainlab/compos/utils_conv_get_flat_dim.html +++ b/docs/build/html/_modules/domainlab/compos/utils_conv_get_flat_dim.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -384,7 +376,7 @@

    Source code Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/a_model_builder.html b/docs/build/html/_modules/domainlab/compos/vae/a_model_builder.html index dc4ce4e5c..eb516cf4c 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/a_model_builder.html +++ b/docs/build/html/_modules/domainlab/compos/vae/a_model_builder.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -388,7 +380,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/a_vae_builder.html b/docs/build/html/_modules/domainlab/compos/vae/a_vae_builder.html index 2755cb044..2e6596f3b 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/a_vae_builder.html +++ b/docs/build/html/_modules/domainlab/compos/vae/a_vae_builder.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -409,7 +401,7 @@

    Source code for d Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/c_vae_adaptor_model_recon.html b/docs/build/html/_modules/domainlab/compos/vae/c_vae_adaptor_model_recon.html index 172a81f4b..5f213d327 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/c_vae_adaptor_model_recon.html +++ b/docs/build/html/_modules/domainlab/compos/vae/c_vae_adaptor_model_recon.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -415,7 +407,7 @@

    Sourc Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/c_vae_builder_classif.html b/docs/build/html/_modules/domainlab/compos/vae/c_vae_builder_classif.html index 85384d4b3..d38299dfc 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/c_vae_builder_classif.html +++ b/docs/build/html/_modules/domainlab/compos/vae/c_vae_builder_classif.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -406,7 +398,7 @@

    Source co Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/c_vae_recon.html b/docs/build/html/_modules/domainlab/compos/vae/c_vae_recon.html index 91bf67965..a46d1ae9d 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/c_vae_recon.html +++ b/docs/build/html/_modules/domainlab/compos/vae/c_vae_recon.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -475,7 +467,7 @@

    Source code for dom Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_concat_vec_reshape_conv.html b/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_concat_vec_reshape_conv.html index dd32567e1..6a771c4fe 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_concat_vec_reshape_conv.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_concat_vec_reshape_conv.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -438,7 +430,7 @@

    Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_concat_vec_reshape_conv_gated_conv.html b/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_concat_vec_reshape_conv_gated_conv.html index 7171d0dc1..77e37dcdf 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_concat_vec_reshape_conv_gated_conv.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_concat_vec_reshape_conv_gated_conv.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -429,7 +421,7 @@

    Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_cond_prior.html b/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_cond_prior.html index 829afb17c..27d4051e5 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_cond_prior.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_cond_prior.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -403,7 +395,7 @@

    Sourc Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_losses.html b/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_losses.html index 1049202b6..aa0cc6474 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_losses.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/decoder_losses.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -410,7 +402,7 @@

    Source co Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder.html b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder.html index 0c5c1692d..cae92810f 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -458,7 +450,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_dirichlet.html b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_dirichlet.html index 70eb148a5..2edb7e5ad 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_dirichlet.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_dirichlet.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -391,7 +383,7 @@

    Source Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic.html b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic.html index 35a822c63..0598161d4 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -422,7 +414,7 @@

    Sou Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic_img2topic.html b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic_img2topic.html index dd7965c67..4b8a40817 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic_img2topic.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic_img2topic.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -421,7 +413,7 @@

    Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic_img_topic2zd.html b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic_img_topic2zd.html index 7856cd8ff..6b53cad46 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic_img_topic2zd.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_domain_topic_img_topic2zd.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -422,7 +414,7 @@

    Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_xyd_parallel.html b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_xyd_parallel.html index 01ddb0f9a..76d4ba160 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_xyd_parallel.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_xyd_parallel.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -529,7 +521,7 @@

    Sou Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_xydt_elevator.html b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_xydt_elevator.html index 49f6491d4..201d071ac 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_xydt_elevator.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_xydt_elevator.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -453,7 +445,7 @@

    So Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_zy.html b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_zy.html index 99e0772a8..8767fd181 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_zy.html +++ b/docs/build/html/_modules/domainlab/compos/vae/compos/encoder_zy.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -420,7 +412,7 @@

    Source code f Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/utils_request_chain_builder.html b/docs/build/html/_modules/domainlab/compos/vae/utils_request_chain_builder.html index a3c177d4f..81c7835c3 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/utils_request_chain_builder.html +++ b/docs/build/html/_modules/domainlab/compos/vae/utils_request_chain_builder.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -401,7 +393,7 @@

    Sou Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/zoo_vae_builders_classif.html b/docs/build/html/_modules/domainlab/compos/vae/zoo_vae_builders_classif.html index 4b12f15fe..bd5e6132c 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/zoo_vae_builders_classif.html +++ b/docs/build/html/_modules/domainlab/compos/vae/zoo_vae_builders_classif.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -509,7 +501,7 @@

    Source Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/vae/zoo_vae_builders_classif_topic.html b/docs/build/html/_modules/domainlab/compos/vae/zoo_vae_builders_classif_topic.html index 7de5e09eb..20a096070 100644 --- a/docs/build/html/_modules/domainlab/compos/vae/zoo_vae_builders_classif_topic.html +++ b/docs/build/html/_modules/domainlab/compos/vae/zoo_vae_builders_classif_topic.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -418,7 +410,7 @@

    Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/compos/zoo_nn.html b/docs/build/html/_modules/domainlab/compos/zoo_nn.html index 64807da3c..40f1392e1 100644 --- a/docs/build/html/_modules/domainlab/compos/zoo_nn.html +++ b/docs/build/html/_modules/domainlab/compos/zoo_nn.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -407,7 +399,7 @@

    Source code for domainlab.co Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/dsets/a_dset_mnist_color_rgb_solo.html b/docs/build/html/_modules/domainlab/dsets/a_dset_mnist_color_rgb_solo.html index 2312938c7..28ef16ad2 100644 --- a/docs/build/html/_modules/domainlab/dsets/a_dset_mnist_color_rgb_solo.html +++ b/docs/build/html/_modules/domainlab/dsets/a_dset_mnist_color_rgb_solo.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -531,7 +523,7 @@

    Source c Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/dsets/dset_img_path_list.html b/docs/build/html/_modules/domainlab/dsets/dset_img_path_list.html index 4a8c03122..78c699278 100644 --- a/docs/build/html/_modules/domainlab/dsets/dset_img_path_list.html +++ b/docs/build/html/_modules/domainlab/dsets/dset_img_path_list.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -401,7 +393,7 @@

    Source code for d Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/dsets/dset_mnist_color_solo_default.html b/docs/build/html/_modules/domainlab/dsets/dset_mnist_color_solo_default.html index a60fa0614..6e127f3ee 100644 --- a/docs/build/html/_modules/domainlab/dsets/dset_mnist_color_solo_default.html +++ b/docs/build/html/_modules/domainlab/dsets/dset_mnist_color_solo_default.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -389,7 +381,7 @@

    Source Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/dsets/dset_poly_domains_mnist_color_default.html b/docs/build/html/_modules/domainlab/dsets/dset_poly_domains_mnist_color_default.html index feaa9d26a..22f05f523 100644 --- a/docs/build/html/_modules/domainlab/dsets/dset_poly_domains_mnist_color_default.html +++ b/docs/build/html/_modules/domainlab/dsets/dset_poly_domains_mnist_color_default.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -409,7 +401,7 @@

    Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/dsets/dset_subfolder.html b/docs/build/html/_modules/domainlab/dsets/dset_subfolder.html index e11b14d10..511cbf200 100644 --- a/docs/build/html/_modules/domainlab/dsets/dset_subfolder.html +++ b/docs/build/html/_modules/domainlab/dsets/dset_subfolder.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -526,7 +518,7 @@

    Source code for domai Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/dsets/utils_data.html b/docs/build/html/_modules/domainlab/dsets/utils_data.html index 09c9bc429..e5efda3d1 100644 --- a/docs/build/html/_modules/domainlab/dsets/utils_data.html +++ b/docs/build/html/_modules/domainlab/dsets/utils_data.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -461,7 +453,7 @@

    Source code for domainlab Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/dsets/utils_wrapdset_patches.html b/docs/build/html/_modules/domainlab/dsets/utils_wrapdset_patches.html index 56fe5bbcc..e5981a62d 100644 --- a/docs/build/html/_modules/domainlab/dsets/utils_wrapdset_patches.html +++ b/docs/build/html/_modules/domainlab/dsets/utils_wrapdset_patches.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -525,7 +517,7 @@

    Source code f Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/exp/exp_cuda_seed.html b/docs/build/html/_modules/domainlab/exp/exp_cuda_seed.html index 659af1a09..f3d5264c3 100644 --- a/docs/build/html/_modules/domainlab/exp/exp_cuda_seed.html +++ b/docs/build/html/_modules/domainlab/exp/exp_cuda_seed.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -383,7 +375,7 @@

    Source code for domainla Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/exp/exp_main.html b/docs/build/html/_modules/domainlab/exp/exp_main.html deleted file mode 100644 index e0f6d77da..000000000 --- a/docs/build/html/_modules/domainlab/exp/exp_main.html +++ /dev/null @@ -1,513 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - domainlab.exp.exp_main — domainlab documentation - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Skip to content -
    - -
    - - -
    - - - - -
    -
    - -
    -
    -
    - -
    -
    -
    -
    -
    -
    - - -
    -
    -
    - -
    -
    - -

    Source code for domainlab.exp.exp_main

    -"""
    -experiment
    -"""
    -import datetime
    -import os
    -import warnings
    -
    -from domainlab.algos.zoo_algos import AlgoBuilderChainNodeGetter
    -from domainlab.exp.exp_utils import AggWriter
    -from domainlab.tasks.zoo_tasks import TaskChainNodeGetter
    -from domainlab.utils.logger import Logger
    -from domainlab.utils.sanity_check import SanityCheck
    -os.environ["CUDA_LAUNCH_BLOCKING"] = "1"  # debug
    -
    -
    -
    [docs]class Exp: - """ - Exp is combination of Task, Algorithm, and Configuration (including random seed) - """ - - def __init__(self, args, task=None, model=None, observer=None, visitor=AggWriter): - """ - :param args: - :param task: default None - :param model: default None - """ - self.task = task - self.curr_dir = os.getcwd() - if task is None: - self.task = TaskChainNodeGetter(args)() - - self.args = args - algo_builder = AlgoBuilderChainNodeGetter( - self.args.model, self.args.apath - )() # request - # the critical logic below is to avoid circular dependence between task initialization - # and trainer initialization: - - self.trainer, self.model, observer_default, device = algo_builder.init_business( - self - ) - # sanity check has to be done after init_business - # jigen algorithm builder has method dset_decoration_args_algo, which could AOP - # into the task intilization process - if args.san_check: - sancheck = SanityCheck(args, self.task) - sancheck.dataset_sanity_check() - - if model is not None: - self.model = model - self.epochs = self.args.epos - self.epoch_counter = 1 - self.val_threshold = args.val_threshold - if observer is None: - observer = observer_default - if not self.trainer.flag_initialized: - self.trainer.init_business(self.model, self.task, observer, device, args) - self.visitor = visitor(self) # visitor depends on task initialization first - # visitor must be initialized last after trainer is initialized - self.experiment_duration = None - self.model.set_saver(self.visitor) - -
    [docs] def execute(self, num_epochs=None): - """ - train model - check performance by loading persisted model - """ - self.model.save() # cause CI infinite loop when put in initializer? - if num_epochs is None: - num_epochs = self.epochs + 1 - t_0 = datetime.datetime.now() - logger = Logger.get_logger() - logger.info(f"\n Experiment start at: {str(t_0)}") - t_c = t_0 - self.trainer.before_tr() - for epoch in range(1, num_epochs): - t_before_epoch = t_c - flag_stop = self.trainer.tr_epoch(epoch) - t_c = datetime.datetime.now() - logger.info( - f"after epoch: {epoch}," - f"now: {str(t_c)}," - f"epoch time: {t_c - t_before_epoch}," - f"used: {t_c - t_0}," - f"model: {self.visitor.model_name}" - ) - logger.info(f"working direcotry: {self.curr_dir}") - # current time, time since experiment start, epoch time - if flag_stop: - self.epoch_counter = epoch - logger.info("early stop trigger") - break - if epoch == self.epochs: - self.epoch_counter = self.epochs - else: - self.epoch_counter += 1 - logger.info( - f"Experiment finished at epoch: {self.epoch_counter} " - f"with time: {t_c - t_0} at {t_c}" - ) - self.experiment_duration = t_c - t_0 - self.trainer.post_tr()
    - -
    [docs] def clean_up(self): - """ - to be called by a decorator - """ - try: - # oracle means use out-of-domain test accuracy to select the model - self.visitor.remove("oracle") # pylint: disable=E1101 - except FileNotFoundError: - pass - - try: - # the last epoch: - # have a model to evaluate in case the training stops in between - self.visitor.remove("epoch") # pylint: disable=E1101 - except FileNotFoundError: - logger = Logger.get_logger() - logger.warn("failed to remove model_epoch: file not found") - warnings.warn("failed to remove model_epoch: file not found") - - try: - # without suffix: the selected model - self.visitor.remove() # pylint: disable=E1101 - except FileNotFoundError: - logger = Logger.get_logger() - logger.warn("failed to remove model") - warnings.warn("failed to remove model") - - try: - # for matchdg - self.visitor.remove("ctr") # pylint: disable=E1101 - except FileNotFoundError: - pass
    -
    - -
    -
    -
    -
    -
    -
    - - -
    - - - - \ No newline at end of file diff --git a/docs/build/html/_modules/domainlab/exp/exp_utils.html b/docs/build/html/_modules/domainlab/exp/exp_utils.html deleted file mode 100644 index 1a7180a60..000000000 --- a/docs/build/html/_modules/domainlab/exp/exp_utils.html +++ /dev/null @@ -1,668 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - domainlab.exp.exp_utils — domainlab documentation - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Skip to content -
    - -
    - - -
    - - - - -
    -
    - -
    -
    -
    - -
    -
    -
    -
    -
    -
    - - -
    -
    -
    - -
    -
    - -

    Source code for domainlab.exp.exp_utils

    -"""
    -This module contains 3 classes inheriting:
    -    ExpProtocolAggWriter(AggWriter(ExpModelPersistVisitor))
    -"""
    -import copy
    -import datetime
    -import os
    -from pathlib import Path
    -
    -import numpy as np
    -import torch
    -from sklearn.metrics import ConfusionMatrixDisplay
    -
    -from domainlab.utils.get_git_tag import get_git_tag
    -from domainlab.utils.logger import Logger
    -
    -
    -
    [docs]class ExpModelPersistVisitor: - """ - This class couples with Task class attributes - """ - - model_dir = "saved_models" - model_suffix = ".model" - - def __init__(self, host): - """ - 1. create new attributes like model names - 2. all dependencies in contructor - 3. do not change the sequence of the lines! - since later lines depends on earlier definitions - """ - self.host = host - self.out = host.args.out - self.model_dir = os.path.join(self.out, ExpModelPersistVisitor.model_dir) - self.git_tag = get_git_tag() - self.task_name = self.host.task.get_na(self.host.args.tr_d, self.host.args.te_d) - self.algo_name = self.host.args.model - self.seed = self.host.args.seed - self.model_name = self.mk_model_na(self.git_tag) - self.model_path = os.path.join( - self.model_dir, self.model_name + ExpModelPersistVisitor.model_suffix - ) - - Path(os.path.dirname(self.model_path)).mkdir(parents=True, exist_ok=True) - self.model = copy.deepcopy(self.host.trainer.model) - # although deepcopy in contructor is expensive, but - # execute copy.deepcopy(self.host.trainer.model) after training will cause thread lock - # if self.host.trainer has tensorboard writer, see - # https://github.com/marrlab/DomainLab/issues/673 - -
    [docs] def mk_model_na(self, tag=None, dd_cut=19): - """ - :param tag: for git commit hash for example - """ - if tag is None: - tag = "tag" - suffix_t = str(datetime.datetime.now())[:dd_cut].replace(" ", "_") - suffix_t = suffix_t.replace("-", "md_") - suffix_t = suffix_t.replace(":", "_") - list4mname = [ - self.task_name, - self.algo_name, - tag, - suffix_t, - "seed", - str(self.seed), - ] - # the sequence of components (e.g. seed in the last place) - # in model name is not crutial - model_name = "_".join(list4mname) - if self.host.args.debug: - model_name = "debug_" + model_name - slurm = os.environ.get("SLURM_JOB_ID") - if slurm: - model_name = model_name + "_" + slurm - logger = Logger.get_logger() - logger.info(f"model name: {model_name}") - return model_name
    - -
    [docs] def save(self, model, suffix=None): - """ - :param model: - """ - file_na = self.model_path - if suffix is not None: - file_na = "_".join([file_na, suffix]) - torch.save(copy.deepcopy(model.state_dict()), file_na)
    - # checkpoint = {'model': Net(), ' - # state_dict': model.state_dict(), - # 'optimizer' :optimizer.state_dict()} - # torch.save(checkpoint, 'Checkpoint.pth') - -
    [docs] def remove(self, suffix=None): - """ - remove model after use - """ - file_na = self.model_path - if suffix is not None: - file_na = "_".join([file_na, suffix]) - os.remove(file_na)
    - -
    [docs] def load(self, suffix=None): - """ - load pre-defined model name from disk - the save function is the same class so to ensure load will ways work - """ - path = self.model_path - if suffix is not None: - path = "_".join([self.model_path, suffix]) - # due to tensorboard writer in trainer.scheduler, - # copy.deepcopy(self.host.trainer.model) will cause thread lock - # see https://github.com/marrlab/DomainLab/issues/673 - self.model.load_state_dict(torch.load(path, map_location="cpu")) - # without separate self.model and self.model_suffixed, - # it will cause accuracy inconsistent problems since the content of self.model - # can be overwritten when the current function is called another time and self.model - # is not deepcopied - # However, deepcopy is also problematic when it is executed too many times - return copy.deepcopy(self.model)
    - # instead of deepcopy, one could also have multiple copies of model in constructor, but this - # does not adhere the lazy principle. - -
    [docs] def clean_up(self): - self.host.clean_up()
    - - -
    [docs]class AggWriter(ExpModelPersistVisitor): - """ - 1. aggregate results to text file. - 2. all dependencies are in the constructor - """ - - def __init__(self, host): - super().__init__(host) - self.agg_tag = self.host.args.aggtag - self.exp_tag = self.host.args.exptag - self.debug = self.host.args.debug - self.has_first_line = False - self.list_cols = None - -
    [docs] def first_line(self, dict_cols): - """ - generate header of the results aggregation file - """ - self.list_cols = list(dict_cols.keys()) - # @FIXME: will be list be the same order each time? - str_line = ", ".join(self.list_cols) - if not os.path.isfile(self.get_fpath()): - self.to_file(str_line) - self.has_first_line = True
    - - def __call__(self, dict_metric): - line, confmat, confmat_filename = self._gen_line(dict_metric) - self.to_file(line) - if not self.host.args.no_dump: - self.confmat_to_file(confmat, confmat_filename) - -
    [docs] def get_cols(self): - """ - call the same function to always get the same columns configuration - """ - epos_name = "epos" - dict_cols = { - "algo": self.algo_name, - epos_name: None, - "seed": self.seed, - "aggtag": self.agg_tag, - # algorithm configuration for instance - "mname": "mname_" + self.model_name, - "commit": "commit_" + self.git_tag, - } - return dict_cols, epos_name
    - - def _gen_line(self, dict_metric): - dict_cols, epos_name = self.get_cols() - dict_cols.update(dict_metric) - confmat = dict_cols.pop("confmat") - confmat_filename = dict_cols.get("mname", None) # return None if not found - # @FIXME: strong dependency on host attribute name - dict_cols.update({epos_name: self.host.epoch_counter}) - if self.host.experiment_duration is not None: - dict_cols.update({"experiment_duration": self.host.experiment_duration}) - else: - print("Does not have attribute") - print(self.host) - if not self.has_first_line: - self.first_line(dict_cols) - list_str = [str(dict_cols[key]) for key in self.list_cols] - str_line = ", ".join(list_str) - return str_line, confmat, confmat_filename - -
    [docs] def get_fpath(self, dirname="aggrsts"): - """ - for writing and reading, the same function is called to ensure name - change in the future will not break the software - """ - list4fname = [ - self.task_name, - self.exp_tag, - ] - fname = "_".join(list4fname) + ".csv" - if self.debug: - fname = "_".join(["debug_agg", fname]) - file_path = os.path.join(self.out, dirname, fname) - return file_path
    - -
    [docs] def to_file(self, str_line): - """ - :param str_line: - """ - file_path = self.get_fpath() - Path(os.path.dirname(file_path)).mkdir(parents=True, exist_ok=True) - logger = Logger.get_logger() - logger.info(f"results aggregation path: {file_path}") - with open(file_path, "a", encoding="utf8") as f_h: - print(str_line, file=f_h)
    - -
    [docs] def confmat_to_file(self, confmat, confmat_filename): - """Save confusion matrix as a figure - - Args: - confmat: confusion matrix. - """ - disp = ConfusionMatrixDisplay(confmat) - disp = disp.plot(cmap="gray") - file_path = self.get_fpath() - # @FIXME: although removesuffix is safe when suffix does not exist, - # we would like to have ".csv" live in some configuraiton file in the future. - file_path = file_path.removesuffix(".csv") - # if prefix does not exist, string remain unchanged. - # @FIXME: still we want to have mname_ as a variable defined in some - # configuration file in the future. - confmat_filename = confmat_filename.removeprefix("mname_") - file_path = os.path.join( - os.path.dirname(file_path), f"{confmat_filename}_conf_mat.png" - ) - logger = Logger.get_logger() - logger.info(f"confusion matrix saved in file: {file_path}") - disp.figure_.savefig(file_path)
    - - -
    [docs]class ExpProtocolAggWriter(AggWriter): - """ - AggWriter tailored to experimental protocol - Output contains additionally index, exp task, te_d and params. - """ - -
    [docs] def get_cols(self): - """columns""" - epos_name = "epos" - dict_cols = { - "param_index": self.host.args.param_index, - "method": self.host.args.benchmark_task_name, - "mname": "mname_" + self.model_name, - "commit": "commit_" + self.git_tag, - "algo": self.algo_name, - epos_name: None, - "te_d": self.host.args.te_d, - "seed": self.seed, - "params": f'"{self.host.args.params}"', - } - return dict_cols, epos_name
    - -
    [docs] def get_fpath(self, dirname=None): - """filepath""" - if dirname is None: - return self.host.args.result_file - return super().get_fpath(dirname)
    - -
    [docs] def confmat_to_file(self, confmat, confmat_filename): - """Save confusion matrix as a figure - - Args: - confmat: confusion matrix. - """ - path4file = self.get_fpath() - index = os.path.basename(path4file) - path4file = os.path.dirname(os.path.dirname(path4file)) - # if prefix does not exist, string remain unchanged. - confmat_filename = confmat_filename.removeprefix("mname_") - path4file = os.path.join(path4file, "confusion_matrix") - os.makedirs(path4file, exist_ok=True) - file_path = os.path.join(path4file, f"{index}.txt") - with open(file_path, "a", encoding="utf8") as f_h: - print(confmat_filename, file=f_h) - for line in np.matrix(confmat): - np.savetxt(f_h, line, fmt="%.2f") - logger = Logger.get_logger() - logger.info(f"confusion matrix saved in file: {file_path}")
    -
    - -
    -
    -
    -
    -
    -
    - - -
    - - - - \ No newline at end of file diff --git a/docs/build/html/_modules/domainlab/exp_protocol/aggregate_results.html b/docs/build/html/_modules/domainlab/exp_protocol/aggregate_results.html index c65ddaa51..cf94fabca 100644 --- a/docs/build/html/_modules/domainlab/exp_protocol/aggregate_results.html +++ b/docs/build/html/_modules/domainlab/exp_protocol/aggregate_results.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -414,7 +406,7 @@

    Source code Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/exp_protocol/run_experiment.html b/docs/build/html/_modules/domainlab/exp_protocol/run_experiment.html deleted file mode 100644 index 039e9d4ed..000000000 --- a/docs/build/html/_modules/domainlab/exp_protocol/run_experiment.html +++ /dev/null @@ -1,565 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - domainlab.exp_protocol.run_experiment — domainlab documentation - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Skip to content -
    - -
    - - -
    - - - - -
    -
    - -
    -
    -
    - -
    -
    -
    -
    -
    -
    - - -
    -
    -
    - -
    -
    - -

    Source code for domainlab.exp_protocol.run_experiment

    -"""
    -Runs one task for a single hyperparameter sample for each leave-out-domain
    -and each random seed.
    -"""
    -import ast
    -import copy
    -import gc
    -
    -import numpy as np
    -import pandas as pd
    -import torch
    -
    -from domainlab.arg_parser import apply_dict_to_args, mk_parser_main
    -from domainlab.exp.exp_cuda_seed import set_seed
    -from domainlab.exp.exp_main import Exp
    -from domainlab.exp.exp_utils import ExpProtocolAggWriter
    -from domainlab.utils.hyperparameter_sampling import G_METHOD_NA
    -from domainlab.utils.logger import Logger
    -
    -
    -
    [docs]def load_parameters(file: str, index: int) -> tuple: - """ - Loads a single parameter sample - @param file: csv file - @param index: index of hyper-parameter - """ - param_df = pd.read_csv(file, index_col=0) - row = param_df.loc[index] - params = ast.literal_eval(row.params) - # row.task has nothing to do with DomainLab task, it is - # benchmark task which correspond to one algorithm - return row[G_METHOD_NA], params
    - - -
    [docs]def convert_dict2float(dict_in): - """ - convert scientific notation from 1e5 to 10000 - """ - dict_out = copy.deepcopy(dict_in) - for key, val in dict_out.items(): - if isinstance(val, str): - try: - val_float = float(val) - dict_out[key] = val_float - except: - pass - return dict_out
    - - -
    [docs]def run_experiment( - config: dict, - param_file: str, - param_index: int, - out_file: str, - start_seed=None, - misc=None, - num_gpus=1, -): - """ - Runs the experiment several times: - - for test_domain in test_domains: - for seed from startseed to endseed: - evaluate the algorithm with test_domain, initialization with seed - - :param config: dictionary from the benchmark yaml - :param param_file: path to the csv with the parameter samples - :param param_index: parameter index that should be covered by this task, - currently this correspond to the line number in the csv file, or row number - in the resulting pandas dataframe - :param out_file: path to the output csv - :param start_seed: random seed to start for stochastic variations of pytorch - :param misc: optional dictionary of additional parameters, if any. - - # FIXME: we might want to run the experiment using commandline arguments - """ - - if misc is None: - misc = {} - str_algo_as_task, hyperparameters = load_parameters(param_file, param_index) - logger = Logger.get_logger() - logger.debug( - "\n*******************************************************************" - ) - logger.debug( - f"{str_algo_as_task}, param_index={param_index}, params={hyperparameters}" - ) - logger.debug( - "*******************************************************************\n" - ) - misc["result_file"] = out_file - misc["params"] = hyperparameters - misc["benchmark_task_name"] = str_algo_as_task - misc["param_index"] = param_index - misc["keep_model"] = False - - parser = mk_parser_main() - args = parser.parse_args(args=[]) - args_algo_specific = config[str_algo_as_task].copy() - if "hyperparameters" in args_algo_specific: - del args_algo_specific["hyperparameters"] - args_domainlab_common_raw = config.get("domainlab_args", {}) - args_domainlab_common = convert_dict2float(args_domainlab_common_raw) - # check if some of the hyperparameters are already specified - # in args_domainlab_common or args_algo_specific - if ( - np.intersect1d( - list(args_algo_specific.keys()), list(hyperparameters.keys()) - ).shape[0] - > 0 - ): - logger.error( - f"the hyperparameter " - f"{np.intersect1d(list(args_algo_specific.keys()), list(hyperparameters.keys()))}" - f" has already been fixed to a value in the algorithm section." - ) - raise RuntimeError( - f"the hyperparameter " - f"{np.intersect1d(list(args_algo_specific.keys()), list(hyperparameters.keys()))}" - f" has already been fixed to a value in the algorithm section." - ) - if ( - np.intersect1d( - list(args_domainlab_common.keys()), list(hyperparameters.keys()) - ).shape[0] - > 0 - ): - logger.error( - f"the hyperparameter " - f"{np.intersect1d(list(args_algo_specific.keys()), list(hyperparameters.keys()))}" - f" has already been fixed to a value in the domainlab_args section." - ) - raise RuntimeError( - f"the hyperparameter " - f"{np.intersect1d(list(args_algo_specific.keys()), list(hyperparameters.keys()))}" - f" has already been fixed to a value in the domainlab_args section." - ) - apply_dict_to_args(args, args_domainlab_common) - args_algo_specific_scientific_notation = convert_dict2float(args_algo_specific) - apply_dict_to_args(args, args_algo_specific_scientific_notation, extend=True) - apply_dict_to_args(args, hyperparameters) - apply_dict_to_args(args, misc, extend=True) - gpu_ind = param_index % num_gpus - args.device = str(gpu_ind) - - if torch.cuda.is_available(): - torch.cuda.init() - logger.info("before experiment loop: ") - logger.info(str(torch.cuda.memory_summary())) - if start_seed is None: - start_seed = config["startseed"] - end_seed = config["endseed"] - else: - end_seed = start_seed + (config["endseed"] - config["startseed"]) - for seed in range(start_seed, end_seed + 1): - for te_d in config["test_domains"]: - args.te_d = te_d - set_seed(seed) - args.seed = seed - try: - if torch.cuda.is_available(): - logger.info("before experiment starts") - logger.info(str(torch.cuda.memory_summary())) - except KeyError as ex: - logger.error(str(ex)) - args.lr = float(args.lr) - # <=' not supported between instances of 'float' and 'str - exp = Exp(args=args, visitor=ExpProtocolAggWriter) - # NOTE: if key "testing" is set in benchmark, then do not execute - # experiment - if not misc.get("testing", False): - exp.execute() - try: - if torch.cuda.is_available(): - logger.info("before torch memory clean up") - logger.info(str(torch.cuda.memory_summary())) - except KeyError as ex: - logger.error(str(ex)) - del exp - torch.cuda.empty_cache() - gc.collect() - try: - if torch.cuda.is_available(): - logger.info("after torch memory clean up") - logger.info(str(torch.cuda.memory_summary())) - except KeyError as ex: - logger.error(str(ex))
    -
    - -
    -
    -
    -
    -
    -
    - - -
    - - - - \ No newline at end of file diff --git a/docs/build/html/_modules/domainlab/mk_exp.html b/docs/build/html/_modules/domainlab/mk_exp.html deleted file mode 100644 index f4697e0aa..000000000 --- a/docs/build/html/_modules/domainlab/mk_exp.html +++ /dev/null @@ -1,411 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - domainlab.mk_exp — domainlab documentation - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Skip to content -
    - -
    - - -
    - - - - -
    -
    - -
    -
    -
    - -
    -
    -
    -
    -
    -
    - - -
    -
    -
    - -
    -
    - -

    Source code for domainlab.mk_exp

    -"""
    -make an experiment
    -"""
    -from domainlab.arg_parser import mk_parser_main
    -from domainlab.exp.exp_main import Exp
    -
    -
    -
    [docs]def mk_exp(task, model, trainer: str, test_domain: str, batchsize: int, nocu=False): - """ - Creates a custom experiment. The user can specify the input parameters. - - Input Parameters: - - task: create a task to a custom dataset by importing "mk_task_dset" function from - "domainlab.tasks.task_dset". For more explanation on the input params refer to the - documentation found in "domainlab.tasks.task_dset.py". - - model: create a model [NameOfModel] by importing "mk_[NameOfModel]" function from - "domainlab.models.model_[NameOfModel]". For a concrete example and explanation of the input - params refer to the documentation found in "domainlab.models.model_[NameOfModel].py" - - trainer: string, - - test_domain: string, - - batch size: int - - Returns: experiment - """ - str_arg = ( - f"--model=apimodel --trainer={trainer} --te_d={test_domain} --bs={batchsize}" - ) - if nocu: - str_arg += " --nocu" - parser = mk_parser_main() - conf = parser.parse_args(str_arg.split()) - exp = Exp(conf, task, model=model) - return exp
    -
    - -
    -
    -
    -
    -
    -
    - - -
    - - - - \ No newline at end of file diff --git a/docs/build/html/_modules/domainlab/models/a_model.html b/docs/build/html/_modules/domainlab/models/a_model.html index 2e270487c..96f148b8b 100644 --- a/docs/build/html/_modules/domainlab/models/a_model.html +++ b/docs/build/html/_modules/domainlab/models/a_model.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -576,7 +568,7 @@

    Source code for domainlab.m Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/a_model_classif.html b/docs/build/html/_modules/domainlab/models/a_model_classif.html index 89e672964..c49431094 100644 --- a/docs/build/html/_modules/domainlab/models/a_model_classif.html +++ b/docs/build/html/_modules/domainlab/models/a_model_classif.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -610,7 +602,7 @@

    Source code for dom Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/args_jigen.html b/docs/build/html/_modules/domainlab/models/args_jigen.html index 8fcd63446..6cdae6f8d 100644 --- a/docs/build/html/_modules/domainlab/models/args_jigen.html +++ b/docs/build/html/_modules/domainlab/models/args_jigen.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -393,7 +385,7 @@

    Source code for domainla Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/args_vae.html b/docs/build/html/_modules/domainlab/models/args_vae.html index ffa03f24e..5d78fa41b 100644 --- a/docs/build/html/_modules/domainlab/models/args_vae.html +++ b/docs/build/html/_modules/domainlab/models/args_vae.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -440,7 +432,7 @@

    Source code for domainlab. Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/interface_vae_xyd.html b/docs/build/html/_modules/domainlab/models/interface_vae_xyd.html index cf7bbe2fb..2ca349eb6 100644 --- a/docs/build/html/_modules/domainlab/models/interface_vae_xyd.html +++ b/docs/build/html/_modules/domainlab/models/interface_vae_xyd.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -411,7 +403,7 @@

    Source code for d Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/model_custom.html b/docs/build/html/_modules/domainlab/models/model_custom.html index e178a4ae8..45fcba94e 100644 --- a/docs/build/html/_modules/domainlab/models/model_custom.html +++ b/docs/build/html/_modules/domainlab/models/model_custom.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -380,7 +372,7 @@

    Source code for domain Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/model_dann.html b/docs/build/html/_modules/domainlab/models/model_dann.html index a9f51cdab..8166dcb0a 100644 --- a/docs/build/html/_modules/domainlab/models/model_dann.html +++ b/docs/build/html/_modules/domainlab/models/model_dann.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -474,7 +466,7 @@

    Source code for domainla Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/model_diva.html b/docs/build/html/_modules/domainlab/models/model_diva.html index 690020df0..e55b1dc5c 100644 --- a/docs/build/html/_modules/domainlab/models/model_diva.html +++ b/docs/build/html/_modules/domainlab/models/model_diva.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -530,7 +522,7 @@

    Source code for domainla Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/model_hduva.html b/docs/build/html/_modules/domainlab/models/model_hduva.html index 570f65872..cd9688224 100644 --- a/docs/build/html/_modules/domainlab/models/model_hduva.html +++ b/docs/build/html/_modules/domainlab/models/model_hduva.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -580,7 +572,7 @@

    Source code for domainl Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/model_jigen.html b/docs/build/html/_modules/domainlab/models/model_jigen.html index 24880d47e..a80dd4e38 100644 --- a/docs/build/html/_modules/domainlab/models/model_jigen.html +++ b/docs/build/html/_modules/domainlab/models/model_jigen.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -490,7 +482,7 @@

    Source code for domainl Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/models/model_vae_xyd_classif.html b/docs/build/html/_modules/domainlab/models/model_vae_xyd_classif.html index 964578aba..a81619002 100644 --- a/docs/build/html/_modules/domainlab/models/model_vae_xyd_classif.html +++ b/docs/build/html/_modules/domainlab/models/model_vae_xyd_classif.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -407,7 +399,7 @@

    Source code f Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/a_task.html b/docs/build/html/_modules/domainlab/tasks/a_task.html index 8467de262..f97847264 100644 --- a/docs/build/html/_modules/domainlab/tasks/a_task.html +++ b/docs/build/html/_modules/domainlab/tasks/a_task.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -550,7 +542,7 @@

    Source code for domainlab.tas Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/a_task_classif.html b/docs/build/html/_modules/domainlab/tasks/a_task_classif.html index 05fd3909c..ee9230a5f 100644 --- a/docs/build/html/_modules/domainlab/tasks/a_task_classif.html +++ b/docs/build/html/_modules/domainlab/tasks/a_task_classif.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -442,7 +434,7 @@

    Source code for domai Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/b_task.html b/docs/build/html/_modules/domainlab/tasks/b_task.html index e7c44c3cb..e0945d81c 100644 --- a/docs/build/html/_modules/domainlab/tasks/b_task.html +++ b/docs/build/html/_modules/domainlab/tasks/b_task.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -438,7 +430,7 @@

    Source code for domainlab.tas Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/b_task_classif.html b/docs/build/html/_modules/domainlab/tasks/b_task_classif.html index 89607de97..09dd1fd0c 100644 --- a/docs/build/html/_modules/domainlab/tasks/b_task_classif.html +++ b/docs/build/html/_modules/domainlab/tasks/b_task_classif.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -407,7 +399,7 @@

    Source code for domai Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/task_dset.html b/docs/build/html/_modules/domainlab/tasks/task_dset.html index 1f6e7bd09..0d9e2dc1a 100644 --- a/docs/build/html/_modules/domainlab/tasks/task_dset.html +++ b/docs/build/html/_modules/domainlab/tasks/task_dset.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -431,7 +423,7 @@

    Source code for domainlab. Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/task_folder.html b/docs/build/html/_modules/domainlab/tasks/task_folder.html index 3fb509c10..c18353b46 100644 --- a/docs/build/html/_modules/domainlab/tasks/task_folder.html +++ b/docs/build/html/_modules/domainlab/tasks/task_folder.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -475,7 +467,7 @@

    Source code for domainla Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/task_folder_mk.html b/docs/build/html/_modules/domainlab/tasks/task_folder_mk.html index 96598eaa9..1911cf998 100644 --- a/docs/build/html/_modules/domainlab/tasks/task_folder_mk.html +++ b/docs/build/html/_modules/domainlab/tasks/task_folder_mk.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -425,7 +417,7 @@

    Source code for domai Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/task_mini_vlcs.html b/docs/build/html/_modules/domainlab/tasks/task_mini_vlcs.html index 1f42be05f..8a7bb2ca7 100644 --- a/docs/build/html/_modules/domainlab/tasks/task_mini_vlcs.html +++ b/docs/build/html/_modules/domainlab/tasks/task_mini_vlcs.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -425,7 +417,7 @@

    Source code for domai Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/task_mnist_color.html b/docs/build/html/_modules/domainlab/tasks/task_mnist_color.html index 16c790499..604546920 100644 --- a/docs/build/html/_modules/domainlab/tasks/task_mnist_color.html +++ b/docs/build/html/_modules/domainlab/tasks/task_mnist_color.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -428,7 +420,7 @@

    Source code for dom Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/task_pathlist.html b/docs/build/html/_modules/domainlab/tasks/task_pathlist.html index 654e98cea..e905301a1 100644 --- a/docs/build/html/_modules/domainlab/tasks/task_pathlist.html +++ b/docs/build/html/_modules/domainlab/tasks/task_pathlist.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -500,7 +492,7 @@

    Source code for domain Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/task_utils.html b/docs/build/html/_modules/domainlab/tasks/task_utils.html index f24ea6f4c..eff3bfd66 100644 --- a/docs/build/html/_modules/domainlab/tasks/task_utils.html +++ b/docs/build/html/_modules/domainlab/tasks/task_utils.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -391,7 +383,7 @@

    Source code for domainlab Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/utils_task.html b/docs/build/html/_modules/domainlab/tasks/utils_task.html index 95a993eb7..47b7a2717 100644 --- a/docs/build/html/_modules/domainlab/tasks/utils_task.html +++ b/docs/build/html/_modules/domainlab/tasks/utils_task.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -653,7 +645,7 @@

    Source code for domainlab Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/utils_task_dset.html b/docs/build/html/_modules/domainlab/tasks/utils_task_dset.html index 66e1137bd..06ba29dc0 100644 --- a/docs/build/html/_modules/domainlab/tasks/utils_task_dset.html +++ b/docs/build/html/_modules/domainlab/tasks/utils_task_dset.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -442,7 +434,7 @@

    Source code for doma Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/tasks/zoo_tasks.html b/docs/build/html/_modules/domainlab/tasks/zoo_tasks.html index cfeb49952..684d0da04 100644 --- a/docs/build/html/_modules/domainlab/tasks/zoo_tasks.html +++ b/docs/build/html/_modules/domainlab/tasks/zoo_tasks.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -424,7 +416,7 @@

    Source code for domainlab. Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/flows_gen_img_model.html b/docs/build/html/_modules/domainlab/utils/flows_gen_img_model.html index 22ebed58c..54544da85 100644 --- a/docs/build/html/_modules/domainlab/utils/flows_gen_img_model.html +++ b/docs/build/html/_modules/domainlab/utils/flows_gen_img_model.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -488,7 +480,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/generate_benchmark_plots.html b/docs/build/html/_modules/domainlab/utils/generate_benchmark_plots.html index c3b2fc080..5ecebede9 100644 --- a/docs/build/html/_modules/domainlab/utils/generate_benchmark_plots.html +++ b/docs/build/html/_modules/domainlab/utils/generate_benchmark_plots.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -1029,7 +1021,7 @@

    Source code Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/get_git_tag.html b/docs/build/html/_modules/domainlab/utils/get_git_tag.html index 4210ad43c..35abef38d 100644 --- a/docs/build/html/_modules/domainlab/utils/get_git_tag.html +++ b/docs/build/html/_modules/domainlab/utils/get_git_tag.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -403,7 +395,7 @@

    Source code for domainla Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/hyperparameter_gridsearch.html b/docs/build/html/_modules/domainlab/utils/hyperparameter_gridsearch.html index 27c0bdb96..31e03b90e 100644 --- a/docs/build/html/_modules/domainlab/utils/hyperparameter_gridsearch.html +++ b/docs/build/html/_modules/domainlab/utils/hyperparameter_gridsearch.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -800,7 +792,7 @@

    Source cod Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/hyperparameter_sampling.html b/docs/build/html/_modules/domainlab/utils/hyperparameter_sampling.html index 169c08ac4..dc4cdd94a 100644 --- a/docs/build/html/_modules/domainlab/utils/hyperparameter_sampling.html +++ b/docs/build/html/_modules/domainlab/utils/hyperparameter_sampling.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -853,7 +845,7 @@

    Source code Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/logger.html b/docs/build/html/_modules/domainlab/utils/logger.html index 41fd8a1f4..fac2da90b 100644 --- a/docs/build/html/_modules/domainlab/utils/logger.html +++ b/docs/build/html/_modules/domainlab/utils/logger.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -410,7 +402,7 @@

    Source code for domainlab.uti Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/override_interface.html b/docs/build/html/_modules/domainlab/utils/override_interface.html index 48cda0541..8ef7de58f 100644 --- a/docs/build/html/_modules/domainlab/utils/override_interface.html +++ b/docs/build/html/_modules/domainlab/utils/override_interface.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -390,7 +382,7 @@

    Source code for d Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/perf.html b/docs/build/html/_modules/domainlab/utils/perf.html index 18a7e5b93..a4055ac7a 100644 --- a/docs/build/html/_modules/domainlab/utils/perf.html +++ b/docs/build/html/_modules/domainlab/utils/perf.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -429,7 +421,7 @@

    Source code for domainlab.utils Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/perf_metrics.html b/docs/build/html/_modules/domainlab/utils/perf_metrics.html index 97dfeb915..9e9c82e23 100644 --- a/docs/build/html/_modules/domainlab/utils/perf_metrics.html +++ b/docs/build/html/_modules/domainlab/utils/perf_metrics.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -462,7 +454,7 @@

    Source code for domainl Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/sanity_check.html b/docs/build/html/_modules/domainlab/utils/sanity_check.html index 6d876d659..defe0e7b2 100644 --- a/docs/build/html/_modules/domainlab/utils/sanity_check.html +++ b/docs/build/html/_modules/domainlab/utils/sanity_check.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -457,7 +449,7 @@

    Source code for domainl Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/test_img.html b/docs/build/html/_modules/domainlab/utils/test_img.html index 16c100906..7ebcfd95c 100644 --- a/docs/build/html/_modules/domainlab/utils/test_img.html +++ b/docs/build/html/_modules/domainlab/utils/test_img.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -392,7 +384,7 @@

    Source code for domainlab.u Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/u_import.html b/docs/build/html/_modules/domainlab/utils/u_import.html index 7f8fdd8c1..f31d7a3e4 100644 --- a/docs/build/html/_modules/domainlab/utils/u_import.html +++ b/docs/build/html/_modules/domainlab/utils/u_import.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -376,7 +368,7 @@

    Source code for domainlab.u Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/u_import_net_module.html b/docs/build/html/_modules/domainlab/utils/u_import_net_module.html index 063faaa42..5c732732b 100644 --- a/docs/build/html/_modules/domainlab/utils/u_import_net_module.html +++ b/docs/build/html/_modules/domainlab/utils/u_import_net_module.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -416,7 +408,7 @@

    Source code for Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/utils_class.html b/docs/build/html/_modules/domainlab/utils/utils_class.html index 23034d5f0..e2f7179c1 100644 --- a/docs/build/html/_modules/domainlab/utils/utils_class.html +++ b/docs/build/html/_modules/domainlab/utils/utils_class.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -391,7 +383,7 @@

    Source code for domainla Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/utils_classif.html b/docs/build/html/_modules/domainlab/utils/utils_classif.html index 98934ba22..9b2d12286 100644 --- a/docs/build/html/_modules/domainlab/utils/utils_classif.html +++ b/docs/build/html/_modules/domainlab/utils/utils_classif.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -406,7 +398,7 @@

    Source code for domain Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/utils_cuda.html b/docs/build/html/_modules/domainlab/utils/utils_cuda.html index d95260091..d19b0beb4 100644 --- a/docs/build/html/_modules/domainlab/utils/utils_cuda.html +++ b/docs/build/html/_modules/domainlab/utils/utils_cuda.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -387,7 +379,7 @@

    Source code for domainlab Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/domainlab/utils/utils_img_sav.html b/docs/build/html/_modules/domainlab/utils/utils_img_sav.html index 5ff9004ff..31454bf1c 100644 --- a/docs/build/html/_modules/domainlab/utils/utils_img_sav.html +++ b/docs/build/html/_modules/domainlab/utils/utils_img_sav.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -309,13 +308,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -404,7 +396,7 @@

    Source code for domain Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_modules/index.html b/docs/build/html/_modules/index.html index 31c12b121..7c9ad56ad 100644 --- a/docs/build/html/_modules/index.html +++ b/docs/build/html/_modules/index.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -307,13 +306,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -376,7 +368,6 @@

    All modules for which code is available

  • domainlab.algos.utils
  • domainlab.algos.zoo_algos
  • domainlab.arg_parser
  • -
  • domainlab.cli
  • domainlab.compos.a_nn_builder
  • domainlab.compos.builder_nn_alex
  • domainlab.compos.builder_nn_conv_bn_relu_2
  • @@ -420,11 +411,7 @@

    All modules for which code is available

  • domainlab.dsets.utils_data
  • domainlab.dsets.utils_wrapdset_patches
  • domainlab.exp.exp_cuda_seed
  • -
  • domainlab.exp.exp_main
  • -
  • domainlab.exp.exp_utils
  • domainlab.exp_protocol.aggregate_results
  • -
  • domainlab.exp_protocol.run_experiment
  • -
  • domainlab.mk_exp
  • domainlab.models.a_model
  • domainlab.models.a_model_classif
  • domainlab.models.args_jigen
  • @@ -493,7 +480,7 @@

    All modules for which code is available

    Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/_static/basic.css b/docs/build/html/_static/basic.css index 088967717..bf18350b6 100644 --- a/docs/build/html/_static/basic.css +++ b/docs/build/html/_static/basic.css @@ -222,7 +222,7 @@ table.modindextable td { /* -- general body styles --------------------------------------------------- */ div.body { - min-width: 360px; + min-width: 450px; max-width: 800px; } @@ -237,6 +237,16 @@ a.headerlink { visibility: hidden; } +a.brackets:before, +span.brackets > a:before{ + content: "["; +} + +a.brackets:after, +span.brackets > a:after { + content: "]"; +} + h1:hover > a.headerlink, h2:hover > a.headerlink, h3:hover > a.headerlink, @@ -324,16 +334,12 @@ aside.sidebar { p.sidebar-title { font-weight: bold; } -nav.contents, -aside.topic, div.admonition, div.topic, blockquote { clear: left; } /* -- topics ---------------------------------------------------------------- */ -nav.contents, -aside.topic, div.topic { border: 1px solid #ccc; @@ -373,9 +379,6 @@ div.body p.centered { div.sidebar > :last-child, aside.sidebar > :last-child, -nav.contents > :last-child, -aside.topic > :last-child, - div.topic > :last-child, div.admonition > :last-child { margin-bottom: 0; @@ -383,9 +386,6 @@ div.admonition > :last-child { div.sidebar::after, aside.sidebar::after, -nav.contents::after, -aside.topic::after, - div.topic::after, div.admonition::after, blockquote::after { @@ -428,6 +428,10 @@ table.docutils td, table.docutils th { border-bottom: 1px solid #aaa; } +table.footnote td, table.footnote th { + border: 0 !important; +} + th { text-align: left; padding-right: 5px; @@ -611,7 +615,6 @@ ul.simple p { margin-bottom: 0; } -/* Docutils 0.17 and older (footnotes & citations) */ dl.footnote > dt, dl.citation > dt { float: left; @@ -629,33 +632,6 @@ dl.citation > dd:after { clear: both; } -/* Docutils 0.18+ (footnotes & citations) */ -aside.footnote > span, -div.citation > span { - float: left; -} -aside.footnote > span:last-of-type, -div.citation > span:last-of-type { - padding-right: 0.5em; -} -aside.footnote > p { - margin-left: 2em; -} -div.citation > p { - margin-left: 4em; -} -aside.footnote > p:last-of-type, -div.citation > p:last-of-type { - margin-bottom: 0em; -} -aside.footnote > p:last-of-type:after, -div.citation > p:last-of-type:after { - content: ""; - clear: both; -} - -/* Footnotes & citations ends */ - dl.field-list { display: grid; grid-template-columns: fit-content(30%) auto; diff --git a/docs/build/html/_static/doctools.js b/docs/build/html/_static/doctools.js index c3db08d1c..e509e4834 100644 --- a/docs/build/html/_static/doctools.js +++ b/docs/build/html/_static/doctools.js @@ -2,263 +2,325 @@ * doctools.js * ~~~~~~~~~~~ * - * Base JavaScript utilities for all Sphinx HTML documentation. + * Sphinx JavaScript utilities for all documentation. * * :copyright: Copyright 2007-2022 by the Sphinx team, see AUTHORS. * :license: BSD, see LICENSE for details. * */ -"use strict"; -const _ready = (callback) => { - if (document.readyState !== "loading") { - callback(); - } else { - document.addEventListener("DOMContentLoaded", callback); +/** + * select a different prefix for underscore + */ +$u = _.noConflict(); + +/** + * make the code below compatible with browsers without + * an installed firebug like debugger +if (!window.console || !console.firebug) { + var names = ["log", "debug", "info", "warn", "error", "assert", "dir", + "dirxml", "group", "groupEnd", "time", "timeEnd", "count", "trace", + "profile", "profileEnd"]; + window.console = {}; + for (var i = 0; i < names.length; ++i) + window.console[names[i]] = function() {}; +} + */ + +/** + * small helper function to urldecode strings + * + * See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/decodeURIComponent#Decoding_query_parameters_from_a_URL + */ +jQuery.urldecode = function(x) { + if (!x) { + return x } + return decodeURIComponent(x.replace(/\+/g, ' ')); }; /** - * highlight a given string on a node by wrapping it in - * span elements with the given class name. + * small helper function to urlencode strings */ -const _highlight = (node, addItems, text, className) => { - if (node.nodeType === Node.TEXT_NODE) { - const val = node.nodeValue; - const parent = node.parentNode; - const pos = val.toLowerCase().indexOf(text); - if ( - pos >= 0 && - !parent.classList.contains(className) && - !parent.classList.contains("nohighlight") - ) { - let span; +jQuery.urlencode = encodeURIComponent; - const closestNode = parent.closest("body, svg, foreignObject"); - const isInSVG = closestNode && closestNode.matches("svg"); - if (isInSVG) { - span = document.createElementNS("http://www.w3.org/2000/svg", "tspan"); - } else { - span = document.createElement("span"); - span.classList.add(className); - } +/** + * This function returns the parsed url parameters of the + * current request. Multiple values per key are supported, + * it will always return arrays of strings for the value parts. + */ +jQuery.getQueryParameters = function(s) { + if (typeof s === 'undefined') + s = document.location.search; + var parts = s.substr(s.indexOf('?') + 1).split('&'); + var result = {}; + for (var i = 0; i < parts.length; i++) { + var tmp = parts[i].split('=', 2); + var key = jQuery.urldecode(tmp[0]); + var value = jQuery.urldecode(tmp[1]); + if (key in result) + result[key].push(value); + else + result[key] = [value]; + } + return result; +}; - span.appendChild(document.createTextNode(val.substr(pos, text.length))); - parent.insertBefore( - span, - parent.insertBefore( +/** + * highlight a given string on a jquery object by wrapping it in + * span elements with the given class name. + */ +jQuery.fn.highlightText = function(text, className) { + function highlight(node, addItems) { + if (node.nodeType === 3) { + var val = node.nodeValue; + var pos = val.toLowerCase().indexOf(text); + if (pos >= 0 && + !jQuery(node.parentNode).hasClass(className) && + !jQuery(node.parentNode).hasClass("nohighlight")) { + var span; + var isInSVG = jQuery(node).closest("body, svg, foreignObject").is("svg"); + if (isInSVG) { + span = document.createElementNS("http://www.w3.org/2000/svg", "tspan"); + } else { + span = document.createElement("span"); + span.className = className; + } + span.appendChild(document.createTextNode(val.substr(pos, text.length))); + node.parentNode.insertBefore(span, node.parentNode.insertBefore( document.createTextNode(val.substr(pos + text.length)), - node.nextSibling - ) - ); - node.nodeValue = val.substr(0, pos); - - if (isInSVG) { - const rect = document.createElementNS( - "http://www.w3.org/2000/svg", - "rect" - ); - const bbox = parent.getBBox(); - rect.x.baseVal.value = bbox.x; - rect.y.baseVal.value = bbox.y; - rect.width.baseVal.value = bbox.width; - rect.height.baseVal.value = bbox.height; - rect.setAttribute("class", className); - addItems.push({ parent: parent, target: rect }); + node.nextSibling)); + node.nodeValue = val.substr(0, pos); + if (isInSVG) { + var rect = document.createElementNS("http://www.w3.org/2000/svg", "rect"); + var bbox = node.parentElement.getBBox(); + rect.x.baseVal.value = bbox.x; + rect.y.baseVal.value = bbox.y; + rect.width.baseVal.value = bbox.width; + rect.height.baseVal.value = bbox.height; + rect.setAttribute('class', className); + addItems.push({ + "parent": node.parentNode, + "target": rect}); + } } } - } else if (node.matches && !node.matches("button, select, textarea")) { - node.childNodes.forEach((el) => _highlight(el, addItems, text, className)); + else if (!jQuery(node).is("button, select, textarea")) { + jQuery.each(node.childNodes, function() { + highlight(this, addItems); + }); + } } -}; -const _highlightText = (thisNode, text, className) => { - let addItems = []; - _highlight(thisNode, addItems, text, className); - addItems.forEach((obj) => - obj.parent.insertAdjacentElement("beforebegin", obj.target) - ); + var addItems = []; + var result = this.each(function() { + highlight(this, addItems); + }); + for (var i = 0; i < addItems.length; ++i) { + jQuery(addItems[i].parent).before(addItems[i].target); + } + return result; }; +/* + * backward compatibility for jQuery.browser + * This will be supported until firefox bug is fixed. + */ +if (!jQuery.browser) { + jQuery.uaMatch = function(ua) { + ua = ua.toLowerCase(); + + var match = /(chrome)[ \/]([\w.]+)/.exec(ua) || + /(webkit)[ \/]([\w.]+)/.exec(ua) || + /(opera)(?:.*version|)[ \/]([\w.]+)/.exec(ua) || + /(msie) ([\w.]+)/.exec(ua) || + ua.indexOf("compatible") < 0 && /(mozilla)(?:.*? rv:([\w.]+)|)/.exec(ua) || + []; + + return { + browser: match[ 1 ] || "", + version: match[ 2 ] || "0" + }; + }; + jQuery.browser = {}; + jQuery.browser[jQuery.uaMatch(navigator.userAgent).browser] = true; +} + /** * Small JavaScript module for the documentation. */ -const Documentation = { - init: () => { - Documentation.highlightSearchWords(); - Documentation.initDomainIndexTable(); - Documentation.initOnKeyListeners(); +var Documentation = { + + init : function() { + this.fixFirefoxAnchorBug(); + this.highlightSearchWords(); + this.initIndexTable(); + if (DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS) { + this.initOnKeyListeners(); + } }, /** * i18n support */ - TRANSLATIONS: {}, - PLURAL_EXPR: (n) => (n === 1 ? 0 : 1), - LOCALE: "unknown", + TRANSLATIONS : {}, + PLURAL_EXPR : function(n) { return n === 1 ? 0 : 1; }, + LOCALE : 'unknown', // gettext and ngettext don't access this so that the functions // can safely bound to a different name (_ = Documentation.gettext) - gettext: (string) => { - const translated = Documentation.TRANSLATIONS[string]; - switch (typeof translated) { - case "undefined": - return string; // no translation - case "string": - return translated; // translation exists - default: - return translated[0]; // (singular, plural) translation tuple exists - } + gettext : function(string) { + var translated = Documentation.TRANSLATIONS[string]; + if (typeof translated === 'undefined') + return string; + return (typeof translated === 'string') ? translated : translated[0]; }, - ngettext: (singular, plural, n) => { - const translated = Documentation.TRANSLATIONS[singular]; - if (typeof translated !== "undefined") - return translated[Documentation.PLURAL_EXPR(n)]; - return n === 1 ? singular : plural; + ngettext : function(singular, plural, n) { + var translated = Documentation.TRANSLATIONS[singular]; + if (typeof translated === 'undefined') + return (n == 1) ? singular : plural; + return translated[Documentation.PLURALEXPR(n)]; }, - addTranslations: (catalog) => { - Object.assign(Documentation.TRANSLATIONS, catalog.messages); - Documentation.PLURAL_EXPR = new Function( - "n", - `return (${catalog.plural_expr})` - ); - Documentation.LOCALE = catalog.locale; + addTranslations : function(catalog) { + for (var key in catalog.messages) + this.TRANSLATIONS[key] = catalog.messages[key]; + this.PLURAL_EXPR = new Function('n', 'return +(' + catalog.plural_expr + ')'); + this.LOCALE = catalog.locale; }, /** - * highlight the search words provided in the url in the text + * add context elements like header anchor links */ - highlightSearchWords: () => { - const highlight = - new URLSearchParams(window.location.search).get("highlight") || ""; - const terms = highlight.toLowerCase().split(/\s+/).filter(x => x); - if (terms.length === 0) return; // nothing to do - - // There should never be more than one element matching "div.body" - const divBody = document.querySelectorAll("div.body"); - const body = divBody.length ? divBody[0] : document.querySelector("body"); - window.setTimeout(() => { - terms.forEach((term) => _highlightText(body, term, "highlighted")); - }, 10); - - const searchBox = document.getElementById("searchbox"); - if (searchBox === null) return; - searchBox.appendChild( - document - .createRange() - .createContextualFragment( - '" - ) - ); + addContextElements : function() { + $('div[id] > :header:first').each(function() { + $('\u00B6'). + attr('href', '#' + this.id). + attr('title', _('Permalink to this headline')). + appendTo(this); + }); + $('dt[id]').each(function() { + $('\u00B6'). + attr('href', '#' + this.id). + attr('title', _('Permalink to this definition')). + appendTo(this); + }); }, /** - * helper function to hide the search marks again + * workaround a firefox stupidity + * see: https://bugzilla.mozilla.org/show_bug.cgi?id=645075 */ - hideSearchWords: () => { - document - .querySelectorAll("#searchbox .highlight-link") - .forEach((el) => el.remove()); - document - .querySelectorAll("span.highlighted") - .forEach((el) => el.classList.remove("highlighted")); - const url = new URL(window.location); - url.searchParams.delete("highlight"); - window.history.replaceState({}, "", url); + fixFirefoxAnchorBug : function() { + if (document.location.hash && $.browser.mozilla) + window.setTimeout(function() { + document.location.href += ''; + }, 10); }, /** - * helper function to focus on search bar + * highlight the search words provided in the url in the text */ - focusSearchBar: () => { - document.querySelectorAll("input[name=q]")[0]?.focus(); + highlightSearchWords : function() { + var params = $.getQueryParameters(); + var terms = (params.highlight) ? params.highlight[0].split(/\s+/) : []; + if (terms.length) { + var body = $('div.body'); + if (!body.length) { + body = $('body'); + } + window.setTimeout(function() { + $.each(terms, function() { + body.highlightText(this.toLowerCase(), 'highlighted'); + }); + }, 10); + $('') + .appendTo($('#searchbox')); + } }, /** - * Initialise the domain index toggle buttons + * init the domain index toggle buttons */ - initDomainIndexTable: () => { - const toggler = (el) => { - const idNumber = el.id.substr(7); - const toggledRows = document.querySelectorAll(`tr.cg-${idNumber}`); - if (el.src.substr(-9) === "minus.png") { - el.src = `${el.src.substr(0, el.src.length - 9)}plus.png`; - toggledRows.forEach((el) => (el.style.display = "none")); - } else { - el.src = `${el.src.substr(0, el.src.length - 8)}minus.png`; - toggledRows.forEach((el) => (el.style.display = "")); - } - }; - - const togglerElements = document.querySelectorAll("img.toggler"); - togglerElements.forEach((el) => - el.addEventListener("click", (event) => toggler(event.currentTarget)) - ); - togglerElements.forEach((el) => (el.style.display = "")); - if (DOCUMENTATION_OPTIONS.COLLAPSE_INDEX) togglerElements.forEach(toggler); + initIndexTable : function() { + var togglers = $('img.toggler').click(function() { + var src = $(this).attr('src'); + var idnum = $(this).attr('id').substr(7); + $('tr.cg-' + idnum).toggle(); + if (src.substr(-9) === 'minus.png') + $(this).attr('src', src.substr(0, src.length-9) + 'plus.png'); + else + $(this).attr('src', src.substr(0, src.length-8) + 'minus.png'); + }).css('display', ''); + if (DOCUMENTATION_OPTIONS.COLLAPSE_INDEX) { + togglers.click(); + } }, - initOnKeyListeners: () => { - // only install a listener if it is really needed - if ( - !DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS && - !DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS - ) - return; + /** + * helper function to hide the search marks again + */ + hideSearchWords : function() { + $('#searchbox .highlight-link').fadeOut(300); + $('span.highlighted').removeClass('highlighted'); + var url = new URL(window.location); + url.searchParams.delete('highlight'); + window.history.replaceState({}, '', url); + }, - const blacklistedElements = new Set([ - "TEXTAREA", - "INPUT", - "SELECT", - "BUTTON", - ]); - document.addEventListener("keydown", (event) => { - if (blacklistedElements.has(document.activeElement.tagName)) return; // bail for input elements - if (event.altKey || event.ctrlKey || event.metaKey) return; // bail with special keys + /** + * make the url absolute + */ + makeURL : function(relativeURL) { + return DOCUMENTATION_OPTIONS.URL_ROOT + '/' + relativeURL; + }, - if (!event.shiftKey) { - switch (event.key) { - case "ArrowLeft": - if (!DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS) break; + /** + * get the current relative url + */ + getCurrentURL : function() { + var path = document.location.pathname; + var parts = path.split(/\//); + $.each(DOCUMENTATION_OPTIONS.URL_ROOT.split(/\//), function() { + if (this === '..') + parts.pop(); + }); + var url = parts.join('/'); + return path.substring(url.lastIndexOf('/') + 1, path.length - 1); + }, - const prevLink = document.querySelector('link[rel="prev"]'); - if (prevLink && prevLink.href) { - window.location.href = prevLink.href; - event.preventDefault(); + initOnKeyListeners: function() { + $(document).keydown(function(event) { + var activeElementType = document.activeElement.tagName; + // don't navigate when in search box, textarea, dropdown or button + if (activeElementType !== 'TEXTAREA' && activeElementType !== 'INPUT' && activeElementType !== 'SELECT' + && activeElementType !== 'BUTTON' && !event.altKey && !event.ctrlKey && !event.metaKey + && !event.shiftKey) { + switch (event.keyCode) { + case 37: // left + var prevHref = $('link[rel="prev"]').prop('href'); + if (prevHref) { + window.location.href = prevHref; + return false; } break; - case "ArrowRight": - if (!DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS) break; - - const nextLink = document.querySelector('link[rel="next"]'); - if (nextLink && nextLink.href) { - window.location.href = nextLink.href; - event.preventDefault(); + case 39: // right + var nextHref = $('link[rel="next"]').prop('href'); + if (nextHref) { + window.location.href = nextHref; + return false; } break; - case "Escape": - if (!DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS) break; - Documentation.hideSearchWords(); - event.preventDefault(); } } - - // some keyboard layouts may need Shift to get / - switch (event.key) { - case "/": - if (!DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS) break; - Documentation.focusSearchBar(); - event.preventDefault(); - } }); - }, + } }; // quick alias for translations -const _ = Documentation.gettext; +_ = Documentation.gettext; -_ready(Documentation.init); +$(document).ready(function() { + Documentation.init(); +}); diff --git a/docs/build/html/_static/documentation_options.js b/docs/build/html/_static/documentation_options.js index a750e4d5e..2fa8c97fe 100644 --- a/docs/build/html/_static/documentation_options.js +++ b/docs/build/html/_static/documentation_options.js @@ -1,14 +1,12 @@ var DOCUMENTATION_OPTIONS = { URL_ROOT: document.getElementById("documentation_options").getAttribute('data-url_root'), VERSION: '', - LANGUAGE: 'en', + LANGUAGE: 'None', COLLAPSE_INDEX: false, BUILDER: 'html', FILE_SUFFIX: '.html', LINK_SUFFIX: '.html', HAS_SOURCE: true, SOURCELINK_SUFFIX: '.txt', - NAVIGATION_WITH_KEYS: false, - SHOW_SEARCH_SUMMARY: true, - ENABLE_SEARCH_SHORTCUTS: false, + NAVIGATION_WITH_KEYS: false }; \ No newline at end of file diff --git a/docs/build/html/_static/language_data.js b/docs/build/html/_static/language_data.js index 2e22b06ab..ebe2f03bf 100644 --- a/docs/build/html/_static/language_data.js +++ b/docs/build/html/_static/language_data.js @@ -10,7 +10,7 @@ * */ -var stopwords = ["a", "and", "are", "as", "at", "be", "but", "by", "for", "if", "in", "into", "is", "it", "near", "no", "not", "of", "on", "or", "such", "that", "the", "their", "then", "there", "these", "they", "this", "to", "was", "will", "with"]; +var stopwords = ["a","and","are","as","at","be","but","by","for","if","in","into","is","it","near","no","not","of","on","or","such","that","the","their","then","there","these","they","this","to","was","will","with"]; /* Non-minified version is copied as a separate JS file, is available */ @@ -197,3 +197,101 @@ var Stemmer = function() { } } + + + +var splitChars = (function() { + var result = {}; + var singles = [96, 180, 187, 191, 215, 247, 749, 885, 903, 907, 909, 930, 1014, 1648, + 1748, 1809, 2416, 2473, 2481, 2526, 2601, 2609, 2612, 2615, 2653, 2702, + 2706, 2729, 2737, 2740, 2857, 2865, 2868, 2910, 2928, 2948, 2961, 2971, + 2973, 3085, 3089, 3113, 3124, 3213, 3217, 3241, 3252, 3295, 3341, 3345, + 3369, 3506, 3516, 3633, 3715, 3721, 3736, 3744, 3748, 3750, 3756, 3761, + 3781, 3912, 4239, 4347, 4681, 4695, 4697, 4745, 4785, 4799, 4801, 4823, + 4881, 5760, 5901, 5997, 6313, 7405, 8024, 8026, 8028, 8030, 8117, 8125, + 8133, 8181, 8468, 8485, 8487, 8489, 8494, 8527, 11311, 11359, 11687, 11695, + 11703, 11711, 11719, 11727, 11735, 12448, 12539, 43010, 43014, 43019, 43587, + 43696, 43713, 64286, 64297, 64311, 64317, 64319, 64322, 64325, 65141]; + var i, j, start, end; + for (i = 0; i < singles.length; i++) { + result[singles[i]] = true; + } + var ranges = [[0, 47], [58, 64], [91, 94], [123, 169], [171, 177], [182, 184], [706, 709], + [722, 735], [741, 747], [751, 879], [888, 889], [894, 901], [1154, 1161], + [1318, 1328], [1367, 1368], [1370, 1376], [1416, 1487], [1515, 1519], [1523, 1568], + [1611, 1631], [1642, 1645], [1750, 1764], [1767, 1773], [1789, 1790], [1792, 1807], + [1840, 1868], [1958, 1968], [1970, 1983], [2027, 2035], [2038, 2041], [2043, 2047], + [2070, 2073], [2075, 2083], [2085, 2087], [2089, 2307], [2362, 2364], [2366, 2383], + [2385, 2391], [2402, 2405], [2419, 2424], [2432, 2436], [2445, 2446], [2449, 2450], + [2483, 2485], [2490, 2492], [2494, 2509], [2511, 2523], [2530, 2533], [2546, 2547], + [2554, 2564], [2571, 2574], [2577, 2578], [2618, 2648], [2655, 2661], [2672, 2673], + [2677, 2692], [2746, 2748], [2750, 2767], [2769, 2783], [2786, 2789], [2800, 2820], + [2829, 2830], [2833, 2834], [2874, 2876], [2878, 2907], [2914, 2917], [2930, 2946], + [2955, 2957], [2966, 2968], [2976, 2978], [2981, 2983], [2987, 2989], [3002, 3023], + [3025, 3045], [3059, 3076], [3130, 3132], [3134, 3159], [3162, 3167], [3170, 3173], + [3184, 3191], [3199, 3204], [3258, 3260], [3262, 3293], [3298, 3301], [3312, 3332], + [3386, 3388], [3390, 3423], [3426, 3429], [3446, 3449], [3456, 3460], [3479, 3481], + [3518, 3519], [3527, 3584], [3636, 3647], [3655, 3663], [3674, 3712], [3717, 3718], + [3723, 3724], [3726, 3731], [3752, 3753], [3764, 3772], [3774, 3775], [3783, 3791], + [3802, 3803], [3806, 3839], [3841, 3871], [3892, 3903], [3949, 3975], [3980, 4095], + [4139, 4158], [4170, 4175], [4182, 4185], [4190, 4192], [4194, 4196], [4199, 4205], + [4209, 4212], [4226, 4237], [4250, 4255], [4294, 4303], [4349, 4351], [4686, 4687], + [4702, 4703], [4750, 4751], [4790, 4791], [4806, 4807], [4886, 4887], [4955, 4968], + [4989, 4991], [5008, 5023], [5109, 5120], [5741, 5742], [5787, 5791], [5867, 5869], + [5873, 5887], [5906, 5919], [5938, 5951], [5970, 5983], [6001, 6015], [6068, 6102], + [6104, 6107], [6109, 6111], [6122, 6127], [6138, 6159], [6170, 6175], [6264, 6271], + [6315, 6319], [6390, 6399], [6429, 6469], [6510, 6511], [6517, 6527], [6572, 6592], + [6600, 6607], [6619, 6655], [6679, 6687], [6741, 6783], [6794, 6799], [6810, 6822], + [6824, 6916], [6964, 6980], [6988, 6991], [7002, 7042], [7073, 7085], [7098, 7167], + [7204, 7231], [7242, 7244], [7294, 7400], [7410, 7423], [7616, 7679], [7958, 7959], + [7966, 7967], [8006, 8007], [8014, 8015], [8062, 8063], [8127, 8129], [8141, 8143], + [8148, 8149], [8156, 8159], [8173, 8177], [8189, 8303], [8306, 8307], [8314, 8318], + [8330, 8335], [8341, 8449], [8451, 8454], [8456, 8457], [8470, 8472], [8478, 8483], + [8506, 8507], [8512, 8516], [8522, 8525], [8586, 9311], [9372, 9449], [9472, 10101], + [10132, 11263], [11493, 11498], [11503, 11516], [11518, 11519], [11558, 11567], + [11622, 11630], [11632, 11647], [11671, 11679], [11743, 11822], [11824, 12292], + [12296, 12320], [12330, 12336], [12342, 12343], [12349, 12352], [12439, 12444], + [12544, 12548], [12590, 12592], [12687, 12689], [12694, 12703], [12728, 12783], + [12800, 12831], [12842, 12880], [12896, 12927], [12938, 12976], [12992, 13311], + [19894, 19967], [40908, 40959], [42125, 42191], [42238, 42239], [42509, 42511], + [42540, 42559], [42592, 42593], [42607, 42622], [42648, 42655], [42736, 42774], + [42784, 42785], [42889, 42890], [42893, 43002], [43043, 43055], [43062, 43071], + [43124, 43137], [43188, 43215], [43226, 43249], [43256, 43258], [43260, 43263], + [43302, 43311], [43335, 43359], [43389, 43395], [43443, 43470], [43482, 43519], + [43561, 43583], [43596, 43599], [43610, 43615], [43639, 43641], [43643, 43647], + [43698, 43700], [43703, 43704], [43710, 43711], [43715, 43738], [43742, 43967], + [44003, 44015], [44026, 44031], [55204, 55215], [55239, 55242], [55292, 55295], + [57344, 63743], [64046, 64047], [64110, 64111], [64218, 64255], [64263, 64274], + [64280, 64284], [64434, 64466], [64830, 64847], [64912, 64913], [64968, 65007], + [65020, 65135], [65277, 65295], [65306, 65312], [65339, 65344], [65371, 65381], + [65471, 65473], [65480, 65481], [65488, 65489], [65496, 65497]]; + for (i = 0; i < ranges.length; i++) { + start = ranges[i][0]; + end = ranges[i][1]; + for (j = start; j <= end; j++) { + result[j] = true; + } + } + return result; +})(); + +function splitQuery(query) { + var result = []; + var start = -1; + for (var i = 0; i < query.length; i++) { + if (splitChars[query.charCodeAt(i)]) { + if (start !== -1) { + result.push(query.slice(start, i)); + start = -1; + } + } else if (start === -1) { + start = i; + } + } + if (start !== -1) { + result.push(query.slice(start)); + } + return result; +} + + diff --git a/docs/build/html/_static/pygments.css b/docs/build/html/_static/pygments.css index 84ab3030a..08bec689d 100644 --- a/docs/build/html/_static/pygments.css +++ b/docs/build/html/_static/pygments.css @@ -17,7 +17,6 @@ span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: .highlight .cs { color: #3D7B7B; font-style: italic } /* Comment.Special */ .highlight .gd { color: #A00000 } /* Generic.Deleted */ .highlight .ge { font-style: italic } /* Generic.Emph */ -.highlight .ges { font-weight: bold; font-style: italic } /* Generic.EmphStrong */ .highlight .gr { color: #E40000 } /* Generic.Error */ .highlight .gh { color: #000080; font-weight: bold } /* Generic.Heading */ .highlight .gi { color: #008400 } /* Generic.Inserted */ diff --git a/docs/build/html/_static/searchtools.js b/docs/build/html/_static/searchtools.js index ac4d5861f..2d7785937 100644 --- a/docs/build/html/_static/searchtools.js +++ b/docs/build/html/_static/searchtools.js @@ -8,20 +8,18 @@ * :license: BSD, see LICENSE for details. * */ -"use strict"; -/** - * Simple result scoring code. - */ -if (typeof Scorer === "undefined") { +if (!Scorer) { + /** + * Simple result scoring code. + */ var Scorer = { // Implement the following function to further tweak the score for each result - // The function takes a result array [docname, title, anchor, descr, score, filename] + // The function takes a result array [filename, title, anchor, descr, score] // and returns the new score. /* - score: result => { - const [docname, title, anchor, descr, score, filename] = result - return score + score: function(result) { + return result[4]; }, */ @@ -30,11 +28,9 @@ if (typeof Scorer === "undefined") { // or matches in the last dotted part of the object name objPartialMatch: 6, // Additive scores depending on the priority of the object - objPrio: { - 0: 15, // used to be importantResults - 1: 5, // used to be objectResults - 2: -5, // used to be unimportantResults - }, + objPrio: {0: 15, // used to be importantResults + 1: 5, // used to be objectResults + 2: -5}, // used to be unimportantResults // Used when the priority is not in the mapping. objPrioDefault: 0, @@ -43,455 +39,456 @@ if (typeof Scorer === "undefined") { partialTitle: 7, // query found in terms term: 5, - partialTerm: 2, + partialTerm: 2 }; } -const _removeChildren = (element) => { - while (element && element.lastChild) element.removeChild(element.lastChild); -}; - -/** - * See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_Expressions#escaping - */ -const _escapeRegExp = (string) => - string.replace(/[.*+\-?^${}()|[\]\\]/g, "\\$&"); // $& means the whole matched string - -const _displayItem = (item, highlightTerms, searchTerms) => { - const docBuilder = DOCUMENTATION_OPTIONS.BUILDER; - const docUrlRoot = DOCUMENTATION_OPTIONS.URL_ROOT; - const docFileSuffix = DOCUMENTATION_OPTIONS.FILE_SUFFIX; - const docLinkSuffix = DOCUMENTATION_OPTIONS.LINK_SUFFIX; - const showSearchSummary = DOCUMENTATION_OPTIONS.SHOW_SEARCH_SUMMARY; - - const [docName, title, anchor, descr] = item; - - let listItem = document.createElement("li"); - let requestUrl; - let linkUrl; - if (docBuilder === "dirhtml") { - // dirhtml builder - let dirname = docName + "/"; - if (dirname.match(/\/index\/$/)) - dirname = dirname.substring(0, dirname.length - 6); - else if (dirname === "index/") dirname = ""; - requestUrl = docUrlRoot + dirname; - linkUrl = requestUrl; - } else { - // normal html builders - requestUrl = docUrlRoot + docName + docFileSuffix; - linkUrl = docName + docLinkSuffix; - } - const params = new URLSearchParams(); - params.set("highlight", [...highlightTerms].join(" ")); - let linkEl = listItem.appendChild(document.createElement("a")); - linkEl.href = linkUrl + "?" + params.toString() + anchor; - linkEl.innerHTML = title; - if (descr) - listItem.appendChild(document.createElement("span")).innerText = - " (" + descr + ")"; - else if (showSearchSummary) - fetch(requestUrl) - .then((responseData) => responseData.text()) - .then((data) => { - if (data) - listItem.appendChild( - Search.makeSearchSummary(data, searchTerms, highlightTerms) - ); - }); - Search.output.appendChild(listItem); -}; -const _finishSearch = (resultCount) => { - Search.stopPulse(); - Search.title.innerText = _("Search Results"); - if (!resultCount) - Search.status.innerText = Documentation.gettext( - "Your search did not match any documents. Please make sure that all words are spelled correctly and that you've selected enough categories." - ); - else - Search.status.innerText = _( - `Search finished, found ${resultCount} page(s) matching the search query.` - ); -}; -const _displayNextItem = ( - results, - resultCount, - highlightTerms, - searchTerms -) => { - // results left, load the summary and display it - // this is intended to be dynamic (don't sub resultsCount) - if (results.length) { - _displayItem(results.pop(), highlightTerms, searchTerms); - setTimeout( - () => _displayNextItem(results, resultCount, highlightTerms, searchTerms), - 5 - ); +if (!splitQuery) { + function splitQuery(query) { + return query.split(/\s+/); } - // search finished, update title and status message - else _finishSearch(resultCount); -}; - -/** - * Default splitQuery function. Can be overridden in ``sphinx.search`` with a - * custom function per language. - * - * The regular expression works by splitting the string on consecutive characters - * that are not Unicode letters, numbers, underscores, or emoji characters. - * This is the same as ``\W+`` in Python, preserving the surrogate pair area. - */ -if (typeof splitQuery === "undefined") { - var splitQuery = (query) => query - .split(/[^\p{Letter}\p{Number}_\p{Emoji_Presentation}]+/gu) - .filter(term => term) // remove remaining empty strings } /** * Search Module */ -const Search = { - _index: null, - _queued_query: null, - _pulse_status: -1, - - htmlToText: (htmlString) => { - const htmlElement = document - .createRange() - .createContextualFragment(htmlString); - _removeChildren(htmlElement.querySelectorAll(".headerlink")); - const docContent = htmlElement.querySelector('[role="main"]'); - if (docContent !== undefined) return docContent.textContent; - console.warn( - "Content block not found. Sphinx search tries to obtain it via '[role=main]'. Could you check your theme or template." - ); - return ""; +var Search = { + + _index : null, + _queued_query : null, + _pulse_status : -1, + + htmlToText : function(htmlString) { + var virtualDocument = document.implementation.createHTMLDocument('virtual'); + var htmlElement = $(htmlString, virtualDocument); + htmlElement.find('.headerlink').remove(); + docContent = htmlElement.find('[role=main]')[0]; + if(docContent === undefined) { + console.warn("Content block not found. Sphinx search tries to obtain it " + + "via '[role=main]'. Could you check your theme or template."); + return ""; + } + return docContent.textContent || docContent.innerText; }, - init: () => { - const query = new URLSearchParams(window.location.search).get("q"); - document - .querySelectorAll('input[name="q"]') - .forEach((el) => (el.value = query)); - if (query) Search.performSearch(query); + init : function() { + var params = $.getQueryParameters(); + if (params.q) { + var query = params.q[0]; + $('input[name="q"]')[0].value = query; + this.performSearch(query); + } }, - loadIndex: (url) => - (document.body.appendChild(document.createElement("script")).src = url), + loadIndex : function(url) { + $.ajax({type: "GET", url: url, data: null, + dataType: "script", cache: true, + complete: function(jqxhr, textstatus) { + if (textstatus != "success") { + document.getElementById("searchindexloader").src = url; + } + }}); + }, - setIndex: (index) => { - Search._index = index; - if (Search._queued_query !== null) { - const query = Search._queued_query; - Search._queued_query = null; - Search.query(query); + setIndex : function(index) { + var q; + this._index = index; + if ((q = this._queued_query) !== null) { + this._queued_query = null; + Search.query(q); } }, - hasIndex: () => Search._index !== null, - - deferQuery: (query) => (Search._queued_query = query), + hasIndex : function() { + return this._index !== null; + }, - stopPulse: () => (Search._pulse_status = -1), + deferQuery : function(query) { + this._queued_query = query; + }, - startPulse: () => { - if (Search._pulse_status >= 0) return; + stopPulse : function() { + this._pulse_status = 0; + }, - const pulse = () => { + startPulse : function() { + if (this._pulse_status >= 0) + return; + function pulse() { + var i; Search._pulse_status = (Search._pulse_status + 1) % 4; - Search.dots.innerText = ".".repeat(Search._pulse_status); - if (Search._pulse_status >= 0) window.setTimeout(pulse, 500); - }; + var dotString = ''; + for (i = 0; i < Search._pulse_status; i++) + dotString += '.'; + Search.dots.text(dotString); + if (Search._pulse_status > -1) + window.setTimeout(pulse, 500); + } pulse(); }, /** * perform a search for something (or wait until index is loaded) */ - performSearch: (query) => { + performSearch : function(query) { // create the required interface elements - const searchText = document.createElement("h2"); - searchText.textContent = _("Searching"); - const searchSummary = document.createElement("p"); - searchSummary.classList.add("search-summary"); - searchSummary.innerText = ""; - const searchList = document.createElement("ul"); - searchList.classList.add("search"); - - const out = document.getElementById("search-results"); - Search.title = out.appendChild(searchText); - Search.dots = Search.title.appendChild(document.createElement("span")); - Search.status = out.appendChild(searchSummary); - Search.output = out.appendChild(searchList); - - const searchProgress = document.getElementById("search-progress"); - // Some themes don't use the search progress node - if (searchProgress) { - searchProgress.innerText = _("Preparing search..."); - } - Search.startPulse(); + this.out = $('#search-results'); + this.title = $('

    ' + _('Searching') + '

    ').appendTo(this.out); + this.dots = $('').appendTo(this.title); + this.status = $('

     

    ').appendTo(this.out); + this.output = $(' @@ -429,13 +421,13 @@
    -

    Trainer DIAL

    +

    Trainer DIAL

    -

    Domain Invariant Adversarial Learning

    +

    Domain Invariant Adversarial Learning

    The algorithm introduced in https://arxiv.org/pdf/2104.00322.pdf uses adversarial learning to tackle the task of domain generalization. Therefore, the source domain is the natural dataset, while the target domain is generated using adversarial attack on the source domain.

    -

    generating the adversarial domain

    +

    generating the adversarial domain

    The generation of adversary images is demonstrated in figure 1. The task is to find an adversary image \(x'\) to the natural image \(x\) with \(||x- x'||\) small, such that the output of a classification network \(\phi\) fulfills \(||\phi(x) - \phi(x')||\) big. In the example in figure 1 you can for example see, that the difference between the left and the right image of the panda is unobservable, but the classifier does still classify them differently.

    In Domainlab the adversary images are created starting from a random perturbation of the natural image \(x'_0 = x + \sigma \tilde{x}~\), \(\tilde{x} \sim \mathcal{N}(0, 1)\) and using \(n\) steps in a gradient descend with step size \(\tau\) to maximize \(||\phi(x) - \phi(x')||\). In general machine learning, the generation of adversary images is used during the training process to make networks more robust to adversarial attacks.

    @@ -445,7 +437,7 @@

    generating the adversarial domain

    -

    network structure

    +

    network structure

    The network consists of three parts. At first a feature extractor, which extracts the main characteristics of the images. This features are then used as the input to a label classifier and a domain classifier. During training the network is optimized to a have low error on the classification task, while ensuring that the internal representation (output of the feature extractor) cannot discriminate between the natural and adversarial domain. This goal can be archived by using a special loss function in combination with a gradient reversal layer.

    @@ -454,7 +446,7 @@

    network structure -

    loss function and gradient reversal layer

    +

    loss function and gradient reversal layer

    The loss function for in the DomainLab package is different to the one described in the paper. It consists of the standard cross entropy loss between the predicted label probabilities and the actual label for the natural domain (\(CE_{nat}\)) and for the adversarial domain (\(CE_{adv}\)). The adversarial domain is weighted by the parameter \(\gamma_\text{reg}\).

    @@ -473,7 +465,7 @@

    loss function and gradient re

    -

    Examples

    +

    Examples

    python main_out.py --te_d=0 --task=mnistcolor10 --model=erm --trainer=dial --nname=conv_bn_pool_2
     
    @@ -481,13 +473,13 @@

    Examples -

    Train DIVA model with DIAL trainer

    +

    Train DIVA model with DIAL trainer

    python main_out.py --te_d 0 1 2 --tr_d 3 7 --task=mnistcolor10 --model=diva --nname=conv_bn_pool_2 --nname_dom=conv_bn_pool_2 --gamma_y=7e5 --gamma_d=1e5 --trainer=dial
     
    -

    Set hyper-parameters for trainer as well

    +

    Set hyper-parameters for trainer as well

    python main_out.py --te_d 0 1 2 --tr_d 3 7 --task=mnistcolor10 --model=diva --nname=conv_bn_pool_2 --nname_dom=conv_bn_pool_2 --gamma_y=7e5 --gamma_d=1e5 --trainer=dial --dial_steps_perturb=1
     
    @@ -542,7 +534,7 @@

    Set hyper-parameters for train Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/docFishr.html b/docs/build/html/docFishr.html index fbeedcb71..e12d5047d 100644 --- a/docs/build/html/docFishr.html +++ b/docs/build/html/docFishr.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -352,13 +351,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -400,9 +392,9 @@
    -

    Trainer Fishr

    +

    Trainer Fishr

    -

    Invariant Gradient Variances for Out-of-distribution Generalization

    +

    Invariant Gradient Variances for Out-of-distribution Generalization

    The goal of the Fishr regularization technique is locally aligning the domain-level loss landscapes around the final weights, finding a minimizer around which the inconsistencies between the domain-level loss landscapes are as small as possible. @@ -415,7 +407,7 @@

    Inv Invariant gradient variances for out-of-distribution generalization")
    -

    Quantifying inconsistency between domains

    +

    Quantifying inconsistency between domains

    Intuitively, two domains are locally inconsistent around a minimizer, if a small perturbation of the minimizer highly affects its optimality in one domain, but only minimally affects its optimality in the other domain. Under certain assumptions, most importantly @@ -435,7 +427,7 @@

    Quantifying inconsistency bet domain-level Hessians, matching the variances across domains.

    -

    Matching the Variances during training

    +

    Matching the Variances during training

    Let \(\mathcal{E}\) be the space of all training domains, and let \(\mathcal{R}_e(\theta)\) be the ERM objective. Fishr minimizes the following objective function during training:

    @@ -452,7 +444,7 @@

    Matching the Variances during tr \(v = \frac{1}{|\mathcal{E}|}\sum_{e\in\mathcal{E}} v_e\).

    -

    Implementation

    +

    Implementation

    The variance of the gradients within each domain can be computed with the BACKPACK package (see: Dangel, Felix, Frederik Kunstner, and Philipp Hennig. “Backpack: Packing more into backprop.” https://arxiv.org/abs/1912.10985). @@ -463,7 +455,7 @@

    Implementation -

    Examples

    +

    Examples

    python main_out.py --te_d=0 --task=mini_vlcs --model=erm --trainer=fishr --nname=alexnet --bs=2 --nocu
     
    @@ -521,7 +513,7 @@

    ExamplesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/docHDUVA.html b/docs/build/html/docHDUVA.html index 0f86b3474..2f5b57b9e 100644 --- a/docs/build/html/docHDUVA.html +++ b/docs/build/html/docHDUVA.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -405,13 +404,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -471,13 +463,13 @@
    -

    Model HDUVA

    +

    Model HDUVA

    -

    HDUVA: HIERARCHICAL VARIATIONAL AUTO-ENCODING FOR UNSUPERVISED DOMAIN GENERALIZATION

    +

    HDUVA: HIERARCHICAL VARIATIONAL AUTO-ENCODING FOR UNSUPERVISED DOMAIN GENERALIZATION

    HDUVA builds on a generative approach within the framework of variational autoencoders to facilitate generalization to new domains without supervision. HDUVA learns representations that disentangle domain-specific information from class-label specific information even in complex settings where domain structure is not observed during training.

    -

    Model Overview

    +

    Model Overview

    More specifically, HDUVA is based on three latent variables that are used to model distinct sources of variation and are denoted as \(z_y\), \(z_d\) and \(z_x\). \(z_y\) represents class specific information, \(z_d\) represents domain specific information and \(z_x\) models residual variance of the input. We introduce an additional hierarchical level and use a continuous latent representation s to model (potentially unobserved) domain structure. This means that we can encourage disentanglement of the latent variables through conditional priors without the need of conditioning on a one-hot-encoded, observed domain label. The model along with its parameters and hyperparameters is shown in Figure 1:

    -

    Evidence lower bound and overall loss

    +

    Evidence lower bound and overall loss

    The ELBO of the model can be decomposed into 4 different terms:

    Likelihood: \(E_{q(z_d, s|x), q(z_x|x), q(z_y|x)}\log p_{\theta}(x|s, z_d, z_x, z_y)\)

    KL divergence weighted as in the Beta-VAE: \(-\beta_x KL(q_{\phi_x}(z_x|x)||p_{\theta_x}(z_x)) - \beta_y KL(q_{\phi_y}(z_y|x)||p_{\theta_y}(z_y|y))\)

    @@ -495,11 +487,11 @@

    Evidence lower bound and overall

    In addition, we construct the overall loss by adding an auxiliary classsifier, by adding an additional term to the ELBO loss, weighted with \(\gamma_y\):

    -

    Hyperparameters loss function

    +

    Hyperparameters loss function

    For fitting the model, we need to specify the 4 \(\beta\)-weights related to the the different terms of the ELBO ( \(\beta_x\) , \(\beta_y\), \(\beta_d\), \(\beta_t\)) as well as \(\gamma_y\).

    -

    Model hyperparameters

    +

    Model hyperparameters

    In addition to these hyperparameters, the following model parameters can be specified:

    • zd_dim: size of latent space for domain-specific information

    • @@ -521,7 +513,7 @@

      Model hyperparameters -

      Hyperparameter for warmup

      +

      Hyperparameter for warmup

      Finally, the number of epochs for hyper-parameter warm-up can be specified via the argument warmup.

      Please cite our paper if you find it useful!

      @inproceedings{sun2021hierarchical,
      @@ -534,45 +526,45 @@ 

      Hyperparameter for warmup -

      Examples

      +

      Examples

      -

      hduva use custom net for sandwich encoder

      +

      hduva use custom net for sandwich encoder

      python main_out.py --te_d=caltech --bs=2 --task=mini_vlcs --model=hduva --nname=conv_bn_pool_2 --gamma_y=7e5 --nname_encoder_x2topic_h=conv_bn_pool_2 --npath_encoder_sandwich_x2h4zd=examples/nets/resnet.py
       
      -

      hduva use custom net for topic encoder

      +

      hduva use custom net for topic encoder

      python main_out.py --te_d=caltech --bs=2 --task=mini_vlcs --model=hduva --nname=conv_bn_pool_2 --gamma_y=7e5 --npath_encoder_x2topic_h=examples/nets/resnet.py --nname_encoder_sandwich_x2h4zd=conv_bn_pool_2
       
      -

      hduva use custom net for classification encoder

      +

      hduva use custom net for classification encoder

      python main_out.py --te_d=caltech --bs=2 --task=mini_vlcs --model=hduva --npath=examples/nets/resnet.py --gamma_y=7e5 --nname_encoder_x2topic_h=conv_bn_pool_2 --nname_encoder_sandwich_x2h4zd=conv_bn_pool_2
       
      -

      use hduva on color mnist, train on 2 domains

      +

      use hduva on color mnist, train on 2 domains

      python main_out.py --tr_d 0 1 2 --te_d 3 --bs=2 --task=mnistcolor10 --model=hduva  --nname=conv_bn_pool_2 --gamma_y=7e5 --nname_encoder_x2topic_h=conv_bn_pool_2 --nname_encoder_sandwich_x2h4zd=conv_bn_pool_2
       
      -

      hduva is domain-unsupervised, so it works also with a single domain

      +

      hduva is domain-unsupervised, so it works also with a single domain

      python main_out.py --tr_d 0  --te_d 3 4 --bs=2 --task=mnistcolor10 --model=hduva --nname=conv_bn_pool_2 --gamma_y=7e5 --nname_encoder_x2topic_h=conv_bn_pool_2 --nname_encoder_sandwich_x2h4zd=conv_bn_pool_2
       
      -

      hduva with implemented neural network

      +

      hduva with implemented neural network

      python main_out.py --te_d=caltech --bs=2 --task=mini_vlcs --model=hduva --nname=conv_bn_pool_2 --gamma_y=7e5 --nname_encoder_x2topic_h=conv_bn_pool_2 --nname_encoder_sandwich_x2h4zd=conv_bn_pool_2
       
      -

      hduva use alex net

      +

      hduva use alex net

      python main_out.py --te_d=caltech --bs=2 --task=mini_vlcs --model=hduva --nname=conv_bn_pool_2 --gamma_y=7e5 --nname_encoder_x2topic_h=conv_bn_pool_2 --nname_encoder_sandwich_x2h4zd=alexnet
       
      @@ -627,7 +619,7 @@

      hduva use alex netSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/docIRM.html b/docs/build/html/docIRM.html index 06cb5ef9a..074c8dad6 100644 --- a/docs/build/html/docIRM.html +++ b/docs/build/html/docIRM.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -344,13 +343,6 @@ Causal IRL - -
    • - - - Invariant Causal Mechanisms through Distribution Matching - -
    @@ -384,7 +376,7 @@
    -

    Trainer IRM (Invariant Risk Minimization)

    +

    Trainer IRM (Invariant Risk Minimization)

    Decompose a classification task into feature extraction \(\Phi(\cdot)\) and classificaiton layer \(w(\cdot)\), then the task loss is

    \(\ell^{(d)} (w \circ \Phi) = \mathbb{E}_{(X, Y) \sim \mathcal{D}_d}[\ell(w \circ \Phi(X), Y)]\) where we use \(\ell\) to denote the cross entropy for a classification task, and \(\mathcal{D}_d\) for distribution of domain \(d\).

    @@ -396,7 +388,7 @@

    Trainer IRM (Invariant Risk Minimization)\(\Phi(\cdot)\) then get optimized under this constraint.

    Thus IRM forms a bi-level optimization by jointly optimize \(\Phi\) and \(w\) which is hard to solve, so in practice IRMv1 is used.

    -

    IRMv1

    +

    IRMv1

    In DomainLab, we write the loss function as $\(\ell(\cdot) + \lambda R(\cdot)\)$, which result in the optmization below:

    \[\min_{\Phi, w} \sum_{d} \ell^{(d)}(w \circ \Phi) + \lambda \sum_{d} \|\nabla_{w|w=1.0} \ell^{(d)}(w \circ \Phi)\|^2\]
    @@ -405,7 +397,7 @@

    IRMv1\(\nabla_{w|w=1} \ell(w \circ \Phi(X^{(d, i)}), Y^{(d, i)})\) of dimension dim(Grad) with \(\nabla_{w|w=1} \ell(w \circ \Phi(X^{(d, j)}), Y^{(d, j)})\) of dimension dim(Grad) For more details, see section 3.2 and Appendix D of : Arjovsky et al., “Invariant Risk Minimization.”

    -

    Examples

    +

    Examples

    python main_out.py --te_d=0 --task=mnistcolor10 --model=erm --trainer=irm --nname=conv_bn_pool_2
     
    @@ -459,7 +451,7 @@

    ExamplesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/docJiGen.html b/docs/build/html/docJiGen.html index 40cf303a5..34ab0e456 100644 --- a/docs/build/html/docJiGen.html +++ b/docs/build/html/docJiGen.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -350,13 +349,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -396,7 +388,7 @@
    -

    Model JiGen

    +

    Model JiGen

    @@ -417,7 +409,7 @@

    Model JiGen -

    Model parameters

    +

    Model parameters

    The following hyperparameters can be specified:

    • nperm: number of patches in a permutation

    • @@ -429,15 +421,15 @@

      Model parameters -

      Examples

      +

      Examples

      -

      model jigen with implemented neural network

      +

      model jigen with implemented neural network

      python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=8 --model=jigen --nname=alexnet --pperm=1 --nperm=100 --grid_len=3
       
      -

      sannity check with jigen tile shuffling

      +

      sannity check with jigen tile shuffling

      python main_out.py --te_d=sketch --tpath=examples/tasks/demo_task_path_list_small.py --debug --bs=8 --model=jigen --nname=alexnet --pperm=1 --nperm=100 --grid_len=3 --san_check
       
      @@ -492,7 +484,7 @@

      sannity check with jigen tile s

    Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/docMA.html b/docs/build/html/docMA.html index 9f7be2a50..b34bac17c 100644 --- a/docs/build/html/docMA.html +++ b/docs/build/html/docMA.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -319,13 +318,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -350,7 +342,7 @@
    -

    Simple Moving Average

    +

    Simple Moving Average

    For each epoch, convex combine the weights for each layey from Paper: Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalization, Devansh Arpit, Huan Wang, Yingbo Zhou, Caiming Xiong, Salesforce Research, USA

    Example:

    @@ -406,7 +398,7 @@

    Simple Moving AverageSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/docMatchDG.html b/docs/build/html/docMatchDG.html index 4a1305da3..8d414b223 100644 --- a/docs/build/html/docMatchDG.html +++ b/docs/build/html/docMatchDG.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -381,13 +380,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -437,13 +429,13 @@
    -

    Trainer MatchDG

    +

    Trainer MatchDG

    -

    Domain Generalization using Causal Matching

    +

    Domain Generalization using Causal Matching

    This algorithm introduced in https://arxiv.org/pdf/2006.07500.pdf is motivated by causality theory. The authors try to enforce, that a model does classify an image only on the object information included in the image and not on the domain information.

    -

    Motivation: causality theory

    +

    Motivation: causality theory

    The authors of the paper motivate their approach by looking at the data-generation process. The underlying causal model (SCM) is given in figure 1. In the graphic one starts of from the object \(O\) and the domain D. The object is directly influenced by its true label \(y_\text{true}\) while label and domain do only correlate with each other. Additionally the object is correlated with the domain conditioned on \(y_\text{true}\). The information from the object \(O\) and the domain \(D\) do together form the image \(x\) which shall be classified by the neuronal network. Doing so, the object does contribute to the image by providing high-level causal features \(x_C\) that are common to any image of the same object. This features are the key for classifying the object, as there is only subliminal influence of the domain, therefore the prediction \(y\) is only depending on this causal features. The second contribution to the image are domain-dependent high-level features of the object \(x_A\), which depend on both, the object \(O\) and the domain \(D\). This domain-dependent features shall not be respected in the classification as there is a high influence of the domain.

    @@ -451,7 +443,7 @@

    Motivation: causality theory -

    Network

    +

    Network

    Before defining the network, one needs to define three sets:

    -

    Training

    +

    Training

    Initialisation: first of all match pairs of same-class data points from different domains are constructed. Given a data point, another data point with the same label from a different domain is selected randomly. The matching across domains is done relative to a base domain, which is chosen as the domain with the highest number of samples for that class. This leads to a matched data matrix \(\mathcal{M}\) of size \((N', K)\) with \(N'\) sum of the size of base domains over all classes and \(K\) number ob domains.

    Phase 1: sample batches \((B, K)\) from \(\mathcal{M}\), with \(B\) is batch size and train a match function \(\Omega: \mathcal{X} \times \mathcal{X} \rightarrow \{0, 1\}\), by adapting the network parameter in \(\phi\) to minimize the contrastive loss for every positive match pair \((x_j, x_k)\)

    @@ -500,27 +492,27 @@

    Training -

    Examples

    +

    Examples

    -

    trainer matchdg with custom neural network

    +

    trainer matchdg with custom neural network

    python main_out.py --te_d=caltech --task=mini_vlcs --bs=2 --model=erm --trainer=matchdg --epochs_ctr=3 --epos=6 --npath=examples/nets/resnet.py
     
    -

    training hduva with matchdg

    +

    training hduva with matchdg

    python main_out.py --te_d 0 1 2 --tr_d 3 7 --task=mnistcolor10 --bs=2 --model=hduva --trainer=matchdg --epochs_ctr=3 --epos=6 --nname=conv_bn_pool_2 --gamma_y=7e5 --nname_encoder_x2topic_h=conv_bn_pool_2 --nname_encoder_sandwich_x2h4zd=conv_bn_pool_2
     
    -

    training implemented neural network with matchdg

    +

    training implemented neural network with matchdg

    python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=2 --model=erm --trainer=matchdg --epochs_ctr=3 --epos=6 --nname=alexnet
     
    -

    trainer matchdg with mnist

    +

    trainer matchdg with mnist

     python main_out.py --te_d 0 1 2 --tr_d 3 7 --task=mnistcolor10 --model=erm --trainer=matchdg --nname=conv_bn_pool_2 --epochs_ctr=2 --epos=6
     
    @@ -575,7 +567,7 @@

    trainer matchdg with mnistSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_MNIST_classification.html b/docs/build/html/doc_MNIST_classification.html index 3023e0e6f..a38d7de5d 100644 --- a/docs/build/html/doc_MNIST_classification.html +++ b/docs/build/html/doc_MNIST_classification.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -358,10 +350,10 @@
    -

    colored MNIST classification

    +

    colored MNIST classification

    We include in the DomainLab package colored verion of MNIST where the color corresponds to the domain and digit corresponds to the semantic concept that we want to classify.

    -

    colored MNIST dataset

    +

    colored MNIST dataset

    We provide 10 different colored version of the MNIST dataset with numbers 0 to 9 as 10 different domains. The digit and background are colored differently, thus a domain correspond to a 2-color combination. An extraction of digit 0 to 9 from domain 0 is shown in Figure 1.

    @@ -397,23 +389,23 @@

    colored MNIST dataset -

    domain generalisation on colored MNIST

    +

    domain generalisation on colored MNIST

    A particular hard task for domain generalization is, if only a few training domains are available and the test domain differs a lot from the train domains. Here we use domain 0 and 3, from Figure 2, for the training, while choosing domain 1 and 2 for testing as the colors appearing here are far different from the ones used in training.

    For our test we like to compare diva and erm, this was done using the following command prompts:

    -

    erm (Emperical Risk Minimization)

    +

    erm (Emperical Risk Minimization)

    python main_out.py --te_d 1 2 --tr_d 0 3 --task=mnistcolor10 --epos=500 --bs=16 --model=erm --nname=conv_bn_pool_2 --lr=1e-3 --seed=0 --san_check --san_num=8
     
    -

    diva

    +

    diva

    python main_out.py --te_d 1 2 --tr_d 0 3 --task=mnistcolor10 --epos=500 --bs=16 --model=diva --nname=conv_bn_pool_2 --nname_dom=conv_bn_pool_2 --lr=1e-3 --seed=0 --gamma_y=1e5 --gamma_d=1e5 --san_check --san_num=8
     
    -

    Results

    +

    Results

    For both algorithms the early stop criterion ended the training. The performance of the trained models on the test domains are summarized in the following table:

    @@ -450,7 +442,7 @@

    Results -

    Detailed prompt explanation

    +

    Detailed prompt explanation

    • --te_d 1 2 sets the test domain to domain 1 and 2

    • --tr_d 0 3 sets the train domain to domain 0 and 3

    • @@ -506,7 +498,7 @@

      Detailed prompt explanationSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_benchmark.html b/docs/build/html/doc_benchmark.html index 335bf64ad..2fdfffa28 100644 --- a/docs/build/html/doc_benchmark.html +++ b/docs/build/html/doc_benchmark.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
    • - - - Invariant Causal Mechanisms through Distribution Matching - -
    @@ -378,7 +370,7 @@
    -

    Benchmarking with DomainLab

    +

    Benchmarking with DomainLab

    Documentation for Benchmark in Markdown

    The package offers the ability to benchmark different user-defined experiments against each other, as well as against different hyperparameter settings and random seeds. @@ -390,10 +382,10 @@

    Benchmarking with DomainLab -

    Dependencies installation

    +

    Dependencies installation

    DomainLab relies on Snakemake for its benchmark functionality.

    -

    Unix installation

    +

    Unix installation

    snakemake depends on pulp due to an upgrde of pulp, snakemake becomes unstable, so we recommed install the following version.

    pip install snakemake==7.32.0
     pip install pulp==2.7.0
    @@ -401,7 +393,7 @@ 

    Unix installation -

    Windows installation details

    +

    Windows installation details

    Benchmarking is currently not tested on Windows due to the dependency on Snakemake and datrie One could, however, try install minimal Snakemake via mamba create -c bioconda -c conda-forge -n snakemake snakemake-minimal @@ -409,12 +401,12 @@

    Windows installation details -

    Setting up a benchmark

    +

    Setting up a benchmark

    The benchmark is configured in a yaml file. We refer to doc_benchmark_yaml.md for a documented example.

    -

    Running a benchmark

    +

    Running a benchmark

    For the execution of a benchmark we provide two scripts in our repository:

    • running the benchmark on a standalone machine (computation node): @@ -422,7 +414,7 @@

      Running a benchmarkrun_benchmark_slurm.sh

    -

    Benchmark on a standalone machine/computation node (with or without GPU)

    +

    Benchmark on a standalone machine/computation node (with or without GPU)

    To run the benchmark with a specific configuration on a standalone machine, inside the DomainLab folder, one can execute (we assume you have a machine with 4 cores or more)

    # Note: this has only been tested on Linux based systems and may not work on Windows
    @@ -437,7 +429,7 @@ 

    rm -r .snakemake/

    -

    Benchmark on a HPC cluster with slurm

    +

    Benchmark on a HPC cluster with slurm

    If you have access to an HPC cluster with slurm support: In a submission node, clone the DomainLab repository, cd into the repository and execute the following command:

    Make sure to use tool like nohup or tmux to keep the following command active!

    @@ -449,7 +441,7 @@

    Benchmark on a HPC cluster with s

    Similar to the local version explained above, the user can also specify a random seed for hyperparameter sampling and pytorch.

    -

    Check errors for slurm runs

    +

    Check errors for slurm runs

    The following script will help to find out which job has failed and the error message, so that you could direct to the specific log file

    bash ./sh_list_error.sh ./zoutput/benchmarks/[output folder of the sepcifed benchmark in the yaml file]/slurm_logs
    @@ -457,7 +449,7 @@ 

    Check errors for slurm runs -

    Map between slurm job id and sampled hyperparameter index

    +

    Map between slurm job id and sampled hyperparameter index

    suppose the slurm job id is 14144163, one could the corresponding log file in ./zoutput/[output folder of the sepcifed benchmark in the yaml file]/slurm_logs folder via find . | grep -i "14144163"

    the results can be @@ -467,13 +459,13 @@

    Map between s

    -

    Obtained results

    +

    Obtained results

    All files created by this benchmark are saved in the given output directory (by default ./zoutput/benchmarks/[name of the benchmark defined in YAML file]). The sampled hyperparameters can be found in hyperparameters.csv. The yaml file is translated to config.txt with corresponding to commit in formation in commit.txt (do not update code during benchmark process so results can be reproducible with this commit information), corresponding to each line in hyperparameters.csv, there will be a csv file in directory rule_results.

    -

    Output folder structure

    +

    Output folder structure

    via tree -L 2 in zoutput/benchmarks/[name of the benchmark defined in configuration yaml file], one can get something like below

    ├── commit.txt
     ├── config.txt
    @@ -509,7 +501,7 @@ 

    Output folder structure

    -

    Obtain partial results

    +

    Obtain partial results

    If the benchmark is not yet completed (still running or has some failed jobs, e.g. BrokenPipe Error due to multiprocessing in PIL image reading), the results.csv file containing the aggregated results will not be created. The user can then obtain the aggregated partial results with plots from the partially completed benchmark by running the following after cd into the DomainLab directory:

    @@ -525,7 +517,7 @@

    Obtain partial resultsclean up the extra csv head generated and plot the csv using command below

    -

    Generate plots from .csv file

    +

    Generate plots from .csv file

    If the benchmark is not completed, the graphics subdirectory might not be created. The user can then manually create the graphics from the csv file of the aggregated partial results, which can be obtained as explained above. Here for, the user must cd into the DomainLab directory and run

    @@ -610,7 +602,7 @@

    Generate plots from .csv fileSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_benchmark_further_explanation.html b/docs/build/html/doc_benchmark_further_explanation.html index fb93ca08b..e9ddc2d0b 100644 --- a/docs/build/html/doc_benchmark_further_explanation.html +++ b/docs/build/html/doc_benchmark_further_explanation.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -310,13 +309,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -350,7 +342,7 @@
    -

    Further explanations to Benchmark Setup

    +

    Further explanations to Benchmark Setup

    The user can create different custom experiments, which are to be benchmarked. Each experiment can have a custom name.

    An experiment can specify following arguments:

    @@ -418,7 +410,7 @@

    Further explanations to Be
    -

    Hyperparameter sampling

    +

    Hyperparameter sampling

    The benchmark offers the option to randomly sample hyperparameters from different distributions. An example can be found in demo_hyperparameter_sampling.yml. We offer two sampling techniques, random hyperparameter sampling and grid search. The default sampling mode is random @@ -450,7 +442,7 @@

    Hyperparameter sampling

    -

    Constraints

    +

    Constraints

    The user can specify a list of constraints for the hyperparameters. Please note the following:

    @@ -339,7 +331,7 @@
    -

    Let DomainLab know where your PACS data were located

    +

    Let DomainLab know where your PACS data were located

    We define the diretory containing PACS data here relative directory of PACS dataset with respect to the DomainLab repository To avoid changing the code, which is always discouraged, one could make a symbolic link of your PACS dataset to the required directory in the code above see examples here

    @@ -367,7 +359,7 @@

    Let DomainLab know where your PACS data w Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_benchmark_yaml.html b/docs/build/html/doc_benchmark_yaml.html index 34793deaa..69fd995cd 100644 --- a/docs/build/html/doc_benchmark_yaml.html +++ b/docs/build/html/doc_benchmark_yaml.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -362,12 +354,12 @@
    -

    Benchmark yaml files

    +

    Benchmark yaml files

    yaml files are a powerful tool to specify the details of a benchmark. The following examples on how a yaml file could look like will lead you through constructing your own yaml file for a benchmark.

    We will start with the general setting of the yaml file and then continue with the description of how to define the sampling/gridsearch for the hyperparameters.

    -

    General setup of the yaml file

    +

    General setup of the yaml file

    One may start with a very general setup of the file, defining all fixed information like task description, test and training domains, used algorithms …

    At the top level, we need to decide whether the random sampling or grid search shall be used.

    For random samples we need to define the total number of hyperparameter samples in each sampling task (num_param_samples) and a sampling seed for the hyperparameters (sampling_seed).

    @@ -509,11 +501,11 @@

    General setup of the yaml file -

    Sampling description

    +

    Sampling description

    There are two possible ways of choosing your hyperparameters for the benchmark in domainlab, rand sampling and grid search. The decision about which method to use was already done in the previous section be either setting num_param_samples (for random sampling) or mode: grid (for gridsearch).

    For filling in the sampling description for the into the Shared params and the hyperparameter section you have the following options:

    -

    uniform and loguniform distribution

    +

    uniform and loguniform distribution

    1. uniform samples in the interval [min, max]

    @@ -538,7 +530,7 @@

    uniform and loguniform distribution

    -

    normal and lognormal distribution

    +

    normal and lognormal distribution

    1. normal samples with mean and standard deviation

    @@ -563,7 +555,7 @@

    normal and lognormal distribution

    -

    cathegorical hyperparameters

    +

    cathegorical hyperparameters

    choose the values of the hyperparameter from a predefined list. If one uses grid search, then all values from the list are used as grid points

    nperm:                          # name of the hyperparameter
         distribution: categorical   # name of the distribution
    @@ -576,7 +568,7 @@ 

    cathegorical hyperparameters -

    Referenced hperparameters

    +

    Referenced hperparameters

    If one hyperparameter does directly depend on another hyperparameter you can add a reference to the other parameter. For gridsearch num will be taken from the reference, therefore it cannot be specified here.

    zy_dim:                         # name of the hyperparameter
         reference: 2 * zx_dim       # formular to be evaluated in python
    @@ -584,7 +576,7 @@ 

    Referenced hperparameters -

    Special Arguments in the Sampling description

    +

    Special Arguments in the Sampling description

    1. datatype: specify the datatype of the samples, if int, the values are rounded to the next integer. This works will all distributions mentioned above.

        @@ -617,7 +609,7 @@

        Special Arguments in the

    -

    Combination of Shared and Task Specific Hyperparameter Samples

    +

    Combination of Shared and Task Specific Hyperparameter Samples

    it is possible to have all sorts of combinations:

    1. a task which includes shared and task specific sampled hyperparameters

    2. @@ -701,7 +693,7 @@

      Combinat Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_coral.html b/docs/build/html/doc_coral.html index 19f3e33ab..858c6264a 100644 --- a/docs/build/html/doc_coral.html +++ b/docs/build/html/doc_coral.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -342,13 +341,6 @@ Causal IRL - -
    3. - - - Invariant Causal Mechanisms through Distribution Matching - -
    4. @@ -382,15 +374,15 @@
      -

      Deep CORAL

      +

      Deep CORAL

      -

      Deep CORAL: Correlation Alignment for Deep Domain Adaptation

      +

      Deep CORAL: Correlation Alignment for Deep Domain Adaptation

      nonlinear transformation that aligns correlations of layer activations in deep neural network https://arxiv.org/pdf/1607.01719

      -

      Examples

      +

      Examples

      python main_out.py --te_d 0 --tr_d 3 7 --bs=32 --epos=1 --task=mnistcolor10 --model=erm --nname=conv_bn_pool_2 --trainer=coral
       
      @@ -444,7 +436,7 @@

      ExamplesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_custom_nn.html b/docs/build/html/doc_custom_nn.html index faeb763dd..b7536b45a 100644 --- a/docs/build/html/doc_custom_nn.html +++ b/docs/build/html/doc_custom_nn.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
    5. - - - Invariant Causal Mechanisms through Distribution Matching - -
    6. @@ -362,7 +354,7 @@
      -

      Specify neural network in command line

      +

      Specify neural network in command line

      To use a custom neural network in command line with DomainLab, the user has to implement the following signature in a python file and specify the file path via --npath

      def build_feat_extract_net(dim_y, remove_last_layer=False):
       
      @@ -371,36 +363,36 @@

      Specify neural network in command lineSee examples below from --npath=examples/nets/resnet.py where the examples can be found in the examples folder of the code repository. https://github.com/marrlab/DomainLab/blob/master/examples/nets/resnet.py

      -

      Example use case

      +

      Example use case

      -

      model ‘erm’ with custom neural network

      +

      model ‘erm’ with custom neural network

      python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=2 --model=erm --npath=examples/nets/resnet.py
       
      -

      trainer ‘matchdg’ with custom neural network

      +

      trainer ‘matchdg’ with custom neural network

      python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=2 --model=erm --trainer=matchdg --epochs_ctr=3 --epos=6 --npath=examples/nets/resnet.py
       
      -

      model erm with custom neural network

      +

      model erm with custom neural network

      python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=8 --model=erm --npath=examples/nets/resnet.py
       
      -

      Larger images:

      +

      Larger images:

      -

      model erm with implemented neural network

      +

      model erm with implemented neural network

      python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=8 --model=erm --nname=alexnet
       
      -

      model dann with implemented neural network

      +

      model dann with implemented neural network

      python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=8 --model=dann --nname=alexnet
       
      @@ -431,7 +423,7 @@

      model dann with implemented

      Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_dann.html b/docs/build/html/doc_dann.html index ade1c9321..1d3569ac7 100644 --- a/docs/build/html/doc_dann.html +++ b/docs/build/html/doc_dann.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -333,13 +332,6 @@ Causal IRL - -
    7. - - - Invariant Causal Mechanisms through Distribution Matching - -
    8. @@ -371,9 +363,9 @@
      -

      Model DANN

      +

      Model DANN

      -

      Domain Adversarial Neural Network

      +

      Domain Adversarial Neural Network

      Details explained in the following publication: Ganin, Yaroslav, et al. “Domain-adversarial training of neural networks.” The journal of machine learning research 17.1 (2016): 2096-2030.

      @@ -431,7 +423,7 @@

      Domain Adversarial Neural Network Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_diva.html b/docs/build/html/doc_diva.html index 141880af2..dcc143df3 100644 --- a/docs/build/html/doc_diva.html +++ b/docs/build/html/doc_diva.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -371,13 +370,6 @@ Causal IRL - -
    9. - - - Invariant Causal Mechanisms through Distribution Matching - -
    10. @@ -431,9 +423,9 @@
      -

      Model DIVA

      +

      Model DIVA

      -

      Domain Invariant Variational Autoencoders

      +

      Domain Invariant Variational Autoencoders

      DIVA addresses the domain generalization problem with a variational autoencoder with three latent variables, using three independent encoders.

      By encouraging the network to store each the domain, @@ -456,7 +448,7 @@

      Domain Invariant Variational Since it is not always clear which different domains actually exist, this can lead to problems and a decreased performance.

      -

      Model parameters

      +

      Model parameters

      The following hyperparameters can be specified:

      • zd_dim: size of latent space for domain-specific information

      • @@ -473,36 +465,36 @@

        Model parameters -

        Examples

        +

        Examples

        -

        model diva with implemented neural network

        +

        model diva with implemented neural network

        python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=2 --model=diva --nname=alexnet --npath_dom=examples/nets/resnet.py --gamma_y=7e5 --gamma_d=1e5
         
        -

        model diva with custom neural network

        +

        model diva with custom neural network

        python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=2 --model=diva --npath=examples/nets/resnet.py --npath_dom=examples/nets/resnet.py --gamma_y=7e5 --gamma_d=1e5
         
        -

        generation of images

        +

        generation of images

        python main_out.py --te_d=0 --task=mnistcolor10 --keep_model --model=diva --nname=conv_bn_pool_2 --nname_dom=conv_bn_pool_2 --gamma_y=10e5 --gamma_d=1e5 --gen
         
      -

      Colored version of MNIST

      +

      Colored version of MNIST

      -

      leave one domain out

      +

      leave one domain out

      python main_out.py --te_d=0 --task=mnistcolor10 --keep_model --model=diva --nname=conv_bn_pool_2 --nname_dom=conv_bn_pool_2 --gamma_y=10e5 --gamma_d=1e5
       
      -

      choose train and test

      +

      choose train and test

      python main_out.py --te_d 0 1 2 --tr_d 3 7 --task=mnistcolor10 --model=diva --nname=conv_bn_pool_2 --nname_dom=conv_bn_pool_2 --gamma_y=7e5 --gamma_d=1e5
       
      @@ -559,7 +551,7 @@

      choose train and testSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_extend_contribute.html b/docs/build/html/doc_extend_contribute.html index 2d0b292aa..c1788dee0 100644 --- a/docs/build/html/doc_extend_contribute.html +++ b/docs/build/html/doc_extend_contribute.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
    11. - - - Invariant Causal Mechanisms through Distribution Matching - -
    12. @@ -344,11 +336,11 @@
      -

      Software Architecture and Design

      +

      Software Architecture and Design

      Design Diagram

      -

      Code structure

      +

      Code structure

      domainlab/
       ├── algos
       │   ├── a_algo_builder.py
      @@ -484,7 +476,7 @@ 

      Code structureSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_install.html b/docs/build/html/doc_install.html index debd4a4eb..dd3999c25 100644 --- a/docs/build/html/doc_install.html +++ b/docs/build/html/doc_install.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
    13. - - - Invariant Causal Mechanisms through Distribution Matching - -
    14. @@ -356,14 +348,14 @@
      -

      Installation of DomainLab

      +

      Installation of DomainLab

      -

      make a sanity check for the dataset using 8 instances from each domain and from each class

      +

      make a sanity check for the dataset using 8 instances from each domain and from each class

      python main_out.py --te_d=0 --task=mini_vlcs --debug --bs=2 --model=diva --nname=alexnet --npath_dom=examples/nets/resnet.py --gamma_y=7e5 --gamma_d=1e5 --san_check --san_num=4
       
      -

      sanity check on only 2 train domains and 2 test domain2

      +

      sanity check on only 2 train domains and 2 test domain2

      python main_out.py --te_d 0 1 --tr_d 3 5 --task=mnistcolor10 --debug --bs=2 --model=erm --nname=conv_bn_pool_2 --san_check --san_num=4
       
      @@ -610,7 +602,7 @@

      sanity check on

      Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_trainer.html b/docs/build/html/doc_trainer.html index e30b08f60..e8747c133 100644 --- a/docs/build/html/doc_trainer.html +++ b/docs/build/html/doc_trainer.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
    15. - - - Invariant Causal Mechanisms through Distribution Matching - -
    16. @@ -348,10 +340,10 @@
      -

      Model Specification

      +

      Model Specification

      When developing new trainers, you can extend the TrainerBasic class. This allows you to build upon established training routines while introducing specialized behaviors tailored to your model.

      -

      Steps to Extend TrainerBasic

      +

      Steps to Extend TrainerBasic

      1. Extend the class: Begin by inheriting from TrainerBasic.

      2. Customize Key Methods: You have several methods that you can override to customize the trainer’s behavior. Here’s a brief overview of what they do:

        @@ -368,7 +360,7 @@

        Steps to Extend -

        Example Implementation

        +

        Example Implementation

        Here is a simple example of a custom trainer that logs additional details at the start of each training:

        class MyCustomTrainer(TrainerBasic):
             def before_tr(self):
        @@ -413,7 +405,7 @@ 

        Example Implementation Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/doc_usage_cmd.html b/docs/build/html/doc_usage_cmd.html index ebe9d0f4c..d1706dfbe 100644 --- a/docs/build/html/doc_usage_cmd.html +++ b/docs/build/html/doc_usage_cmd.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL -

      3. -
      4. - - - Invariant Causal Mechanisms through Distribution Matching - -
      5. @@ -374,10 +366,10 @@
        -

        DomainLab Usage Guide

        +

        DomainLab Usage Guide

        Given the repository and the dependencies are set up, here is how can use DomainLab:

        -

        Essential Commands

        +

        Essential Commands

        To run DomainLab, the minimum necessary parameters are:

        1. Task Specification (--tpath or --task): This is to specify a task. You can eiter give a path to as Python file which specifies the task, or use a predfined set. You can find more about specifying tasks here.

        2. @@ -386,14 +378,14 @@

          Essential Commands--nname or --npath): Specifies which neural network is used for feature extraction, either through a path or predefined options.

        -

        Example Command

        +

        Example Command

        python3 main_out.py --te_d 0 1 2  --task=mnistcolor10 --model=diva --nname=conv_bn_pool_2  --gamma_y=7e5 --gamma_d=1e5

        -

        Optional Commands

        +

        Optional Commands

        -

        Advanced Configuration

        +

        Advanced Configuration

        • Learning Rate (--lr): Set the training learning rate.

        • Regularization (--gamma_reg): Sets the weight of the regularization @@ -462,7 +454,7 @@

          Advanced Configuration

        -

        Task-Specific Arguments

        +

        Task-Specific Arguments

        • Batch Size (--bs): Loader batch size for mixed domains.

        • Training-Validation Split (--split): Proportion of training, a value between 0 and 1.

        • @@ -474,9 +466,9 @@

          Task-Specific Arguments

        -

        Model-Specific Hyperparameters

        +

        Model-Specific Hyperparameters

        -

        VAE Model Parameters

        +

        VAE Model Parameters

        • Latent Space Dimensions (--zd_dim, --zx_dim, --zy_dim): Set the size of latent spaces for domain, unobserved, and class features.

        • Topic Dimension (--topic_dim): Number of topics for HDUVA.

        • @@ -490,7 +482,7 @@

          VAE Model Parameters -

          MatchDG Parameters

          +

          MatchDG Parameters

          • Cosine Similarity Factor (--tau): Magnify cosine similarity.

          • Match Tensor Update Frequency (--epos_per_match_update): Epochs before updating the match tensor.

          • @@ -498,14 +490,14 @@

            MatchDG Parameters -

            Jigen Parameters

            +

            Jigen Parameters

            • Permutation Settings (--nperm, --pperm, --jigen_ppath): Configure image tile permutations.

            • Grid Length (--grid_len): Length of image in tile units.

        -

        DIAL Parameters

        +

        DIAL Parameters

        • Adversarial Image Generation (--dial_steps_perturb, --dial_noise_scale, --dial_lr, --dial_epsilon): Configure parameters for generating adversarial images.

        @@ -517,7 +509,7 @@

        DIAL Parameters -

        Example

        +

        Example

        DomainLab comes with some minimal toy-dataset to test its basis functionality, see a minimal subsample of the VLCS dataset and an example configuration file for vlcs_mini.

        To train a domain invariant model on the vlcs_mini task:

        python main_out.py --te_d=caltech --tpath=examples/tasks/task_vlcs.py --config=examples/yaml/demo_config_single_run_diva.yaml
        @@ -533,7 +525,7 @@ 

        Examplemore examples.

        -

        Further Resources

        +

        Further Resources

        @@ -381,9 +373,9 @@
        -

        domainlab.algos package

        +

        domainlab.algos package

        -

        Subpackages

        +

        Subpackages

        -

        Submodules

        +

        Submodules

        -

        domainlab.algos.a_algo_builder module

        +

        domainlab.algos.a_algo_builder module

        parent class for combing model, trainer, task, observer

        class domainlab.algos.a_algo_builder.NodeAlgoBuilder(success_node=None)[source]
        -

        Bases: AbstractChainNodeHandler

        +

        Bases: domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler

        Base class for Algorithm Builder

        @@ -455,7 +447,7 @@

        Submodules -
        Parameters:
        +
        Parameters

        node – The node to be added to the algorithm builder.

        @@ -474,7 +466,7 @@

        Submodules is_myjob(request)[source]
        -
        Parameters:
        +
        Parameters

        request – string

        @@ -491,12 +483,12 @@

        Submodules -

        domainlab.algos.builder_api_model module

        +

        domainlab.algos.builder_api_model module

        build algorithm from API coded model with custom backbone

        class domainlab.algos.builder_api_model.NodeAlgoBuilderAPIModel(success_node=None)[source]
        -

        Bases: NodeAlgoBuilder

        +

        Bases: domainlab.algos.a_algo_builder.NodeAlgoBuilder

        build algorithm from API coded model with custom backbone

        @@ -506,7 +498,7 @@

        Submodules -

        domainlab.algos.builder_custom module

        +

        domainlab.algos.builder_custom module

        domainlab.algos.builder_custom.make_basic_trainer(class_name_model)[source]
        @@ -515,12 +507,12 @@

        Submodules -

        domainlab.algos.builder_dann module

        +

        domainlab.algos.builder_dann module

        builder for Domain Adversarial Neural Network: accept different training scheme

        class domainlab.algos.builder_dann.NodeAlgoBuilderDANN(success_node=None)[source]
        -

        Bases: NodeAlgoBuilder

        +

        Bases: domainlab.algos.a_algo_builder.NodeAlgoBuilder

        init_business(exp)[source]
        @@ -535,12 +527,12 @@

        Submodules -

        domainlab.algos.builder_diva module

        +

        domainlab.algos.builder_diva module

        Builder pattern to build different component for experiment with DIVA

        class domainlab.algos.builder_diva.NodeAlgoBuilderDIVA(success_node=None)[source]
        -

        Bases: NodeAlgoBuilder

        +

        Bases: domainlab.algos.a_algo_builder.NodeAlgoBuilder

        Builder pattern to build different component for experiment with DIVA

        @@ -555,12 +547,12 @@

        Submodules -

        domainlab.algos.builder_erm module

        +

        domainlab.algos.builder_erm module

        builder for erm

        class domainlab.algos.builder_erm.NodeAlgoBuilderERM(success_node=None)[source]
        -

        Bases: NodeAlgoBuilder

        +

        Bases: domainlab.algos.a_algo_builder.NodeAlgoBuilder

        builder for erm

        @@ -570,12 +562,12 @@

        Submodules -

        domainlab.algos.builder_hduva module

        +

        domainlab.algos.builder_hduva module

        build hduva model, get trainer from cmd arguments

        class domainlab.algos.builder_hduva.NodeAlgoBuilderHDUVA(success_node=None)[source]
        -

        Bases: NodeAlgoBuilder

        +

        Bases: domainlab.algos.a_algo_builder.NodeAlgoBuilder

        init_business(exp)[source]
        @@ -584,12 +576,12 @@

        Submodules -

        domainlab.algos.builder_jigen1 module

        +

        domainlab.algos.builder_jigen1 module

        builder for JiGen

        class domainlab.algos.builder_jigen1.NodeAlgoBuilderJiGen(success_node=None)[source]
        -

        Bases: NodeAlgoBuilder

        +

        Bases: domainlab.algos.a_algo_builder.NodeAlgoBuilder

        init_business(exp)[source]
        @@ -598,7 +590,7 @@

        Submodules -

        domainlab.algos.utils module

        +

        domainlab.algos.utils module

        network builder utils

        @@ -607,7 +599,7 @@

        Submodules -

        domainlab.algos.zoo_algos module

        +

        domainlab.algos.zoo_algos module

        chain of responsibility pattern for algorithm selection

        @@ -623,7 +615,7 @@

        Submodules -

        Module contents

        +

        Module contents

        @@ -674,7 +666,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.algos.msels.html b/docs/build/html/domainlab.algos.msels.html index 59df915fc..2dfedb355 100644 --- a/docs/build/html/domainlab.algos.msels.html +++ b/docs/build/html/domainlab.algos.msels.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -320,13 +319,6 @@ Causal IRL - -
      6. - - - Invariant Causal Mechanisms through Distribution Matching - -
      7. @@ -368,12 +360,12 @@
        -

        domainlab.algos.msels package

        +

        domainlab.algos.msels package

        -

        Submodules

        +

        Submodules

        -

        domainlab.algos.msels.a_model_sel module

        +

        domainlab.algos.msels.a_model_sel module

        Abstract Model Selection

        @@ -453,12 +445,12 @@

        Submodules -

        domainlab.algos.msels.c_msel_oracle module

        +

        domainlab.algos.msels.c_msel_oracle module

        Model Selection should be decoupled from

        class domainlab.algos.msels.c_msel_oracle.MSelOracleVisitor(msel=None, val_threshold=None)[source]
        -

        Bases: AMSel

        +

        Bases: domainlab.algos.msels.a_model_sel.AMSel

        save best out-of-domain test acc model, but do not affect how the final model is selected

        @@ -487,12 +479,12 @@

        Submodules -

        domainlab.algos.msels.c_msel_tr_loss module

        +

        domainlab.algos.msels.c_msel_tr_loss module

        AMSel.accept —> Trainer

        class domainlab.algos.msels.c_msel_tr_loss.MSelTrLoss(max_es, val_threshold=None)[source]
        -

        Bases: AMSel

        +

        Bases: domainlab.algos.msels.a_model_sel.AMSel

        1. Model selection using sum of loss across training domains

        2. Visitor pattern to trainer

        3. @@ -520,12 +512,12 @@

          Submodules -

          domainlab.algos.msels.c_msel_val module

          +

          domainlab.algos.msels.c_msel_val module

          Model Selection should be decoupled from

          class domainlab.algos.msels.c_msel_val.MSelValPerf(max_es, val_threshold=None)[source]
          -

          Bases: MSelTrLoss

          +

          Bases: domainlab.algos.msels.c_msel_tr_loss.MSelTrLoss

          1. Model selection using validation performance

          2. Visitor pattern to trainer

          3. @@ -558,7 +550,7 @@

            Submodules -

            Module contents

            +

            Module contents

        @@ -609,7 +601,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.algos.observers.html b/docs/build/html/domainlab.algos.observers.html index 3a43becdd..eb0584d94 100644 --- a/docs/build/html/domainlab.algos.observers.html +++ b/docs/build/html/domainlab.algos.observers.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -320,13 +319,6 @@ Causal IRL - -
      8. - - - Invariant Causal Mechanisms through Distribution Matching - -
      9. @@ -368,12 +360,12 @@
        -

        domainlab.algos.observers package

        +

        domainlab.algos.observers package

        -

        Submodules

        +

        Submodules

        -

        domainlab.algos.observers.a_observer module

        +

        domainlab.algos.observers.a_observer module

        interface for observer + visitor

        @@ -413,14 +405,14 @@

        Submodules -

        domainlab.algos.observers.b_obvisitor module

        +

        domainlab.algos.observers.b_obvisitor module

        observer and visitor pattern, responsible train, validation, test dispatch performance evaluation to model, dispatch model selection to model selection object

        class domainlab.algos.observers.b_obvisitor.ObVisitor(model_sel)[source]
        -

        Bases: AObVisitor

        +

        Bases: domainlab.algos.observers.a_observer.AObVisitor

        Observer + Visitor pattern for model selection

        @@ -460,11 +452,11 @@

        Submodules -

        domainlab.algos.observers.c_obvisitor_cleanup module

        +

        domainlab.algos.observers.c_obvisitor_cleanup module

        class domainlab.algos.observers.c_obvisitor_cleanup.ObVisitorCleanUp(observer)[source]
        -

        Bases: AObVisitor

        +

        Bases: domainlab.algos.observers.a_observer.AObVisitor

        decorator of observer, instead of using if and else to decide clean up or not, we use decorator

        @@ -493,11 +485,11 @@

        Submodules -

        domainlab.algos.observers.c_obvisitor_gen module

        +

        domainlab.algos.observers.c_obvisitor_gen module

        class domainlab.algos.observers.c_obvisitor_gen.ObVisitorGen(model_sel)[source]
        -

        Bases: ObVisitor

        +

        Bases: domainlab.algos.observers.b_obvisitor.ObVisitor

        For Generative Models

        @@ -507,7 +499,7 @@

        Submodules -

        Module contents

        +

        Module contents

        @@ -558,7 +550,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.algos.trainers.compos.html b/docs/build/html/domainlab.algos.trainers.compos.html index 70985c43a..afac40ef1 100644 --- a/docs/build/html/domainlab.algos.trainers.compos.html +++ b/docs/build/html/domainlab.algos.trainers.compos.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -321,13 +320,6 @@ Causal IRL - -
      10. - - - Invariant Causal Mechanisms through Distribution Matching - -
      11. @@ -367,12 +359,12 @@
        -

        domainlab.algos.trainers.compos package

        +

        domainlab.algos.trainers.compos package

        -

        Submodules

        +

        Submodules

        -

        domainlab.algos.trainers.compos.matchdg_args module

        +

        domainlab.algos.trainers.compos.matchdg_args module

        args for matchdg

        @@ -381,7 +373,7 @@

        Submodules -

        domainlab.algos.trainers.compos.matchdg_match module

        +

        domainlab.algos.trainers.compos.matchdg_match module

        class domainlab.algos.trainers.compos.matchdg_match.MatchPair(dim_y, i_c, i_h, i_w, bs_match, virtual_ref_dset_size, num_domains_tr, list_tr_domain_size)[source]
        @@ -390,7 +382,7 @@

        Submodules -

        domainlab.algos.trainers.compos.matchdg_utils module

        +

        domainlab.algos.trainers.compos.matchdg_utils module

        create dictionary for matching

        @@ -405,7 +397,7 @@

        Submodules
        class domainlab.algos.trainers.compos.matchdg_utils.MatchDictNumDomain2SizeDomain(num_domains_tr, list_tr_domain_size, i_c, i_h, i_w)[source]
        -

        Bases: MatchDictInit

        +

        Bases: domainlab.algos.trainers.compos.matchdg_utils.MatchDictInit

        tensor dimension for the kth domain: [num_domains_tr, (size_domain_k, i_c, i_h, i_w)]

        @@ -415,7 +407,7 @@

        Submodules
        class domainlab.algos.trainers.compos.matchdg_utils.MatchDictVirtualRefDset2EachDomain(virtual_ref_dset_size, num_domains_tr, i_c, i_h, i_w)[source]
        -

        Bases: MatchDictInit

        +

        Bases: domainlab.algos.trainers.compos.matchdg_utils.MatchDictInit

        dict[0:virtual_ref_dset_size] has tensor dimension: (num_domains_tr, i_c, i_h, i_w)

        @@ -445,7 +437,7 @@

        Submodules -

        Module contents

        +

        Module contents

        @@ -496,7 +488,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.algos.trainers.html b/docs/build/html/domainlab.algos.trainers.html index 6abf208b6..b7998cd10 100644 --- a/docs/build/html/domainlab.algos.trainers.html +++ b/docs/build/html/domainlab.algos.trainers.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -320,13 +319,6 @@ Causal IRL - -
      12. - - - Invariant Causal Mechanisms through Distribution Matching - -
      13. @@ -392,9 +384,9 @@
        -

        domainlab.algos.trainers package

        +

        domainlab.algos.trainers package

        -

        Subpackages

        +

        Subpackages

        -

        Submodules

        +

        Submodules

        -

        domainlab.algos.trainers.a_trainer module

        +

        domainlab.algos.trainers.a_trainer module

        Base Class for trainer

        class domainlab.algos.trainers.a_trainer.AbstractTrainer(successor_node=None, extend=None)[source]
        -

        Bases: AbstractChainNodeHandler

        +

        Bases: domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler

        Algorithm director that controls the data flow

        after_batch(epoch, ind_batch)[source]
        -
        Parameters:
        +
        Parameters
        @@ -1073,7 +1065,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.compos.html b/docs/build/html/domainlab.compos.html index a064d6d98..1921d345c 100644 --- a/docs/build/html/domainlab.compos.html +++ b/docs/build/html/domainlab.compos.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -319,13 +318,6 @@ Causal IRL - -
      14. - - - Invariant Causal Mechanisms through Distribution Matching - -
      15. @@ -373,9 +365,9 @@
        -

        domainlab.compos package

        +

        domainlab.compos package

        -

        Subpackages

        +

        Subpackages

        -

        Submodules

        +

        Submodules

        -

        domainlab.compos.a_nn_builder module

        +

        domainlab.compos.a_nn_builder module

        Integrate Chain-of-Responsibility and Builder Pattern for feature extract

        class domainlab.compos.a_nn_builder.AbstractFeatExtractNNBuilderChainNode(successor_node)[source]
        -

        Bases: AbstractChainNodeHandler

        +

        Bases: domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler

        to ensure chain of responsibility node AbstractChainNodeHandler always work even some node can not start their heavy weight business object, avoid override the initializer so that node construction is always @@ -460,7 +452,7 @@

        Submodules is_myjob(args)[source]
        -
        Parameters:
        +
        Parameters

        args_nname – command line arguments: “–nname”: name of the torchvision model “–npath”: path to the user specified python file with neural @@ -471,14 +463,14 @@

        Submodules -

        domainlab.compos.builder_nn_alex module

        +

        domainlab.compos.builder_nn_alex module

        domainlab.compos.builder_nn_alex.mkNodeFeatExtractNNBuilderNameAlex(arg_name4net, arg_val)[source]

        -

        domainlab.compos.builder_nn_conv_bn_relu_2 module

        +

        domainlab.compos.builder_nn_conv_bn_relu_2 module

        domainlab.compos.builder_nn_conv_bn_relu_2.mkNodeFeatExtractNNBuilderNameConvBnRelu2(arg_name4net, arg_val, conv_stride)[source]
        @@ -494,7 +486,7 @@

        Submodules -

        domainlab.compos.builder_nn_external_from_file module

        +

        domainlab.compos.builder_nn_external_from_file module

        domainlab.compos.builder_nn_external_from_file.mkNodeFeatExtractNNBuilderExternFromFile(arg_name_net_path)[source]
        @@ -504,7 +496,7 @@

        Submodules -

        domainlab.compos.utils_conv_get_flat_dim module

        +

        domainlab.compos.utils_conv_get_flat_dim module

        domainlab.compos.utils_conv_get_flat_dim.get_flat_dim(module, i_channel, i_h, i_w, batchsize=5)[source]
        @@ -518,7 +510,7 @@

        Submodules -

        domainlab.compos.zoo_nn module

        +

        domainlab.compos.zoo_nn module

        class domainlab.compos.zoo_nn.FeatExtractNNBuilderChainNodeGetter(args, arg_name_of_net, arg_path_of_net)[source]
        @@ -528,7 +520,7 @@

        Submodules -

        Module contents

        +

        Module contents

        @@ -579,7 +571,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.compos.nn_zoo.html b/docs/build/html/domainlab.compos.nn_zoo.html index 3571ecbe6..49f7b3746 100644 --- a/docs/build/html/domainlab.compos.nn_zoo.html +++ b/docs/build/html/domainlab.compos.nn_zoo.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -320,13 +319,6 @@ Causal IRL - -
      16. - - - Invariant Causal Mechanisms through Distribution Matching - -
      17. @@ -374,16 +366,16 @@
        -

        domainlab.compos.nn_zoo package

        +

        domainlab.compos.nn_zoo package

        -

        Submodules

        +

        Submodules

        -

        domainlab.compos.nn_zoo.net_adversarial module

        +

        domainlab.compos.nn_zoo.net_adversarial module

        class domainlab.compos.nn_zoo.net_adversarial.AutoGradFunMultiply(*args, **kwargs)[source]
        -

        Bases: Function

        +

        Bases: torch.autograd.function.Function

        static backward(ctx, grad_output)[source]
        @@ -424,7 +416,7 @@

        Submodules
        class domainlab.compos.nn_zoo.net_adversarial.AutoGradFunReverseMultiply(*args, **kwargs)[source]
        -

        Bases: Function

        +

        Bases: torch.autograd.function.Function

        https://pytorch.org/docs/stable/autograd.html https://pytorch.org/docs/stable/notes/extending.html#extending-autograd

        @@ -467,7 +459,7 @@

        Submodules
        class domainlab.compos.nn_zoo.net_adversarial.Flatten[source]
        -

        Bases: Module

        +

        Bases: torch.nn.modules.module.Module

        forward(x)[source]
        @@ -488,18 +480,18 @@

        Submodules -

        domainlab.compos.nn_zoo.net_classif module

        +

        domainlab.compos.nn_zoo.net_classif module

        Classifier

        class domainlab.compos.nn_zoo.net_classif.ClassifDropoutReluLinear(z_dim, target_dim)[source]
        -

        Bases: Module

        +

        Bases: torch.nn.modules.module.Module

        first apply dropout, then relu, then linearly fully connected, without activation

        forward(z_vec)[source]
        -
        Parameters:
        +
        Parameters

        z_vec

        @@ -511,17 +503,17 @@

        Submodules -

        domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2 module

        +

        domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2 module

        In PyTorch, images are represented as [channels, height, width]

        class domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2.NetConvBnReluPool2L(isize, conv_stride, dim_out_h)[source]
        -

        Bases: Module

        +

        Bases: torch.nn.modules.module.Module

        forward(tensor_x)[source]
        -
        Parameters:
        +
        Parameters

        tensor_x – image

        @@ -534,7 +526,7 @@

        Submodules
        class domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2.NetConvDense(isize, conv_stride, dim_out_h, args, dense_layer=None)[source]
        -

        Bases: Module

        +

        Bases: torch.nn.modules.module.Module

        @@ -821,7 +813,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.compos.pcr.html b/docs/build/html/domainlab.compos.pcr.html index 924d9d4b6..b136f1698 100644 --- a/docs/build/html/domainlab.compos.pcr.html +++ b/docs/build/html/domainlab.compos.pcr.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -320,13 +319,6 @@ Causal IRL - -
      18. - - - Invariant Causal Mechanisms through Distribution Matching - -
      19. @@ -364,12 +356,12 @@
        -

        domainlab.compos.pcr package

        +

        domainlab.compos.pcr package

        -

        Submodules

        +

        Submodules

        -

        domainlab.compos.pcr.p_chain_handler module

        +

        domainlab.compos.pcr.p_chain_handler module

        Chain of Responsibility

        @@ -440,7 +432,7 @@

        Submodules
        class domainlab.compos.pcr.p_chain_handler.DummyChainNodeHandlerBeaver(success_node=None)[source]
        -

        Bases: AbstractChainNodeHandler

        +

        Bases: domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler

        Dummy class to show how to inherit from Chain of Responsibility

        @@ -461,7 +453,7 @@

        Submodules
        class domainlab.compos.pcr.p_chain_handler.DummyChainNodeHandlerLazy(success_node=None)[source]
        -

        Bases: AbstractChainNodeHandler

        +

        Bases: domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler

        Dummy class to show how to inherit from Chain of Responsibility

        @@ -492,7 +484,7 @@

        Submodules -

        domainlab.compos.pcr.request module

        +

        domainlab.compos.pcr.request module

        class domainlab.compos.pcr.request.RequestArgs2ExpCmd(args)[source]
        @@ -520,7 +512,7 @@

        Submodules -

        Module contents

        +

        Module contents

        @@ -571,7 +563,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.compos.vae.compos.html b/docs/build/html/domainlab.compos.vae.compos.html index 7c3da267e..3e2295fad 100644 --- a/docs/build/html/domainlab.compos.vae.compos.html +++ b/docs/build/html/domainlab.compos.vae.compos.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -321,13 +320,6 @@ Causal IRL - -
      20. - - - Invariant Causal Mechanisms through Distribution Matching - -
      21. @@ -385,17 +377,17 @@
        -

        domainlab.compos.vae.compos package

        +

        domainlab.compos.vae.compos package

        -

        Submodules

        +

        Submodules

        -

        domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv module

        +

        domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv module

        decoder which takes concatenated latent representation

        class domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv.DecoderConcatLatentFcReshapeConv(z_dim, i_c, i_h, i_w, cls_fun_nll_p_x, net_fc_z2flat_img, net_conv, net_p_x_mean, net_p_x_log_var)[source]
        -

        Bases: Module

        +

        Bases: torch.nn.modules.module.Module

        Latent vector re-arranged to image-size directly, then convolute to get the textures of the original image

        @@ -416,7 +408,7 @@

        Submodules forward(z, img)[source]
        -
        Parameters:
        +
        Parameters

        @@ -792,7 +784,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.compos.vae.html b/docs/build/html/domainlab.compos.vae.html index 3f10e6c13..84256035d 100644 --- a/docs/build/html/domainlab.compos.vae.html +++ b/docs/build/html/domainlab.compos.vae.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -320,13 +319,6 @@ Causal IRL - -
      22. - - - Invariant Causal Mechanisms through Distribution Matching - -
      23. @@ -378,9 +370,9 @@
        -

        domainlab.compos.vae package

        +

        domainlab.compos.vae package

        -

        Subpackages

        +

        Subpackages

        -

        Submodules

        +

        Submodules

        -

        domainlab.compos.vae.a_model_builder module

        +

        domainlab.compos.vae.a_model_builder module

        Integrate Chain-of-Responsibility and Builder Patter to construct VAE encoder and decoder

        class domainlab.compos.vae.a_model_builder.AbstractModelBuilderChainNode(success_node=None)[source]
        -

        Bases: AbstractChainNodeHandler

        +

        Bases: domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler

        to ensure chain of responsibility node AbstractChainNodeHandler always work even some node can not start their heavy weight business object, avoid override the initializer so that node construction is always light weight.

        @@ -425,13 +417,13 @@

        Submodules -

        domainlab.compos.vae.a_vae_builder module

        +

        domainlab.compos.vae.a_vae_builder module

        Integrate Chain-of-Responsibility and Builder Patter to construct VAE encoder and decoder

        class domainlab.compos.vae.a_vae_builder.AbstractVAEBuilderChainNode(successor_node)[source]
        -

        Bases: AbstractChainNodeHandler

        +

        Bases: domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler

        to ensure chain of responsibility node AbstractChainNodeHandler always work even some node can not start their heavy weight business object, avoid override the @@ -454,7 +446,7 @@

        Submodules -

        domainlab.compos.vae.c_vae_adaptor_model_recon module

        +

        domainlab.compos.vae.c_vae_adaptor_model_recon module

        This adaptor couples intensively with the heavy-weight model class The model class can be refactored, we do want to use the trained old-version model, which we only need to change this adaptor class.

        @@ -498,14 +490,14 @@

        Submodules -

        domainlab.compos.vae.c_vae_builder_classif module

        +

        domainlab.compos.vae.c_vae_builder_classif module

        Builder 1. classifier for domain and class 2. p(z_y|y) and p(z_d|d)

        class domainlab.compos.vae.c_vae_builder_classif.ChainNodeVAEBuilderClassifCondPrior(successor_node)[source]
        -

        Bases: AbstractVAEBuilderChainNode

        +

        Bases: domainlab.compos.vae.a_vae_builder.AbstractVAEBuilderChainNode

        1. This class defines common methods shared by child classes:
        @@ -809,7 +801,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.dsets.html b/docs/build/html/domainlab.dsets.html index d712f5243..635f9f31b 100644 --- a/docs/build/html/domainlab.dsets.html +++ b/docs/build/html/domainlab.dsets.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -319,13 +318,6 @@ Causal IRL - -
      24. - - - Invariant Causal Mechanisms through Distribution Matching - -
      25. @@ -375,17 +367,17 @@
        -

        domainlab.dsets package

        +

        domainlab.dsets package

        -

        Submodules

        +

        Submodules

        -

        domainlab.dsets.a_dset_mnist_color_rgb_solo module

        +

        domainlab.dsets.a_dset_mnist_color_rgb_solo module

        Color MNIST with single color

        class domainlab.dsets.a_dset_mnist_color_rgb_solo.ADsetMNISTColorRGBSolo(ind_color, path='zoutput', subset_step=100, color_scheme='both', label_transform=<function mk_fun_label2onehot.<locals>.fun_label2onehot>, list_transforms=None, raw_split='train', flag_rand_color=False)[source]
        -

        Bases: Dataset

        +

        Bases: torch.utils.data.dataset.Dataset

        Color MNIST with single color 1. nominal domains: color palettes/range/spectrum 2. subdomains: color(foreground, background) @@ -406,11 +398,11 @@

        Submodules -

        domainlab.dsets.dset_img_path_list module

        +

        domainlab.dsets.dset_img_path_list module

        class domainlab.dsets.dset_img_path_list.DsetImPathList(root_img, path2filelist, trans_img=None, trans_target=None)[source]
        -

        Bases: Dataset

        +

        Bases: torch.utils.data.dataset.Dataset

        get_list_tuple_img_label()[source]
        @@ -418,11 +410,11 @@

        Submodules -

        domainlab.dsets.dset_mnist_color_solo_default module

        +

        domainlab.dsets.dset_mnist_color_solo_default module

        class domainlab.dsets.dset_mnist_color_solo_default.DsetMNISTColorSoloDefault(ind_color, path='zoutput', subset_step=100, color_scheme='both', label_transform=<function mk_fun_label2onehot.<locals>.fun_label2onehot>, list_transforms=None, raw_split='train', flag_rand_color=False)[source]
        -

        Bases: ADsetMNISTColorRGBSolo

        +

        Bases: domainlab.dsets.a_dset_mnist_color_rgb_solo.ADsetMNISTColorRGBSolo

        get_background_color(ind)[source]
        @@ -442,28 +434,28 @@

        Submodules -

        domainlab.dsets.dset_poly_domains_mnist_color_default module

        +

        domainlab.dsets.dset_poly_domains_mnist_color_default module

        merge several solo-color mnist to form a mixed dataset

        class domainlab.dsets.dset_poly_domains_mnist_color_default.DsetMNISTColorMix(n_domains, path, color_scheme='both')[source]
        -

        Bases: Dataset

        +

        Bases: torch.utils.data.dataset.Dataset

        merge several solo-color mnist to form a mixed dataset

        class domainlab.dsets.dset_poly_domains_mnist_color_default.DsetMNISTColorMixNoDomainLabel(n_domains, path, color_scheme='both')[source]
        -

        Bases: DsetMNISTColorMix

        +

        Bases: domainlab.dsets.dset_poly_domains_mnist_color_default.DsetMNISTColorMix

        -

        domainlab.dsets.dset_subfolder module

        +

        domainlab.dsets.dset_subfolder module

        https://github.com/pytorch/vision/blob/bb5af1d77658133af8be8c9b1a13139722315c3a/torchvision/datasets/folder.py#L93 https://pytorch.org/vision/stable/_modules/torchvision/datasets/folder.html#DatasetFolder.fetch_img_paths

        class domainlab.dsets.dset_subfolder.DsetSubFolder(root, loader, list_class_dir, extensions=None, transform=None, target_transform=None, is_valid_file=None)[source]
        -

        Bases: DatasetFolder

        +

        Bases: torchvision.datasets.folder.DatasetFolder

        Only use user provided class names, ignore the other subfolders :param list_class_dir: list of class directories to use as classes

        @@ -471,7 +463,7 @@

        Submodules domainlab.dsets.dset_subfolder.fetch_img_paths(path_dir, class_to_idx, extensions=None, is_valid_file=None)[source]
        -
        Parameters:
        +
        Parameters

        -

        domainlab.dsets.utils_data module

        +

        domainlab.dsets.utils_data module

        Utilities for dataset

        class domainlab.dsets.utils_data.DsetInMemDecorator(dset, name=None)[source]
        -

        Bases: Dataset

        +

        Bases: torch.utils.data.dataset.Dataset

        fetch all items of a dataset into memory

        @@ -520,7 +512,7 @@

        Submodules domainlab.dsets.utils_data.plot_ds(dset, f_name, batchsize=32)[source]
        -
        Parameters:
        +
        Parameters

        @@ -609,7 +601,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.exp.html b/docs/build/html/domainlab.exp.html index 305b4d5a3..dcf8410d9 100644 --- a/docs/build/html/domainlab.exp.html +++ b/docs/build/html/domainlab.exp.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -319,13 +318,6 @@ Causal IRL - -
      26. - - - Invariant Causal Mechanisms through Distribution Matching - -
      27. @@ -347,9 +339,9 @@
      28. domainlab.exp.exp_cuda_seed module
      29. -
      30. domainlab.exp.exp_main module +
      31. domainlab.exp.exp_main module
      32. -
      33. domainlab.exp.exp_utils module +
      34. domainlab.exp.exp_utils module
      35. Module contents
      36. @@ -365,12 +357,12 @@
        -

        domainlab.exp package

        +

        domainlab.exp package

        -

        Submodules

        +

        Submodules

        -

        domainlab.exp.exp_cuda_seed module

        +

        domainlab.exp.exp_cuda_seed module

        Random seed should be set from command line to ensure reproducibility: https://pytorch.org/docs/stable/notes/randomness.html https://discuss.pytorch.org/t/difference-between-torch-manual-seed-and-torch-cuda-manual-seed/13848/6

        @@ -379,154 +371,14 @@

        Submodulesdomainlab.exp.exp_cuda_seed.set_seed(seed)[source]

        -
        -

        domainlab.exp.exp_main module

        -

        experiment

        -
        -
        -class domainlab.exp.exp_main.Exp(args, task=None, model=None, observer=None, visitor=<class 'domainlab.exp.exp_utils.AggWriter'>)[source]
        -

        Bases: object

        -

        Exp is combination of Task, Algorithm, and Configuration (including random seed)

        -
        -
        -clean_up()[source]
        -

        to be called by a decorator

        -
        -
        -
        -execute(num_epochs=None)[source]
        -

        train model -check performance by loading persisted model

        -
        -
        +
        +

        domainlab.exp.exp_main module

        -
        -

        domainlab.exp.exp_utils module

        -
        -
        This module contains 3 classes inheriting:

        ExpProtocolAggWriter(AggWriter(ExpModelPersistVisitor))

        -
        -
        -
        -
        -class domainlab.exp.exp_utils.AggWriter(host)[source]
        -

        Bases: ExpModelPersistVisitor

        -
          -
        1. aggregate results to text file.

        2. -
        3. all dependencies are in the constructor

        4. -
        -
        -
        -confmat_to_file(confmat, confmat_filename)[source]
        -

        Save confusion matrix as a figure

        -
        -
        Parameters:
        -

        confmat – confusion matrix.

        -
        -
        -
        -
        -
        -first_line(dict_cols)[source]
        -

        generate header of the results aggregation file

        -
        -
        -
        -get_cols()[source]
        -

        call the same function to always get the same columns configuration

        -
        -
        -
        -get_fpath(dirname='aggrsts')[source]
        -

        for writing and reading, the same function is called to ensure name -change in the future will not break the software

        -
        -
        -
        -to_file(str_line)[source]
        -
        -
        Parameters:
        -

        str_line

        -
        -
        -
        -
        -
        -
        -class domainlab.exp.exp_utils.ExpModelPersistVisitor(host)[source]
        -

        Bases: object

        -

        This class couples with Task class attributes

        -
        -
        -clean_up()[source]
        -
        -
        -
        -load(suffix=None)[source]
        -

        load pre-defined model name from disk -the save function is the same class so to ensure load will ways work

        -
        -
        -
        -mk_model_na(tag=None, dd_cut=19)[source]
        -
        -
        Parameters:
        -

        tag – for git commit hash for example

        -
        -
        -
        -
        -
        -model_dir = 'saved_models'
        -
        -
        -
        -model_suffix = '.model'
        -
        -
        -
        -remove(suffix=None)[source]
        -

        remove model after use

        -
        -
        -
        -save(model, suffix=None)[source]
        -
        -
        Parameters:
        -

        model

        -
        -
        -
        -
        -
        -
        -class domainlab.exp.exp_utils.ExpProtocolAggWriter(host)[source]
        -

        Bases: AggWriter

        -

        AggWriter tailored to experimental protocol -Output contains additionally index, exp task, te_d and params.

        -
        -
        -confmat_to_file(confmat, confmat_filename)[source]
        -

        Save confusion matrix as a figure

        -
        -
        Parameters:
        -

        confmat – confusion matrix.

        -
        -
        -
        -
        -
        -get_cols()[source]
        -

        columns

        -
        -
        -
        -get_fpath(dirname=None)[source]
        -

        filepath

        -
        -
        +
        +

        domainlab.exp.exp_utils module

        -

        Module contents

        +

        Module contents

        @@ -577,7 +429,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.exp_protocol.html b/docs/build/html/domainlab.exp_protocol.html index 0beac9db8..dcf5c67a9 100644 --- a/docs/build/html/domainlab.exp_protocol.html +++ b/docs/build/html/domainlab.exp_protocol.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -319,13 +318,6 @@ Causal IRL - -
      37. - - - Invariant Causal Mechanisms through Distribution Matching - -
      38. @@ -347,7 +339,7 @@
      39. domainlab.exp_protocol.aggregate_results module
      40. -
      41. domainlab.exp_protocol.run_experiment module +
      42. domainlab.exp_protocol.run_experiment module
      43. Module contents
      44. @@ -363,12 +355,12 @@
        -

        domainlab.exp_protocol package

        +

        domainlab.exp_protocol package

        -

        Submodules

        +

        Submodules

        -

        domainlab.exp_protocol.aggregate_results module

        +

        domainlab.exp_protocol.aggregate_results module

        Functions to join the csv result files generate by different jobs of the benchmarks into a single csv file.

        @@ -387,7 +379,7 @@

        Submodules -
        Parameters:
        +
        Parameters
        • input_files – List of csv files with identical header.

        • output_file – Output csv file.

        • @@ -396,52 +388,11 @@

          Submodules -

          domainlab.exp_protocol.run_experiment module

          -

          Runs one task for a single hyperparameter sample for each leave-out-domain -and each random seed.

          -
          -
          -domainlab.exp_protocol.run_experiment.convert_dict2float(dict_in)[source]
          -

          convert scientific notation from 1e5 to 10000

          -
          -
          -
          -domainlab.exp_protocol.run_experiment.load_parameters(file: str, index: int) tuple[source]
          -

          Loads a single parameter sample -@param file: csv file -@param index: index of hyper-parameter

          -
          -
          -
          -domainlab.exp_protocol.run_experiment.run_experiment(config: dict, param_file: str, param_index: int, out_file: str, start_seed=None, misc=None, num_gpus=1)[source]
          -

          Runs the experiment several times:

          -
          -
          for test_domain in test_domains:
          -
          for seed from startseed to endseed:

          evaluate the algorithm with test_domain, initialization with seed

          -
          -
          -
          -
          -
          -
          Parameters:
          -
            -
          • config – dictionary from the benchmark yaml

          • -
          • param_file – path to the csv with the parameter samples

          • -
          • param_index – parameter index that should be covered by this task,

          • -
          -
          -
          -

          currently this correspond to the line number in the csv file, or row number -in the resulting pandas dataframe -:param out_file: path to the output csv -:param start_seed: random seed to start for stochastic variations of pytorch -:param misc: optional dictionary of additional parameters, if any.

          -

          # FIXME: we might want to run the experiment using commandline arguments

          -
          +
          +

          domainlab.exp_protocol.run_experiment module

          -

          Module contents

          +

          Module contents

        @@ -492,7 +443,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.html b/docs/build/html/domainlab.html index f09ed15ce..d7b576f4d 100644 --- a/docs/build/html/domainlab.html +++ b/docs/build/html/domainlab.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -246,9 +245,9 @@
      45. domainlab.arg_parser module
      46. -
      47. domainlab.cli module +
      48. domainlab.cli module
      49. -
      50. domainlab.mk_exp module +
      51. domainlab.mk_exp module
      52. Module contents
      53. @@ -344,13 +343,6 @@ Causal IRL - -
      54. - - - Invariant Causal Mechanisms through Distribution Matching - -
      55. @@ -374,9 +366,9 @@
      56. domainlab.arg_parser module
      57. -
      58. domainlab.cli module +
      59. domainlab.cli module
      60. -
      61. domainlab.mk_exp module +
      62. domainlab.mk_exp module
      63. Module contents
      64. @@ -392,9 +384,9 @@
        -

        domainlab package

        +

        domainlab package

        -

        Subpackages

        +

        Subpackages

        -

        Submodules

        +

        Submodules

        -

        domainlab.arg_parser module

        +

        domainlab.arg_parser module

        Command line arguments

        class domainlab.arg_parser.ParseValuesOrKeyValuePairs(option_strings, dest, nargs=None, const=None, default=None, type=None, choices=None, required=False, help=None, metavar=None)[source]
        -

        Bases: Action

        +

        Bases: argparse.Action

        Class used for arg parsing where values are provided in a key value format

        @@ -621,41 +613,14 @@

        Submodules -

        domainlab.cli module

        -

        command-line interface (CLI) for the domainlab package

        -
        -
        -domainlab.cli.domainlab_cli()[source]
        -

        Function used to run domainlab as a command line tool for the package installed with pip.

        -
        +
        +

        domainlab.cli module

        -
        -

        domainlab.mk_exp module

        -

        make an experiment

        -
        -
        -domainlab.mk_exp.mk_exp(task, model, trainer: str, test_domain: str, batchsize: int, nocu=False)[source]
        -

        Creates a custom experiment. The user can specify the input parameters.

        -
        -
        Input Parameters:
          -
        • task: create a task to a custom dataset by importing “mk_task_dset” function from

        • -
        -

        “domainlab.tasks.task_dset”. For more explanation on the input params refer to the -documentation found in “domainlab.tasks.task_dset.py”. -- model: create a model [NameOfModel] by importing “mk_[NameOfModel]” function from -“domainlab.models.model_[NameOfModel]”. For a concrete example and explanation of the input -params refer to the documentation found in “domainlab.models.model_[NameOfModel].py” -- trainer: string, -- test_domain: string, -- batch size: int

        -
        -
        -

        Returns: experiment

        -
        +
        +

        domainlab.mk_exp module

        -

        Module contents

        +

        Module contents

        globals for the whole package

        @@ -712,7 +677,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.models.html b/docs/build/html/domainlab.models.html index 3b97e935a..03be90eae 100644 --- a/docs/build/html/domainlab.models.html +++ b/docs/build/html/domainlab.models.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -319,13 +318,6 @@ Causal IRL - -
      65. - - - Invariant Causal Mechanisms through Distribution Matching - -
      66. @@ -383,17 +375,17 @@
        -

        domainlab.models package

        +

        domainlab.models package

        -

        Submodules

        +

        Submodules

        -

        domainlab.models.a_model module

        +

        domainlab.models.a_model module

        operations that all kinds of models should have

        class domainlab.models.a_model.AModel[source]
        -

        Bases: Module

        +

        Bases: torch.nn.modules.module.Module

        operations that all models (classification, segmentation, seq2seq)

        @@ -410,13 +402,13 @@

        Submodulesabstract cal_task_loss(tensor_x, tensor_y)[source]

        Calculate the task loss

        -
        Parameters:
        +
        Parameters
        • tensor_x – input

        • tensor_y – label

        -
        Returns:
        +
        Returns

        task loss

        @@ -445,7 +437,7 @@

        Submodulesforward(tensor_x, tensor_y, tensor_d, others=None)[source]

        forward.

        -
        Parameters:
        +
        Parameters
        -

        domainlab.models.interface_vae_xyd module

        +

        domainlab.models.interface_vae_xyd module

        Base Class for XYD VAE

        @@ -676,7 +668,7 @@

        Submodules -
        Parameters:
        +
        Parameters

        @@ -1020,7 +1012,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.tasks.html b/docs/build/html/domainlab.tasks.html index 707575b8b..47c4a2f83 100644 --- a/docs/build/html/domainlab.tasks.html +++ b/docs/build/html/domainlab.tasks.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -319,13 +318,6 @@ Causal IRL - -
      67. - - - Invariant Causal Mechanisms through Distribution Matching - -
      68. @@ -387,17 +379,17 @@
        -

        domainlab.tasks package

        +

        domainlab.tasks package

        -

        Submodules

        +

        Submodules

        -

        domainlab.tasks.a_task module

        +

        domainlab.tasks.a_task module

        Base class for Task

        class domainlab.tasks.a_task.NodeTaskDG(succ=None)[source]
        -

        Bases: AbstractChainNodeHandler

        +

        Bases: domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler

        Domain Generalization Classification Task

        @@ -417,7 +409,7 @@

        Submodulesget_list_domains_tr_te(tr_id, te_id)[source]

        For static DG task, get train and test domains list.

        -
        Parameters:
        +
        Parameters
        @@ -447,7 +439,7 @@

        Submodules is_myjob(request)[source]
        -
        Parameters:
        +
        Parameters

        request – string

        @@ -457,7 +449,7 @@

        Submodulesproperty isize

        isize

        -
        Type:
        +
        Type

        getter for input size

        @@ -495,12 +487,12 @@

        Submodules -

        domainlab.tasks.a_task_classif module

        +

        domainlab.tasks.a_task_classif module

        Abstract class for TaskClassif

        class domainlab.tasks.a_task_classif.NodeTaskDGClassif(succ=None)[source]
        -

        Bases: NodeTaskDG

        +

        Bases: domainlab.tasks.a_task.NodeTaskDG

        abstract class for classification task

        @@ -520,12 +512,12 @@

        Submodules -

        domainlab.tasks.b_task module

        +

        domainlab.tasks.b_task module

        Use dictionaries to create train and test domain split

        class domainlab.tasks.b_task.NodeTaskDict(succ=None)[source]
        -

        Bases: NodeTaskDG

        +

        Bases: domainlab.tasks.a_task.NodeTaskDG

        Use dictionaries to create train and test domain split

        @@ -545,12 +537,12 @@

        Submodules -

        domainlab.tasks.b_task_classif module

        +

        domainlab.tasks.b_task_classif module

        Use dictionaries to create train and test domain split

        class domainlab.tasks.b_task_classif.NodeTaskDictClassif(succ=None)[source]
        -

        Bases: NodeTaskDict, NodeTaskDGClassif

        +

        Bases: domainlab.tasks.b_task.NodeTaskDict, domainlab.tasks.a_task_classif.NodeTaskDGClassif

        Use dictionaries to create train and test domain split

        @@ -566,7 +558,7 @@

        Submodules -

        domainlab.tasks.task_dset module

        +

        domainlab.tasks.task_dset module

        Use dictionaries to create train and test domain split

        @@ -577,12 +569,12 @@

        Submodules -

        domainlab.tasks.task_folder module

        +

        domainlab.tasks.task_folder module

        When class names and numbers does not match across different domains

        class domainlab.tasks.task_folder.NodeTaskFolder(succ=None)[source]
        -

        Bases: NodeTaskDictClassif

        +

        Bases: domainlab.tasks.b_task_classif.NodeTaskDictClassif

        create dataset by loading files from an organized folder then each domain correspond to one dataset

        @@ -590,7 +582,7 @@

        Submodulesproperty dict_domain2imgroot

        “xx/yy/zz”}

        -
        Type:
        +
        Type

        {“domain name”

        @@ -609,7 +601,7 @@

        Submodules
        class domainlab.tasks.task_folder.NodeTaskFolderClassNaMismatch(succ=None)[source]
        -

        Bases: NodeTaskFolder

        +

        Bases: domainlab.tasks.task_folder.NodeTaskFolder

        when the folder names of the same class from different domains have different names

        @@ -620,7 +612,7 @@

        Submodules -

        domainlab.tasks.task_folder_mk module

        +

        domainlab.tasks.task_folder_mk module

        When class names and numbers does not match across different domains

        @@ -650,7 +642,7 @@

        Submodules -

        domainlab.tasks.task_mini_vlcs module

        +

        domainlab.tasks.task_mini_vlcs module

        test task for image size 224

        @@ -659,12 +651,12 @@

        Submodules -

        domainlab.tasks.task_mnist_color module

        +

        domainlab.tasks.task_mnist_color module

        Color MNIST with palette

        class domainlab.tasks.task_mnist_color.NodeTaskMNISTColor10(succ=None)[source]
        -

        Bases: NodeTaskDictClassif

        +

        Bases: domainlab.tasks.b_task_classif.NodeTaskDictClassif

        Use the deafult palette with 10 colors

        @@ -698,7 +690,7 @@

        Submodules -

        domainlab.tasks.task_pathlist module

        +

        domainlab.tasks.task_pathlist module

        The class TaskPathList provides the user an interface to provide a file with each line consisting of a pair, where the first slot contains the path (either absolute or relative if the user knows from where this package is @@ -707,7 +699,7 @@

        Submodules
        class domainlab.tasks.task_pathlist.NodeTaskPathListDummy(succ=None)[source]
        -

        Bases: NodeTaskDictClassif

        +

        Bases: domainlab.tasks.b_task_classif.NodeTaskDictClassif

        typedef class so that other function can use isinstance

        @@ -720,7 +712,7 @@

        Submodulesdomainlab.tasks.task_pathlist.mk_node_task_path_list(isize, img_trans_te, list_str_y, img_trans_tr, dict_class_label_ind2name, dict_domain2imgroot, dict_d2filepath_list_img_tr, dict_d2filepath_list_img_val, dict_d2filepath_list_img_te, succ=None)[source]

        mk_node_task_path_list.

        -
        Parameters:
        +
        Parameters

        tensor1hot2ind.

        -
        Parameters:
        +
        Parameters

        tensor_label

        -

        domainlab.tasks.utils_task_dset module

        +

        domainlab.tasks.utils_task_dset module

        task specific dataset operation

        class domainlab.tasks.utils_task_dset.DsetIndDecorator4XYD(dset)[source]
        -

        Bases: Dataset

        +

        Bases: torch.utils.data.dataset.Dataset

        For dataset of x, y, d, decorate it wih index

        class domainlab.tasks.utils_task_dset.DsetZip(dset1, dset2, name=None)[source]
        -

        Bases: Dataset

        +

        Bases: torch.utils.data.dataset.Dataset

        enable zip return in getitem: x_1, y_1, x_2, y_2 to avoid always the same match, the second dataset does not use the same idx in __get__item() but instead, a random one

        -

        domainlab.tasks.zoo_tasks module

        +

        domainlab.tasks.zoo_tasks module

        all available tasks for domainlab

        @@ -898,7 +890,7 @@

        Submodules -

        Module contents

        +

        Module contents

        @@ -949,7 +941,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/domainlab.utils.html b/docs/build/html/domainlab.utils.html index 47174311b..ba4aa50c9 100644 --- a/docs/build/html/domainlab.utils.html +++ b/docs/build/html/domainlab.utils.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -319,13 +318,6 @@ Causal IRL - -
      69. - - - Invariant Causal Mechanisms through Distribution Matching - -
      70. @@ -395,12 +387,12 @@
        -

        domainlab.utils package

        +

        domainlab.utils package

        -

        Submodules

        +

        Submodules

        -

        domainlab.utils.flows_gen_img_model module

        +

        domainlab.utils.flows_gen_img_model module

        class domainlab.utils.flows_gen_img_model.FlowGenImgs(model, device)[source]
        @@ -421,7 +413,7 @@

        Submodules -

        domainlab.utils.generate_benchmark_plots module

        +

        domainlab.utils.generate_benchmark_plots module

        generate the benchmark plots by calling the gen_bencmark_plots(…) function

        @@ -470,7 +462,7 @@

        Submodules
        -domainlab.utils.generate_benchmark_plots.gen_plots(dataframe: DataFrame, output_dir: str, use_param_index: bool)[source]
        +domainlab.utils.generate_benchmark_plots.gen_plots(dataframe: pandas.core.frame.DataFrame, output_dir: str, use_param_index: bool)[source]

        dataframe: dataframe with columns [‘param_index’,’task’,’ algo’,’ epos’,’ te_d’,’ seed’,’ params’,’ acc’,’precision’,…]

        @@ -534,20 +526,20 @@

        Submodules -

        domainlab.utils.get_git_tag module

        +

        domainlab.utils.get_git_tag module

        domainlab.utils.get_git_tag.get_git_tag(print_diff=False)[source]
        -

        domainlab.utils.hyperparameter_gridsearch module

        +

        domainlab.utils.hyperparameter_gridsearch module

        gridsearch for the hyperparameter space

        def add_next_param_from_list is an recursive function to make cartesian product along all the scalar hyper-parameters, this resursive function is used in def grid_task

        -domainlab.utils.hyperparameter_gridsearch.add_next_param_from_list(param_grid: dict, grid: dict, grid_df: DataFrame)[source]
        +domainlab.utils.hyperparameter_gridsearch.add_next_param_from_list(param_grid: dict, grid: dict, grid_df: pandas.core.frame.DataFrame)[source]

        can be used in a recoursive fassion to add all combinations of the parameters in param_grid to grid_df param_grid: dictionary with all possible values for each parameter

        @@ -585,7 +577,7 @@

        Submodules
        -domainlab.utils.hyperparameter_gridsearch.grid_task(grid_df: DataFrame, task_name: str, config: dict, shared_df: DataFrame)[source]
        +domainlab.utils.hyperparameter_gridsearch.grid_task(grid_df: pandas.core.frame.DataFrame, task_name: str, config: dict, shared_df: pandas.core.frame.DataFrame)[source]

        create grid for one sampling task for a method and add it to the dataframe

        @@ -634,7 +626,7 @@

        Submodules
        -domainlab.utils.hyperparameter_gridsearch.sample_gridsearch(config: dict, dest: Optional[str] = None) DataFrame[source]
        +domainlab.utils.hyperparameter_gridsearch.sample_gridsearch(config: dict, dest: Optional[str] = None) pandas.core.frame.DataFrame[source]

        create the hyperparameters grid according to the given config, which should be the dictionary of the full benchmark config yaml. @@ -651,7 +643,7 @@

        Submodules -

        domainlab.utils.hyperparameter_retrieval module

        +

        domainlab.utils.hyperparameter_retrieval module

        retrieval for hyperparameters

        @@ -660,7 +652,7 @@

        Submodules -

        domainlab.utils.hyperparameter_sampling module

        +

        domainlab.utils.hyperparameter_sampling module

        Samples the hyperparameters according to a benchmark configuration file.

        # Structure of this file: - Class Hyperparameter @@ -669,7 +661,7 @@

        Submodules
        class domainlab.utils.hyperparameter_sampling.CategoricalHyperparameter(name: str, config: dict)[source]
        -

        Bases: Hyperparameter

        +

        Bases: domainlab.utils.hyperparameter_sampling.Hyperparameter

        A sampled hyperparameter, which is constraint to fixed, user given values and datatype

        @@ -714,7 +706,7 @@

        Submodules
        class domainlab.utils.hyperparameter_sampling.ReferenceHyperparameter(name: str, config: dict)[source]
        -

        Bases: Hyperparameter

        +

        Bases: domainlab.utils.hyperparameter_sampling.Hyperparameter

        Hyperparameter that references only a different one. Thus, this parameter is not sampled but set after sampling.

        @@ -732,7 +724,7 @@

        Submodules
        class domainlab.utils.hyperparameter_sampling.SampledHyperparameter(name: str, config: dict)[source]
        -

        Bases: Hyperparameter

        +

        Bases: domainlab.utils.hyperparameter_sampling.Hyperparameter

        A numeric hyperparameter that shall be sampled

        @@ -748,12 +740,12 @@

        Submodules
        -domainlab.utils.hyperparameter_sampling.check_constraints(params: List[Hyperparameter], constraints) bool[source]
        +domainlab.utils.hyperparameter_sampling.check_constraints(params: List[domainlab.utils.hyperparameter_sampling.Hyperparameter], constraints) bool[source]

        Check if the constraints are fulfilled.

        -domainlab.utils.hyperparameter_sampling.create_samples_from_shared_samples(shared_samples: DataFrame, config: dict, task_name: str)[source]
        +domainlab.utils.hyperparameter_sampling.create_samples_from_shared_samples(shared_samples: pandas.core.frame.DataFrame, config: dict, task_name: str)[source]

        add informations like task, G_MODEL_NA and constrainds to the shared samples Parameters: shared_samples: pd Dataframe with columns [G_METHOD_NA, G_MODEL_NA, ‘params’] @@ -762,12 +754,12 @@

        Submodules
        -domainlab.utils.hyperparameter_sampling.get_hyperparameter(name: str, config: dict) Hyperparameter[source]
        +domainlab.utils.hyperparameter_sampling.get_hyperparameter(name: str, config: dict) domainlab.utils.hyperparameter_sampling.Hyperparameter[source]

        Factory function. Instantiates the correct Hyperparameter

        -domainlab.utils.hyperparameter_sampling.get_shared_samples(shared_samples_full: DataFrame, shared_config_full: dict, task_config: dict)[source]
        +domainlab.utils.hyperparameter_sampling.get_shared_samples(shared_samples_full: pandas.core.frame.DataFrame, shared_config_full: dict, task_config: dict)[source]
        • creates a dataframe with columns [task, G_MODEL_NA, params],

        @@ -785,7 +777,7 @@

        Submodules
        -domainlab.utils.hyperparameter_sampling.sample_hyperparameters(config: dict, dest: Optional[str] = None, sampling_seed: Optional[int] = None) DataFrame[source]
        +domainlab.utils.hyperparameter_sampling.sample_hyperparameters(config: dict, dest: Optional[str] = None, sampling_seed: Optional[int] = None) pandas.core.frame.DataFrame[source]

        Samples the hyperparameters according to the given config, which should be the dictionary of the full benchmark config yaml. @@ -796,7 +788,7 @@

        Submodules
        -domainlab.utils.hyperparameter_sampling.sample_parameters(init_params: List[Hyperparameter], constraints, shared_config=None, shared_samples=None) dict[source]
        +domainlab.utils.hyperparameter_sampling.sample_parameters(init_params: List[domainlab.utils.hyperparameter_sampling.Hyperparameter], constraints, shared_config=None, shared_samples=None) dict[source]

        Tries to sample from the hyperparameter list.

        Errors if in 10_0000 attempts no sample complying with the constraints is found.

        @@ -814,7 +806,7 @@

        Submodules -

        domainlab.utils.logger module

        +

        domainlab.utils.logger module

        A logger for our software

        @@ -823,7 +815,7 @@

        Submodules
        -static get_logger(logger_name='logger_185015', loglevel='INFO')[source]
        +static get_logger(logger_name='logger_18603', loglevel='INFO')[source]

        returns a logger if no logger was created yet, it will create a logger with the name specified in logger_name with the level specified in loglevel. @@ -837,7 +829,7 @@

        Submodules -

        domainlab.utils.override_interface module

        +

        domainlab.utils.override_interface module

        domainlab.utils.override_interface.override_interface(interface_class)[source]
        @@ -856,7 +848,7 @@

        Submodules -

        domainlab.utils.perf module

        +

        domainlab.utils.perf module

        Classification Performance

        @@ -867,7 +859,7 @@

        Submodules classmethod cal_acc(model, loader_te, device)[source]

        -
        Parameters:
        +
        Parameters

        -

        domainlab.utils.u_import_net_module module

        +

        domainlab.utils.u_import_net_module module

        import external neural network implementation

        @@ -978,7 +970,7 @@

        Submodules -

        domainlab.utils.utils_class module

        +

        domainlab.utils.utils_class module

        domainlab.utils.utils_class.store_args(method)[source]
        @@ -986,7 +978,7 @@

        Submodules -

        domainlab.utils.utils_classif module

        +

        domainlab.utils.utils_classif module

        domainlab.utils.utils_classif.get_label_na(tensor_ind, list_str_na)[source]
        @@ -996,10 +988,10 @@

        Submodules domainlab.utils.utils_classif.logit2preds_vpic(logit)[source]

        -
        Logit:
        +
        Logit

        batch of logit vector

        -
        Returns:
        +
        Returns

        vector of one-hot, vector of probability, index, @@ -1014,7 +1006,7 @@

        Submodules -

        domainlab.utils.utils_cuda module

        +

        domainlab.utils.utils_cuda module

        choose devices

        @@ -1023,7 +1015,7 @@

        Submodules -

        domainlab.utils.utils_img_sav module

        +

        domainlab.utils.utils_img_sav module

        domainlab.utils.utils_img_sav.mk_fun_sav_img(path='.', nrow=8, folder_na='')[source]
        @@ -1036,7 +1028,7 @@

        Submodules -

        Module contents

        +

        Module contents

        @@ -1087,7 +1079,7 @@

        SubmodulesSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/genindex.html b/docs/build/html/genindex.html index 24cef9747..40a8b7039 100644 --- a/docs/build/html/genindex.html +++ b/docs/build/html/genindex.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -307,13 +306,6 @@ Causal IRL - -
      71. - - - Invariant Causal Mechanisms through Distribution Matching - -
      72. @@ -434,8 +426,6 @@

        A

      73. agg_main() (in module domainlab.exp_protocol.aggregate_results)
      74. agg_results() (in module domainlab.exp_protocol.aggregate_results) -
      75. -
      76. AggWriter (class in domainlab.exp.exp_utils)
      77. Alex4DeepAll (class in domainlab.compos.nn_zoo.nn_alex)
      78. @@ -609,12 +599,12 @@

        C

      79. (domainlab.models.a_model_classif.AModelClassif method)
      80. + +

    - + - - +
  • concat_ydx() (domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv.DecoderConcatLatentFcReshapeConv method) @@ -643,12 +629,6 @@

    C

  • config_img() (domainlab.compos.vae.zoo_vae_builders_classif.ChainNodeVAEBuilderClassifCondPriorBase method)
  • -
  • confmat_to_file() (domainlab.exp.exp_utils.AggWriter method) - -
  • construct_classifier() (domainlab.compos.vae.c_vae_builder_classif.ChainNodeVAEBuilderClassifCondPrior method)
  • construct_cond_prior() (domainlab.compos.vae.c_vae_builder_classif.ChainNodeVAEBuilderClassifCondPrior method) @@ -656,8 +636,6 @@

    C

  • Conv2d (class in domainlab.compos.nn_zoo.net_gated)
  • convert() (domainlab.compos.pcr.p_chain_handler.Request4Chain method) -
  • -
  • convert_dict2float() (in module domainlab.exp_protocol.run_experiment)
  • count_domain_class() (domainlab.tasks.b_task_classif.NodeTaskDictClassif method)
  • @@ -1008,13 +986,6 @@

    D

    -
  • - domainlab.cli - -
  • @@ -1220,6 +1191,8 @@

    D

  • module
  • +
    • domainlab.compos.vae.compos.encoder_dirichlet @@ -1234,8 +1207,6 @@

      D

    • module
    -
    • domainlab.compos.vae.compos.encoder_domain_topic_img2topic @@ -1374,20 +1345,6 @@

      D

    • -
    • - domainlab.exp.exp_main - -
    • -
    • - domainlab.exp.exp_utils - -
    • @@ -1402,20 +1359,6 @@

      D

    • -
    • - domainlab.exp_protocol.run_experiment - -
    • -
    • - domainlab.mk_exp - -
    • @@ -1747,8 +1690,6 @@

      D

    • module
    -
  • domainlab_cli() (in module domainlab.cli) -
  • dset_decoration_args_algo() (domainlab.algos.trainers.a_trainer.AbstractTrainer method)
  • -
  • first_line() (domainlab.exp.exp_utils.AggWriter method) -
  • Flatten (class in domainlab.compos.nn_zoo.net_adversarial)
  • FlowGenImgs (class in domainlab.utils.flows_gen_img_model) @@ -1965,12 +1896,6 @@

    G

  • get_base_domain_size4match_dg() (in module domainlab.algos.trainers.compos.matchdg_utils)
  • -
  • get_cols() (domainlab.exp.exp_utils.AggWriter method) - -
  • get_device() (in module domainlab.utils.utils_cuda)
  • get_dset_by_domain() (domainlab.tasks.b_task.NodeTaskDict method) @@ -1995,12 +1920,6 @@

    G

  • list_tr_domain_size (domainlab.algos.trainers.a_trainer.AbstractTrainer property)
  • -
  • load() (domainlab.exp.exp_utils.ExpModelPersistVisitor method) - -
  • -
  • load_parameters() (in module domainlab.exp_protocol.run_experiment) +
  • load() (domainlab.models.a_model.AModel method)
  • loader_te (domainlab.tasks.a_task.NodeTaskDG property)
  • -
  • model (domainlab.algos.trainers.a_trainer.AbstractTrainer property) -
  • -
  • model_dir (domainlab.exp.exp_utils.ExpModelPersistVisitor attribute)
  • model_sel (domainlab.algos.observers.c_obvisitor_cleanup.ObVisitorCleanUp property)
  • model_selection_epoch (domainlab.algos.msels.a_model_sel.AMSel property) -
  • -
  • model_suffix (domainlab.exp.exp_utils.ExpModelPersistVisitor attribute)
  • module @@ -2428,8 +2333,6 @@

    M

  • domainlab.algos.zoo_algos
  • domainlab.arg_parser -
  • -
  • domainlab.cli
  • domainlab.compos
  • @@ -2532,18 +2435,10 @@

    M

  • domainlab.exp
  • domainlab.exp.exp_cuda_seed -
  • -
  • domainlab.exp.exp_main -
  • -
  • domainlab.exp.exp_utils
  • domainlab.exp_protocol
  • domainlab.exp_protocol.aggregate_results -
  • -
  • domainlab.exp_protocol.run_experiment -
  • -
  • domainlab.mk_exp
  • domainlab.models
  • @@ -2821,8 +2716,6 @@

    R

  • ReferenceHyperparameter (class in domainlab.utils.hyperparameter_sampling)
  • register_external_node() (domainlab.algos.zoo_algos.AlgoBuilderChainNodeGetter method) -
  • -
  • remove() (domainlab.exp.exp_utils.ExpModelPersistVisitor method)
  • Request4Chain (class in domainlab.compos.pcr.p_chain_handler)
  • @@ -2861,8 +2754,6 @@

    R

  • round_to_discreate_grid_uniform() (in module domainlab.utils.hyperparameter_gridsearch)
  • round_vals_in_dict() (in module domainlab.utils.generate_benchmark_plots) -
  • -
  • run_experiment() (in module domainlab.exp_protocol.run_experiment)
  • @@ -2900,16 +2791,12 @@

    S

  • sav_add_title() (in module domainlab.utils.utils_img_sav)
  • -
  • save() (domainlab.exp.exp_utils.ExpModelPersistVisitor method) - -
  • - - +
    Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/index.html b/docs/build/html/index.html index 7d09a095d..5c713d306 100644 --- a/docs/build/html/index.html +++ b/docs/build/html/index.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -316,13 +315,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -352,7 +344,7 @@
    -

    Welcome to domainlab’s documentation!

    +

    Welcome to domainlab’s documentation!

    Contents:

    @@ -470,8 +462,8 @@

    Welcome to domainlab’s documentation!Examples -
  • Causal IRL
  • -
  • Invariant Causal Mechanisms through Distribution Matching
      +
    • Causal IRL
    • @@ -479,7 +471,7 @@

      Welcome to domainlab’s documentation! -

      Indices and tables

      +

      Indices and tables

      @@ -357,7 +349,7 @@
      -

      domainlab

      +

      domainlab

      • domainlab package
      • @@ -537,7 +529,7 @@

        domainlabSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/objects.inv b/docs/build/html/objects.inv index 43a8cfa24074458fde63243a1bfb5db14b647f05..6797c36343f3f2953ecb51948383900f09f093a1 100644 GIT binary patch delta 22685 zcmV)kK%l?1-T}PX0g!ut+)9=v`oEuo2YY~11$1WDR_z`%HqHznOSZRqB-ta`uI~Cn zP|3&?x0z%X$<)Qa`Z1WimaiLbykqN4kw|YCdAaIBo8B_j`y#> zmk&X}izGf&WShMJdeSsi_Hb+ns(?KK>_b%^NY%W$Knu(`Ib@Q5eY5riZLI70=c}7m zigPZvM#spNt+361v2MyeidQ()dY^nJ@k3H)o4BbcAusdWW3g$nvhdL{O^O0VLjW2( z_454oc2RPvKrvHEDWf7aY2xJgLaUHwp^WS)!SON$nkakq?R=S+6`<=KwO`6-%j}<( zfS8V?!YOJLXRPY`yPHK0o`=LV{xLlwL$-4ZmkQQv8hK2M#sX z_+PV+q*&w#4K9W!_BF}%-Bm_mbOWbKcw-sYhQ{3!bA5huuW!&LUzRvbU zrj0L&kB2mANDz2Z(<+z56t~JYhMw6H8Ba+bQ!j;)K$0qdjcJeN5W+{6id5K9K z#wp!5VV$*qm>kN@V;E88LrtnDQvI?Q1s7ig>l}qh0{TmY(P9uRhlVS`P_UXu>Ym3k zPcf9=#PByRt7MZCv|4z8x(3ZZv7g_vI%~>mah+_yN2{Sta!~7QlQ7JB&m@aS#59c> zn){kVnHBH@v)Bj~nkvrAy2j3W0+yNfrMeapFf!eL6gfyXm=O~o1I$RlYo*jOkm`3K zfL4mw*@r?heZM97XHspk^3Di0NXD7SK=1VF1R?hP6k^cx&Hx>^gV1)B^wnEOWz3^J zWfHf)146IQKRsA9*&`ty=r#>DhLmjsv$T00S`ckJlY$oP&~nUW0YM1NQ#MKcJ+vH) zdeBLK$^(`1)aPE*G9qEG@yp08I(ivVNWm{7^MW=>+y9IYqKOSE4!uau%Sn&>11zhh z8DBLCEUu=#E8eo(495`!VwciQFC}JaB$5cTw9d%rExLe!uLrckYaKbIQW*eu zef~hJ@?_I|>E)>0J{04JFFgK`G<=#^>2nle4`f?b6oLr8>0jilPuY{yY}J1usmp~!(v5a#V;D5C6t z*MD)u*=$`yk(6gJ!&SwcHLkNCC&`^%vV|68v5gKz&LJM`2n;T+WXTpje;g`Uy*MNl z^<&fa4T4HE&&!~QaU1-hh@=v`rDa^>_TS@@4~aB84qcF;u}Cw++JHxwt;Jm@Y08?U zGn%z(whxU9!_j|J?hmD9KbF(l+aTy6_s}0pcg%>B37WV~MA~{zR>4K~zUzF3wvzo_ z-@yN$lamt#A}43dS{zCG?Qws}IM42B?}Gn4rS)_2-_9G~vZ^^I`5oC_(su6beUn@T z8v;CMlX(RtEaNNfNyYW%kw9IEJ@Yd{U;hT*_wDCjl4^fEyni{A1t}WQuRxd&a+JD$ zW|PVVDSvz?q#&e}0)4$pj&+?SMf{yq1>P))N*oOjF5)%bK;?}khyCEkwvba?LT-(C z*NQ<%auvLonnoRFDrJee;*$!)oAXdgf|+8>IccS!yxg5eP=mZ!p%WA=H?Wz?G<(WY zf;UmX1#DB6sF{#SeeCtEbu76o< z;w?!Ub|ZoBt#n}W@(VXRl(lrjl7s8a>S=)271?Dq8}q-={|h<@E6S8O-{w^9V!r#z z{l?Ocb^RIsd`t2psXtZ!i@dU*u;WQ_*Me4&b#gf5uYCO6l--uJ*>u(e(T1$sS;n`d zc!01!@n)YoYogt^XWWA&2$Hxc41agJ`apHtIX}>{^!|nYyuBS~wfWRj@0;)zWYxMhLv17qO{re$LXcwyp;!J^k5(9YWJS%qUYrQ=LF_&ap^vXiut}poC%c?w- z#|9b%-A+J%1C)aB)5a(Q^rDWDcdH#X*M^f=#_O7n!@HAt`zRWTr`2{||u}{oArI!*Q zhnWQ8R(&RM^<(*bDt}pc`R}kXtesE-b5;YQL%expd!3RZ%}F5CZtbF5r}(WGr_98i zw8di9oIsz>h~gX;e=wz5E&U)0?61e+g<23zFiale0kCnXzV)*N83 z_w+xp;apo0c*+_3>>Wv-&w!-oG3!_US@jjDWw0+s*M1zn#BtHy!-; zmr&5E=P#|0+3qSn^$FK%Z1V(8rtTN6mMc)#V-bF9;ZKQSHJ=igbp)dJWhJXEHomA` z>&)JMb=qaqiGReA1;u-t?ewMFMkf6q-DeVJXN6RL(EMd7hwu1cJHOTYMT3bY#fil_ zw3ueG8zweMlLOrH%rj_39a$h{7A5}roQK`P2Q~BG{{3=cesK}SuUkkOV{bcvSUS9* zgg?@^68w5n!7L(vO$|C&?llu!I0Xvf(3@uQJ zY}!LltoQ#GGS`?|AkNE8T+vD0cgOPmmHhX8xotk@$2wHX zI4Kdlgu7`IJTJiICnRhyf{R$+USEICUPyjj!Xd|ObN0{w9W}+}kp;FX0-Gk<`=Twz z(jMUucBo{9$)F~o!Sdlmu*JauBpE02r0a12Pl0ioRoUh-Cr!u_&`Vhh^L{(Klm%la znpR+c&=k~54xgO-hQg*iD6jrSN>NErYw6W~clHGWsKI|t!=$TGY~a*2%Rgxv#~BrV z+SKkv&N!f^_6ARQqXwtT17=lDo*nqFvmfb6b3`Y#^}~Hqq|db4W4-{VoS;UqRphk! z!%Cyz+2rLlW$&HkImv_O3|Y|n-~Rnl82wp)l8v4T3^K!}w@G+bYzeH2!HnM9XmHZZ zHKRa#KgFrj9pt{b@!S%q$C zdP!jK7h&*uwF~86rN9$AR+j!=a`x$LZlwlqTn*w+(!3gU2%h$0Pfe(j^y78i&)^k* zWYfUeFSx>po&P%fYml5}5*jvR8KvW=Dc=bz_U?+U*SI?;H4@g}&fi;INoc*504^t# z_KeiXTc%X-HhWTx=luMwvlWlKfeJ|*q8Cny!b93W2kfUImQI1(Hoy02wn62U%}C3S&0Q8^ zfm)&gPN`a=0ZP&GkbOvUXgolF=%FePq-tKB#U4s{Xo1Dl10B#Zu}Q+nz?3IbNDH2@(u%Rf-$3zvPdCm44|XAyPkoNi7$9pGYo) z{%&OG`|mLbU`^dwS&+vP1ND%XO`O(vNL|0eAMUwI4!kaaN4Q9hkm6&=^fw(*c2W5SJ(}GQF1R2W$gx@5g zer?zX4xEe(*(CLMOib6eHeK;q?IQP6G$ zEu86bgQX_)Rgj=4M3rrR+1I=T%?@guG5LEj8L565Fqpe2;IZ*G6QRH! z;v%6*PPTktg@eiNv4kCGG|5PXRrG1g0?*hSUo_X}`eA}cj}?hUc~#C5xT4p|6TyOl z9#VVjBm$m!8;dP}qG5lBV-wje3voR!dTjp|a~@MPpu4PzUl%%6E1<699Vy`XrEpSc zncc>`eheoBt0`BBs0j2WnJMMN?_q?b*`BBNqKZxdPM(tdm^37e>L@)EH8y%jMInms z%Y2QW4%wy%BZ5D$%^m@UX^kSmtK`O#5W6G_5z_Nu;&Q2f5K?FlAU)fvXA%~|)1nPm z6%cr{7kcgCBw@5`Fr@fB+mT{vlhy&PlaxOXd?wx-?%-z!vvw+~?>Kc*kX-^fQcpnp zDMvehv<^=;ge}g$yrv7aoZzjm+7R8vUG930aoLTRzS-0cJwC)gUVeFi3d>iJC=JKD zDfbIKEkm_`<*LK=@URkZ%^vTDs*S5d>*zChQkz%ROT91@g-|1K@H8n36bAzk%(=IK ze#J~+Wlsr?QWS_OB>$MlHgCx!%Cb&yN+!f0kBB+zSN)56h2 zorN^Eo5fu7YH-;I&jJ~@b890IcT&kl;*vTJR8-A>lCwBX`ttaM4Gj?>*p@xGV*m2v ze14F?tWemAg7`V3>3nQh4@Z&i&cNMHA+etz6mHy9v2sM|g|2L4ozZMA1a}A3Z_o6= z%YLY5VV&Wmw4q6Zh|AaX8-fpsujNyc<6+i;N)oGjwNvm`kcKAErFt7|dFNVcTR$iL zrl5v@$*!g7EX`_KJ%tldCwth2Stn0^nc`u2!}Ku9SJ|fVN~{^2wOGa^Y|HKv3x}{! z$+KnBhjzCc9yUUi|A1Gl9uygfYPzIgnPF+a;#hR;02hGZj9F0#E@v{jOy=wD4h2RT z1IXEthK)|zU6Br2iM@uc`*eFE?bRB=@i_~BPJfo|cU4(7zBQjlO1C|-4gii-HK;|f zTzi|`Wc9wC)Iv;y5qb!F;{A85{79j?qSEJd@4~XDH9`E8Z|ddxbx<(#3n6ElN8P?K zEA{#f)d&jqU|kmIcV_?mC;xn9LdMNgR??RDe(hu}8VcEoSl%gQa$uXY<`9uXMrx6N zI(6H#KJ1A`|7@hk#mz*WELxcqR9xJJ>UTujV6%UCysmils6Op)qV$A z8^c-rEDd%NOi__tG8KU2ldJu`sF0U`qpUXw)(WUcJg8>`sB1RqWITvv56F&nGSIU$ ztCrVNP%m>l94x~L9D6JnNzi7Qf}LdrLUp`+3^~FQC#BM^Wc{LC&UaIi0sl2Scwxc8 z{N<}69m*QI2UaKgZS~^kenk7|mZZA&o;TuZyLkLIFQ56AMkq(R(-Gcy>kl-4>?0{? zJ*zlAB|&>&uG^Tt)Xp@_NHW?DFen0`ccEL}H9%5U!e$&FNU?d`Cl&2R=4G=iL#RO4 z59JH|z*e@!rQ?i}4qM^*mvACp>xS%;5F#L9XOaP3Mi}IM@)G|kenH2rk~FE}AlZiB z0`Qd3c7hI1QZbW{Dx%&bP`d|zQD4H2X^yP21}xsP*_tdx_nt zk$6{Sv?Wg}u`|wBNhK=V^jMG=aZ-gJ!PJhYef4LODzYUNO;S?F30y|;NH*VN{$eXd z@h_ozsr6%05q|sgo_3E)VJoA;m&k8->P5umcBlcqNG3+x74nA8%Uw}_R(m>zhrl5C z$0FNO5rG5*)t=_sIwnHmZjZ$f|Feh10ASXGK`zb1P3LPox>X zK(X2#Cb=Z=zEZ^%5g;>HaK3_w?Iu`FKnKbjtz^GX;N&xs%FNMXD;}1{E#~@y=E7ZB zWt$}bZ9w8RN^+L6hTAxQ8XDv`pmCSSb1km4fQ~zfw?;xB54Qy*l#CF`aM$*S0;Y#T zMxcaINekhI{!=-6g|r`*_O*9YPd>+255 zxoR6A@SPUDTFzC<5Uod=KW||GE1&4lDN9M*9TQySt0|m9f=^n)iTp*{0l@k*{ZG3o z(?Sawgk4FHY&L$0IYEP&QkY!FL=-!F;M{Y}m!$on#qAFD;)G~{n65YM_tIohL$XAY zHFvhXPKvB~<N&pPEL7sbH3r-AHL#4|fF>3vxbKM+$T(t7eh2T97sW!!~j) zRY*EZjiQ@`*WTIB9!aFdxqSFtKDo`V*$_u8y{+Ikx^~-t=)JY?an^NHR-`-fw@|BW zi=|iFFtChWBq+MA*)2wA(9YECUKJcvI1CP(8HKX#*3O0C$27t4C57ReEXh;m-?iaj zu)N-$FtyPUjd&S3tZ8= z_7mnXVLTQNbvv)7Xg_rytFl-ntH+ZX8X8U8=Zi>CaXlH8S;xAejX;vqMrE~6Hjf$o z3omIL8+M`O3O3kssU??Q9-GzfaC9}oSIo%D9wrhn|0Kag02BRznO|}@P@L%F=dXX% zlYtsoIMa&VjAT+l%luJ9MttK3L`G{G8$s-hCi~>DdLYd+A%$?~@8!cPOII}Gt@yM+ zx*AR!P_t9=Vv`XY93qE?=EW5yY_cR@wf7U7lvl8QV!@FJjzq9hEH%(gpv3S=lTRBL zf5gHgaagGfpOH+96QAUuD)$c=F9NG_ zyW;C4tV;unCo8c7YO|Uv8Mb~~DC44MkIHRy&zEwb?*vB?qI_TE z0=1~``gFB3%z(lBu6e|3J*qm5U+!~>N-@3zjJc=Vv$smKny<^~n{7!qGmO}329E3~ zBAG&^VYm5{_IEwI@V;hNorz}G-VK4M>HUUfEgiK`m+%iX!z+c zxWAK43yDi=hc?Z?=df5>t+IWxn|1Y2(2)9?eMS81*BNl8(+cih7z?>efYz>D zvsLSDlVk@d?g@6|D|3ZJYW4@0L*2;5@HjtZ4Io_2|N4*rL+SqOKmOA#*+vo;JJtqi zl*63d%DYjoBS>gML>XH_88a%AfgUme(UY_uF9IIRljR;oe~QQ9Yn=}EAR^hXTFU5K zPZ^y{ltV4m?R>906g7HU5B={qV zpO3=2NzgyE0Fean$BDgzXLV_*TcLOSTZP`WSDKv!b8=?*Oj*2A z6q8C6f02viSknx3b$2Z@{Ip@gEBdA8m5CHEqOsRtv=pK@FyQVxDninpY#tNZYueYV z=CLY|yT`%A5AUGQuE})r{4x8^kTCY9_mkvq`}sSXz7$gNK{z z%Z~#lO~F=AplfBT`BO6t*V*l#FJbS-GwKIF#%%}}Rq zU`4-m!UuFSUgGyFis!yI>NQx_u^tmL*Ng5s?|Mf=3ayZ?W6$DO!*~Y`^f*}RiH+9& zdbI#Ug+!=c6TW^M4_%`K(8GTQ*kLiLMg0v^ zK^&4(odnw5*8r=m{=R|>aQDi}Lant43MQid@V+I8?kGWY zBN6;juc>mckXoabh$7Do}q%A?qPHyX!w%-QGz>$NHDyQ>LWf#B>)m@Ojb2`XL(-(eMOXgAL6ql2?qa5kE zPrgILz)x?k_<04G7lR&&7+kF2EtlQuU-*t)xDk%;;3vuEvD>{sJ28Erq5~w~LmPii zEI`9S6b2pf6Rj{)2Ysh@5XGRQe~CJ1e(V5H+8#ViefEKyzMs{(?oiUs?L(G>1wXIa zRn_65R;y@hQ%XA>n}+5jopTlef9{!~%hRsn4p-G$BU^v8+!_ndSQhLaI=$oCRy&Exm-?O8;+z7G9(+{<;(`=d-3E2OM^2x70$Fud1J~s zm<=Ln@}7y5Y@gqN6g`vE^lWTVn|)}>>MN0c+B-R9dhIRvyG%xlSkqNM{Z#4s=aaT1 zP+lRavyDNdmKDY?>_CAS7mJz6cM!>yIM%d1j5qsqHkJBVzR|E-#Q6#l(Z?el#ZGa4 zO46rl-R)b}o-+S-^njC-RiRCNwx7 z{7hv$z*Y3DE{9tkVVf4KyVb~+0Ph7=#Hb=h#To)Z#hQqbW~$;Y3!g3S9R&Nadd#S9 z2-q753+dzS+bgy!8xk3c(y^$Iw4#2>YC7h#+wUW36G9w=|26vU>iymM-NkQL7qo)@ zc7Att{^jb^E&YA|^~qb4C`AEalb0wm1m|q?xRcE&ECHO8`Y1jMW#ri`qNC*4pTv_wDKb3nW;j5g z0ns1&eyw8v8%I<3C0bZTPeQQMKTkiad3iMbeEM1DYWr>UQ<{x1Zwl;iZ2{a|!e#^K zT>P(-uqi_U@sso^AO`o$an!OWlNu^f4B{$sXUaWr8w@1!kbe%1l8Bc zWlPN0XHFC~BHtDD)xtv$4t~MU8lKLI*$XSN;(Blg!_}D19B#&>Nl}cG6~Y$S!XQU}zdXOa zja(UUEJ4ehDm5+hq#Zkw&20BcoE%@`f;1s)9tv2-?r3xNluT@%c^KQpN`M@z3HS87 zGT>OU-oZwi#dV5G$pL@-5@sZ8(aNVvWy|tu#qMS^srF-&;Sq6v8v$^nd#}- zXD`8ZFjkAn>v4J;w?!Uc3j@WNHvx< zrJd40$}$#1&OV_wkCX-ykG7$WSJU3ZN}jy{Sn&2o4(r2&XBF+*?cgyMS~-iSpks&( zWUNXWBG$ZX<|p+ntW+=>*UmJ{6vJ5!Zym-shG^B=gGSNG@Z4Y5DU_LFfyUxe-^J<* z97FU1MJ>nfHYk7E$R#S;4c2T>w2=!`wCki>fau~D=x5s@wSv*dF3^&91eU8Qa17BN zin2!DUq8-1AL^k^*GBnk)I+Fabeh?&6g#dD1F2YXwFK^{@B80VXfY2_uQ%oXV6zlG zW|%!4J0eoA70-yDEI_w3D^6p#ls%`WHF6kp6g1*)^EiL0cC>?#rtHVrA*pIwFctT% zyV$x!ECvX!m)Kwabx5)*pkufscESB;kmSjSYd#-S!Go9ari3A2oa{p8j(K^DxPv(a z^22jh6Cvv~UKtBr4iUmX+hY;aK~Y^+bUaq=;rW&D{+SiJ93q5&U`C!@_r%%an#%cw zk;G~E;-!C=j))b)b^HrXeHB>~+*SO&5i0Epq%GhL?h2L?#c>~3<#T;nzYIlV0R{CK z6~Z~sAj8f-&>GHZx5qtyo$BnKN*51RA!41}b#IjU=4?mFj7H#~9)8NI=9uIa*`_sV z&PYOoPn<))aTV^!_7YThPgPi!O_hLe&VmbBK@!C(! zX=mgt=B~7JaBh=t`dK&AbepTL;7wBIU8^X*i=oX6e zjSqhg)r+UK14+x+@xAaswq-?5;)a2!h5W%B(Nw4LLQ_RcnDOn0`-v*Dv$(Z&u?q&G z7V?L6L|i`6q6gQ_@y`mLF+QPp=H-U~#rI=MHYJ57B(T4FIZK!W*bke}1wOoZ2OVX$ zIvUy^O1YR|`2Mmv9lF>#ild^|ocLzLK-Yg_o6=&-dpS#(@$GlmQo7g&`mLp|00Deq zb;@+P8&pS59eDAL&tavnspMsuz-lR4{}i$lN&?-cf2NZpV_ePZ`Wf>F^<;Ib2! zz3YKgJSbmJ?e{6$AKHe{Z}>#QwIA?(n9%KfjW@8nKCH2=5N}Rjv)K=RX%(NLOS7je#VN)rY*!2h4BJE? z`#?8Nq{%yz4mEE9u9JQ>wk(l2q1nUt0GN&L10mj=%go+p^<&irvJXUaIk-^!0|R&F zz&4LfYQO>Q_z9C>HYEWwlZG}j14#2sle;!C4VCYjD#C?3h1b~Dt+$i;HbMc%lRP&W z0mrjeH>3eMUwuq4NI9ocK2j+kj>SzYWdUz6)7KBwx6etHp2i2gI1G?*UfGt_37yhP zL%=|HNkt%-oTjMUag#GSDGCtCj`XNz1x)J?lV&+F0e6#_IU@wb|9dEtzd0Hbp#Xdc zI}7QQThd(6kaq>7_GgppIdK6alU+J%0VR{kIw%3Mlkz%t0^?JYYCALoSU#VVp*x`* zF)32G2{C!7X$!t-y1OexgtOv0A?HF!cl*R|lU_U`1nckR!;^$OFMrn{Rm>$IzjLZd z1}vzSt@rdlpKgJ*x}5}nRuqfJnf!skV>kGAkMra^E@vq~KpzBbPBjMkw5Zl_LA@2C zUDaCV)U?_l1#wTF{CExTlWUN&j6In3r!Gj>Q<<1x0sVBcB6K|vhlI1#F%|jSk<`sE zEELrTcLJHIAGx!jT7SlNQIo0(XJIG7pVif3VPX3hE+#x9FNe)OU1VkpD|9>u*!M4o zi+vh7(l0cqN$T&ImV5y`Tf&99KPBYhnB^&4eEN)O#JBVJ?>O+qZ-0rLInc~1iZE8) z-bStr8k%3d0VnEwDQ*No4^VvmMcMK0F46a!?Gq|Ha^M1C}D-*do+xm0r zXHf`H55H0OfKT#xTb28Go8?5~J5CIgK)BR(v!Yd&zGuWADCLK2ZZ!`F*Vci@Cd+G{ zOgqxVaD!2tX8X1L;pHAHU@tCP)8JmvqF8IITK($_T7LrtytXo&yt?4bt1wp4U$ikq z#cJ-n>-xNML$N@MA&x#)+2z&gIjX{!s5*u+5-e4CSeYWJJ_2c0)302H26@ z%QQvZwZteDq&1*-#V!{ z<7@izca=Q7Rwt74cajXv-)A~Br%H(C%~QheP=DP?Q$cz5E`X3X_P*r_wv)PFeb6n0 zl%NT~kLB2C!@0Z_Z&ClEb=-tfEs3{QFCxHz+n`AQ?e+C% z5OysITgJZ37o=eLG^dGQ!^^l49nng&us0#vAVg_@p>MBc&u&xPE z^O)?is$gtEkiBE|jLITss9sPp^8_*V6n|yI^jwNPZx|Pxq6`?{MsZ+9@j{$tRknG| zNz;Zro^?U^o{+=|qkcQP?8Gev2=9?_I)SIc6(_xgE@jIO3CUTECT_|@ zwrK;i{qBqfP(6MG$yBmXVd51P`UL{0rGK@|Qbw3l`6y%oG#cqv46Wf#=DUzZ(tm8N zTfwx)!*^fY`EACLbyH=59Vp44n$F~-BffzXPgInRrhAuaYvi|Z?up8}*$3Xm-X1!P zfBCO7T5k%9Iw{g;+Baa~pxlpB zZ|Lz8RvA_vSyXd4Nx*UX%WDc-jTVwTC;6Iwi10QjfLjL=|Lxx|g{hxglfOY11zure z+Z&VHK_gtl40b*cj}3Sd4~Aa9G+FGvs_BOK22Om#47p(O&lCD2{^ci^PVY{B)HFgv zI-#@PFcZEXDS!`o5-6DUZ+D z^M`Z5DQd=vIg9E*16n{=g)MVayF5#m8Yqx&X#Hn-@5o}eT;a}1Mn$*8L|?yG1;p+a{xY#FN0QbdTaeE zJ3CB`^E5|iY@MKkusbGiIv+GTDS)8Z1k#E~dnCwggp`0dCdhymB#YEn?!?0|MGfvq zj$qBOvE8VH*cUtXmlsIz z;W1k5bnw&X#3#^;1ClNxI+pwXw=5*BNmJ`ilyZX6VNw=2cV4rDla7C~Fw^PLUW}5{ zPrL|AV)z^HiwsI@nbW4C+C9WI{k2XYip8)98ViTq!lVqYJ z+Bu5}fqOI#N%I)zR2U1&_V5j#9?qD-1ck5mpXv8CU^;Uk8N*xzKjP+i$O)t6wgs?K zEesFJ!7p-7f?O1*WSbmw?xMa(K$M8_b&68bFxx_&y%c|vkZ^*!1FMblw9T%e!UPZR zO@tFOH9=?V>qy3eRoK5*eQNOr?nE1@f_TADDziwnN~fJIPb1~ z_TuXvqoko(H`z;sgcH;qMcD!QcOTqsx-Bmm`L((uu@Kb3YjnuJE!nXP7TA9IdZUK} z4eyT+AIg8xl$e>=OZzQpHjiBgtX{7F7|@N}%ZA6Kj<*@@8zy;PK9e+tH*1PIQ%}9x zo1?+@=Cm`CPkg3HlklE=u_t?2B`>{FOw*MXj95dmpEXZTI(55*+H*#ki4tX!x&_VYyy;~47iLZXK=crtTWJ%|u z?sJtKcvF{Ap{`z{M-aL=b^b7y?M?@qk{utavMFiBZ7AYB z-L!wKcX5g=HjpvSsv^oFrw>&y;oxgn9SaD6?Z!0Ym@5c7edkeC+R@Q@PduMX6t)bb zlHIN%@MsY{V?=vHHH5_5s@%h?DP+@>)vJeT9zquI;u+4`h%n|dXlD*Q)5PyU*0tz9 zOj;Y~>d(FuTN|T?(KiJS3QY|~QgH^@n$UlucCczpM*13C6}F#SyFpP9137xqcfWSbaXkwsE|2H1DQ(d@U*y6#p|E8zN6V_oS(^%`f{JrP&?8IneqlPrGT^At>6si~;)0;$rXipc^&%4>Gaf4vi#A=oMabQlDYPe_r-60ygi?!FnWyWz;LjpiL9SL}aca!}WX z9K{MT=-y>b{EBbU0NtMz)6%+*cZ3cB)2CqON+K5W;~9P+)i&lG(Li>~5|**!S&hdew$Zp5L=&L9>WGDj&uy4(-W@ zYhZkcf4uzi03C{ew#rhLL)dyiR7uCWDfew9$v(U+9iWN&jicF>7S7D$W>lIK1suVJzYB~Qz*x{YDw#d8bHea< ziIOsm1&y1Mq}l@`{j)$t3B*X@rru+E;KG`VJaApK^9XHY$Od`;?}I}q2Pp;$Blb!Wk{CvK{^AwjGX(7udI zEL1PtIa@SXo6{5eYRnd5AC7fakQyh04G|sbz0EK@9O{2&!SpFm6RE%6v~@>P8jvjM zeASgo3N0uodXFGo8N4mwH$5X&KRucQ?NZe%RMjg7)t7pm_V_V%2RIjEU+wD+cDr+qCYvaz@Ev`85cGr z+gP5hX$OBRh=CIN0$wCj>tqivl-0>o2va5W1-y2qK8&i3Y}0s{^f0OdvcY92N*Zq! z*Wz9w!fr0dQFw&u%;9egbn)43Ne}In*hm)50Q<4|+qQH7GN+$@2x8=LF=ko^P&of) z=|NF^DX6MpGP7XWv)+`)qU(FJz#t~h$w5Iua|C}*rqyb1lZWjA>?~)PP_SXOq5JT# zPXS~{=Af&1U#U$Z{k{GseFa$S%gM7_Ceo#P^Uc+q{w&+?s-%7Z;*oh$YUXwRLIp0K)dj!dXdsGqf|6elVE6@Lf)^(|mr3#eo;o z)J2nNb~ekRJE@*TqJ)?*B{! zu4%v}L-pW6GYPS#)@y8CPdmdn_=3F(B{su}8HX(OwRaxOp43in{w^7t3InrQ*Y=5Q zjPD9-w3i^^rA^l3{6;o&PYc^I;8?=m_+_DSUcw!3MRG>tnO7@Hbl-^!kPmeFH_dDtVoxsI>NH1bXA%!+rfX7xog^f(Hva1(x1aZ z8+BErUF-DA@ZvCgCV#7bWCtWe{z{Q7>Fm_~QI;|P1w&T&<)P^eh?&GyKlp#2O>M&r z!44i3uNJ8agZ2x+gAx2kj^{)@Z@}cd5Of$IaXy7D)w)A%WnA6 zV#d0S(a#(BTQwH2j2+iz&A1rB;NcBl&b$juJr=R%DkDeJS~Le*JJl&6VaG5YNU?d` zClzf@^Rl6R=kf)9c<670bY;M?B(9s1j=1Ya{h-(aqh$02yjZ5b)(zPwp-k=37w|4L z74rNj4Aa&*_$AB;)&_sE(Zezfd-lmo{HOQ@)0!62m0_VJRgxxEj9pbqs2)5^O)8H? z3a7#AI4fe^u|_Y`rsxZJ7n)jc5}5qNPCgmdhY8QBnAWvb_7cdr7JUJ)t*K26FRM1m zPX6s>);y9Wot?UmuLo*m`|5VQt1>!rO{&c!dzPjkFXBz^cw2v!d+pv$jU^sJmP26J zUy&`TXvQFQoY3ODc_f?f@unc^xaDffO2H!kvYU{vV-(F((Z zTIvBl7{TY4hhvtf5@1|zhw@MzW0r=9wbV5XR5wxR<$@m7gJ-EpdATdfYMQotZ44zRs{IhPt4Dh>S-5Ha;=U0=tfs>&+PwsCRX#|4aU4`c%mS>hUN zl7)t}*eR@oPOJSitHtx#tU};^$MuO`6c$h@kBBmM5|^~H(2$loMclN5UMa}9UR;QV z@sLpF3G@ZLOHB>dba}R0iyN&99$tpqxA}N4+3yp$TY9-|DqufSx7nO~>D4 zS6103$%iD%&Fo}c^fG=M%2t8zwlb)Bwi7suz(hN1bZLrF52Hc-@xt+n%dj{N}b(x*P($cea z!)(&Rw_{5iuOKad*2rwgF;5udOO@8GfMx8cHlsJq%i7}rjc``;_lDrXi1<5sMYMGS z?aR2#!fquYSQT~GO7N_!6#PNV4pDn4^mnhuOh92gf?mi~W;m91J)!2@CmuG0aYbwOaQa-?Pn{){=q=T}36AZQ4#dt2dAZwB7XHV>Ji&j>`L(eg3h^rgQFN1`Yzcid zM$*88y|ot?jvipLjo;;jW5x`w7ni+Z@gCO8HvEJC zUY6lIK+y4q{nllFH93Y2ESrj0bbAKi{;YO()=7~yuYA{}_+Y<1tQtVwvyTe~WlOkB zW7eO7YcXX4L^I~`7jZ8%xK5s6-K7wFUbkgsxWCkY$7;2Zu=AMdWVpu3QpqWv6Sb!SerG>tVQh%(v^-L_Ls>Oz{wLPmNT|_r7RH9i zP0P3->`C-vPQ*52c&Vc*Xti8E{4T}DTaq*oStVWvdKL=Ag|qY#yE%ZhpzFFRE24e< zT4(!zJdhokv#xDdh?Ot(o5^xbH(H>ag zC`^+zIa#dy5Y2&hUiHcCCfd7iBrO$;nhV{3OhV@_FYs1m7R-ibdS{DKZ`J8Wf8~{k z)`VVWxSXT30MW&H=%?K3>#waE(w=g;CEdneG(&nsYr^Gbw3F;;R&;>2bRITX&pzw< z6~P_~dCf7~V_A@RC=GK${%SJr*+5TK945wWP_&UADmqRN*`R16JydK*%l_(h2+9b5 z54GAsqQ6=lf-+)(TGHrrxrQ>B4ULc9{m5d0z7mKoZkc}B7DdCEx%y$Ozm{~EvsgvI z&_sDFCk6yd^=o1Du{tfyw;oEmC~iH>ij=%e?>gkc?1sa?kbG0_Nz{}*@c#|Db93C& zQAyHdW$}MWKtIu6bZAXG$`UhGWIi9X4RCvS6B6Fpo~(Q z6|<$e!huuIwGqSHAqNcopj6E&(MTW_Nzy%uY5(l|RI(h&q+b}i@LZ-#7$M$WOJEmBCjL65I<&+xTPyFmAaXjj`TH~HAaeZk`+ z=IlXy1o2aK(yXR^t%oCns2{U!Q;X8D;^W^u`@!)+tRkmQ)bvHYl=vdf_m3jOM?*$d zA#^W?%-ChCX)oIXY{klCQmqbuuQa0+?Y*lko3mqLbfnZ8niWq;l_fOC6@;ciPX9E| zW%a$b)d)veXV$Ngz>7FudpRsCe=T0?><%yDcu|Wtbq;x9JG@v*9_z{dwG;cM^zV5H zOHvjx6#}m|fVp*(RLywL=S0Mq`Xm32#P<|FD`-@n+Z6Eu>6P>`huT+vxJFcoR9{ik z)IU_^jtN65>c>gnP1GRLh)=K$v~<3?F|yJnYX8i zC`BYuh?-4SB`<5*Lpc_kt|g**KFy$$P!YyR9hJOJr?4cI>@k3UGHNMVUDr}ZEoHkY zOmbVYamoIW)sNMGs2*Tp!-{|fb@bXY*K^Mox(*?OAjObqh%=(R8hj2hyPBoT@;m^B zaF7eiW3wu^t3hXQ)Qi(8f&kxFYaiRP_N~QW`R~&dgZ36=jDgAfq}e>S)laWYdaPJr z+lkUxWr=Vj$XG4btIWx|-r7L7za6Qa+4dE1vH~tHphnSuG0yAifmQ>A6rzsQ<|2EL z|J_qY1IkEHiVAm_58Lvw?TUgzks;xvX#95hm6qFKQDR}dDBZ9E%>p`VR*;^u6i(8u z^75I2KfujPtkQ<5dfJg&u}XTZRu60m@Lo_wj4EPOti~5qtSuNSrYhmG?70%&Av~r> zTG>5jR6s?4v(ARrbS!&89;UyOV^glEPGt&73&rPH)JIw)KV>yZ=LcsxWkQH!@V`dC zUA@0Mzq|PD>Vj6q-_Gx@&c9rJx~0F*zkd0&dVhQI>GJCKWAx$EUG(+#{_E%Wci*n= z-(N~H9vtomsRg%*$)?_5Bnp)7!+L>Ll z2%SMhSv%P~fcVF#E~Bi8_^=Kzc#D-p97F2hj0`^3w6Bzt7Z?&-w5vU(t`}Ibo*3o)vBf)oaG-K{q~?G@y?Jy`9=MEu_!c=5dk26A-1U zEv1`(l1+KYHj6YIA}G~kWE$OA7Vi2BG?NeK9?Ytl0Jvr>mrGQmDT$$CgVk$h>~m7> zj|Z(XN~Y1FNN@^|(Fmijr;d0;Hk5~bg=1S^QqZ=WwBLB-W%R7Mr+CMnV+ zFALh!OrW0yE7yl3U*TC5v~$j7mRvS0v+NE>R}-dER#sw0Huf-)fcYm0CIXn~56t+I zJAvXvA3uNnqs9%;M)CE3!1+x=I^$+|ow7?x4y1s^fsE8E^|q5471W%UQcNVrOvSfn znpW&2(+aM0UTx~9nXP9t8fck6ipYqck%Gu*O=J1Kvxy?*Lj$!O9RxP8D61^tBNY^E zpIC4tf}?7#(Dzbl9qJ}eV)!Hi)H14L!FoZ-YFOXs#LV&B@i1C{2D-4Ruv+YNIv$+l zDcJ7CKB?nDFZEbtFK(H4%JG$k7&7Z~R;1j-W5NWnuQj#bU(Er{ir>6*y<@xB-3bM#Er zA<=7jtfJ577x%2M9z8=?Uc!qL>5M401&YX<{{U*mpjH3{J5Fmh7+3++NBO38Kj%p5jf4VH9(2GBu#+=hVm^j?dWTl!JMm+V# z4i~f(kxZd~P_VmgJEfx<_x!77@2|pPlSqLzqG?jNul0@vJ3xuMS}3W_a_^kRFw^I-C^WCD0fcM$U;pucDBXYk z$A21Dgxs}U{cO}TOMez>z|`de1s12P4bn(cmsI7wsn@ZGvSKBlH8kpK3FBMZ7c5JT zHT77Tq&ZPgTjWMoxJGpzgvlFGF3)doJ7Q^n89_eLVlu14v|#S4`TjV}E!X^Zr1`LF z(n*!5#zwU(EwTq_*uVoEwr zSDqFryzI%x5@cLeO$bCswcIw(Nk!;?C~EVVHH06rW5>`*O*_`};Jn63iRJ{KLg1L~ z0+zZ4C;8^3)k4ozBI(J`*)u(@ts@Gxv)Dg$1&inM+xX#poA|KT0COA4&Xt$kk1MZT_JJ zh$MhNj?EQ6;eav8%LCJ+ZGxV%WVNFO>G)u<$`VAR0ZqD&tZUUMS;Way8qzT)nH_}9 zs?t)oGVl1e0=#P?G|RPfVrKXZ8LgUT(@JDfu`}`N?z$!A^RsMkW6L;Hsp_U+JED0M z(X479S_;wb%FKO-pxtOmHjfE^?VRlERr6St$KB)bh@^5yXD`klv+s1YTRA( z`H;7I#`|3CF_TgN!bW>pulcS;9;NNNbj*sbON&68hon0bZ+RF_0=Hqmp)}or`uR<; z1C_dANd;oN87W9JmmbjYit&Lgr&gcj$UBOr9-te>gb}Dj`yNQ>eUT7P{X?p)k<7iK?Ob3><0Jw+mc2qI>6jh zQt)ASIGontoHjZaw2;HQ>C?%D|g$bIaWuKy^3O{r(m zVI6I<)XV}!RVy!)^#ztwK)c6FR+7WwVe`1Bor^8oCPVAJ_DO+%_S&NP^|4-d1qXfHYHmd-le*{X9{?P#l{s@Wywrqq|P04;|M;nj_iggW0Ebr{%xpTBE5po&+wGS6m_3`ha$l5LKP=%*wIp2?~0AW;{3bQO!#l1{4d>u;DO~&IGYju(X9}H$?MOD*qC7(0rd8 zptLgyF#PEQH(hJedG1iwb@q_4(<48x+IiFAG*+u;Yf?(<)=iVYYJIZ^_;b$;HG4Zx zJDf*rZEW4qa;ptIV>t=3R6aY!rci&xCM!p4c*u>AEw9CYv5a}qe`XlGh*%3AFv(*7 z%*FCjHWWI@jBKlR&Me?p(D0+5ECJ5sBskORpXz5~=4Vx6|M{01(*gz@M4Pzr6RQHQ%8nt5$7x##^ zr*9!u5@yJMzt(VFxoC`75SJibS0uN$JXqfmZ;7N8Ps3e416&nTyY@M~-ba>b)%hQY zbLsv|f}m%`cKWrwcOLbS)YE5eMiRBG&@*E@14VCNjE~+yB-4RiHyCgB={Rw^2(gi% zo4NVQt}Og`q}}o%0@#tHPYK(rGQ4_oV!;qy#da=#f?<bq6WF?4ZIm~|^1cmf?=7uoB&jAx#_%&!By&KmbX&lVzAZ9B)v17X_i9Dz z@>p$%*(Fnsc4zBd8!{C$EMn$efg_m)c2-zdial~O+VFv`OFo!#d=ep=Trk}Drx`Gg zgsJ#{pe^XFl*fqj6k}H(v=?X(Z*gvKQ0=+aFea(zd0Es5LC`C3*894hS5Tlaf>}(# zt3oaxXjApn<`)HeM>H>?$52TTb+J}ROEWK{LokV0!>ORsP7){83vBVGa&>~{kquDm zqKJQ#r52Hs%mcUlO8mp7(t^`?1u}3dVV@FzqJ^dL3RvLO!P@w^CsPYidp@PZ7X?0x z8e?+$DGqr#%%jkVF$U1iBE*NPY|2fUE1^1y8O)x03NFn|PGEW57*KFbp=PFx22o=m zPCv!z-$0w-^v@HF13R1@DcBAE!-~HRZC4ZN&wuF2gdD>g2LJUR|2etVe-GE!O{5)v zURP~I*LR0ULbzP{2?ilq@dki|V-WC}&t(ndNH;3Wvh=t9zx~bnXcz1MfAk`I%zNzr z|7+~k*Vv!p7W=T5*ax}8{zUh>%h#$8YcJ~y-L?0*|LLFq6yz@Vxi=AcnAju4<3B^{ cH`!re8+>!B|EDM;_H5|?{D=P+$5jmhdC2|=_W%F@ delta 23358 zcmV)nK%Kw5*#WiQ0g!ut)i(Fu6fGw7r@N!KNQ^<&JuUO&m&F*B0Ji(PrlRaoqg2szEp`Od}e_q5z6S&=7?(c$ga zeVHC}68$+!+tBq>v(JCx;j8~MTK$(kCdqeYt>9vDCf;=t!{7LS@tEZ)sSu&ow`EGs zX~1ti^d?X0I{JAY*&T9CUK>)ROli_Ql_?^0R%8v0th0jDbvOwHH6f-xC3zSbb-aK3 zqkIekUL^6MBHQd0(37UAvd3dXPzCG>U>~aTK&s}=1zKRn$sv>c>zlPFXk%T^@2+lI zDbBgv8XY57w!$`l$GR!^C|==E>wWT_#E(gxZQ`b)guKjekHx0R%ECv-G${%c4FPEE z)XVeR+eOKx0>w-vrHqQyq=}Q`E3HDBg)*||1jox1Xrk=ZxASFQR)DT|)P5KZiv%zl2$>Z~cN#dWd)AFYNq$w95JO~NqiJ(DaR5z{nk zXzptcWmdos%wi){XsS3b>l!=j30P*@m+D$bz{qreQ{*7oU`9-U3@{@Fua#2EK&sz` z09q+xXCDg5^!=9Pccj{4<((01kc=~tf!^uU2}11oDa4@XodG&-2chjM>8rPn%9uxc z$|P=o2ZUaqe}1%RvPVKZ&}|xQ3@O_NW@+<0v>@7cCIv0nq2-v%0)h~jr)-k?duTZp z^`Mh~lt(J%xzD|*WkkYWdUk_-7*G3q6S+*03U$m{{ z;uk$Dv4hEFtU_TWIb<2xks@gMp!mH(hUJKVB}NQ{;AKs*ps4R~w?sx4b%e=ar7{5S z`uver<;kY`(#uh~eJI8cUwHg6Y4|j;(&s3`9?7<>$cd#jRd`X*iK)o0pTatsw78@N zf%VJzXO+Yg}hPPLeyjWD70GVjCTboI^a=5g1%t$&xL6{y0>ydT~f9 z>c^(-8w8bTo|i!p<2Lw15lJO>OUt;%?Z3w*9};PH9J(MwW07WtwE>SVTZ_9+(v&qx zXEbZoY#$mIhNJ(c+#gEIek`Z8w?WV$@1Z}I?wAoL6Ety~h_v;btb&W|eb@O6Z6*7= zzJdSml9Lk!A}1HiS{zCG?Qws}IL{tv?}Gn4rS)_2-_IN0vZ^^I`90ZQ(su6bLz7$u z8v?v!lX(RtEYln9NyYW%i9lV6J@Yd{U;hr@5AEk)lWKoFe0V*S1t}WQuRxd&a+JD$ zVUx-QDS!M;NI^&`1^RlI9P2tuiugOJ3cOhol{gw6T*OaMk@v{+fCrtn+g`x#25oVA4~;?sd56$efGeu0hd?jxYfWIoDTYmI7wmEKdD-a z(0oMpNe~xN+^*kuj_RKQa=5P2P|Uu3CUHX3AlVUIIoqs$s)8f|LvC&O*or|*hbcVk zbbrCNzxWw)YOx{lu1eB6fWxM8_prq(5j7Xp;Uq#2$s729AVvs#Z9`dmVe|Eyz?A9RJPpm+g zf}0s` zo3*qO_uqMpSlc;?%rIk7wXkqSZ7usQh6DP^0ey~VpNA^bx#X^E?9w~}4_1!Hat;+a zBvjCk>_DIom!u@LNbEfJqbvjMJeQEW$%rTV2d)nKM4D!4AI-cm#iS}<^Q{j#yMJb} ziMJ$a*o_3fx6*;h%P-vQP}b57OAf9xtET~8S7evfY|Q^k|1anutSD3Be4A6Xi}~&= z_Zv$)*7Y6y`Ih8IQh%=g2YF*ZVaJo=t_7_k>*R3A-}v~sDZ4Fcv+1k{q77NMvy5*^ z@c?0e;>|vF)?|}s&Q7H~>+UNS2!GZBbsGhZo4y1=5*LNx5?LRpZvW^ky8;cZp!<87wAMajdkDYnMA4}XmKBc=Q( zl<>2S%B3xl-Uv=(wnflE>ph;r4}`rwKQ`s3DoH;ci<@KKWVBJyRxh%+g>u%9$>#e@ zQl!H5^NA1p{E`$kIfZ+fpp*KQOFWrF>F{*WVTE8*Z_P#KE@?=L zUH}>D8r*f3qGbaa6hFxa=s}W~OZWu;2l0%A9Lhbih|oU5twYAZL4Q3Is*Wgd3u>`u zir$h2$IU$+vUerm08U+BCvT*>O0k?E;Lv^0;s$Vm#Way~=Hin@Fdg*EB_W6NJV$cy z6gZemm=v1r)JIpA2^@vpFI+8GpsvRv{MN#s6T@mg zB{1s;K*}si{Pj5xyN3^I=D+^i<-+{p zB8uPdkTk~Lb^x(-ctHt&r0*s8?WBTPMEsV%;z;dY+!VIi+~Al?N7(DG$VsJ;M_rUf zdIyv33nBqOlMoCoP?>DnQ&z0^{~j{empuN0MPIY#_~YiDQ(Oq8v{}Js0D`3qZ$sl( zcETaf%S~L<5Moa9(<;dsts2~SQLOp9)68mV=Hyyguc{zx#&>PJTa%3p9e; zvIO)}*228s&n{)b*omeU7=JVc^^(IUC%+Q0DG$oKhLKWK64Y9Hwcnk6fdFdo-_kJY zY7`qdbGFVCm6K-&{_E^VdeR)xNp1b`kQC_) zt@fBNz$quF5o{GXZT_&*D0nt`eM{K`Yk5xcpgBVpwEowByA(#>Nq@4@Gl4;7*z`6D z&x$R9RWX>+dmjx>nz?2aXz!;ub-IJxcQ-!#pwCT%cXtTC2ID;(!6m=D1HRUzdJpqK z?n#VqVv>{Rq$#V=Eln>8?ENAPKCgD6{HqjrV#mtTKS<6#oz1P(;Ek(6{7ITugAT#d zUhJs}Rg!+Zt@|0gf`4opIQs=x7_swTXMYQlvrIz6W-Oz0{50h|Va48EvGp2v=cGo$ z`rG*jt2+>_w-Uf@h|->s8hOi<3f^Wuu&m!meQ+$k-0#!=?WlY=HVb~ z?ft_RJ@OeX;ROuUHXIhGQI5waF?&U0U+@>wb9fqI_ZZjoS18=|IXnk{b9?phj_3?3AYUcAC{MA7*Wgm`@A3NZMx{0f)+WJxb zbcP%1Q&O`F@Z%)U%NLTeyYY+melH&Z?acv@a#AMu_AAFrlO{oe;jKz>L-v>a6|&MV zeI!KcCx5BsVhdJ{BjUz$vL7ECFGydD&m(-@825KI5Xn`4^9HBnwy?1;i& zxfF1-iQ56rBem}Zd8-27>kr&3)5Nln4$SLK4u6mN9uJD-u!SM7WOHnYAByhzE<0cF z=ey2#yT4YMa`0)RMUTN|_V)>UAZn}5ua2PGtg4~EULfJ^*u~#@jfb{{`Wgog!E-ck zu{&SxNLACC6mC*fHcblv7Zw98G&FqyVBrZ(cAZ3#gg|G|X;!J%jF-dL$L*GPj`1Z_ z1b^O<58#+_oyY+BjNSQa4sQm#^YshJNPBq}mw zlhofaFq9yeHOYJd8^ zu22-B$~M33YhHq8=ZBp!`O|wDseT#cR21-nR_}OVunbf@Csmh;P+$*q^gxrGh;BFz zCc6@R*t{X6KkZP!i-5;h%}WG+nBdW4MPgB2l?yJe=$es&GZaiVwHJ6J;F)&>+aema z$~rcY-3}YB=S7e0-(t>ViUxF-HGlEjLZ@m4)OEZg1-xV#P6{ov+nCpn;e=o{mbCMsEhJ;ZarDvkXM(?O7MDZgoZ}IaX+Z16$>|vpS znZQy)13bOmON9qY^KjmoWkJjPIhOm8( zuW#u>Ehl*Et2RV;2dcZCV_bH5xNkPKLyr&fkJn!wp~CVNBuc}vZp!^aPs>nkx$1B| zJgmf9Lq!1H_?W9h>*zC@|9{jg84E-q)CgiNBq<6M2LlkyXSINS#jIVRc}{SYqCiY3 z0k}jdQsLx8io@pps{K-pp3) zJPTyp&aI6=+({)HiA(A@P*F8Y&f+xb%i|L^G(>=4ryt<8qRUV7`F}wIvqE9(pz%va z)A`u29*!d2oq@Ye*J5jCDBQTIV&#a^3tid9I-}XVt>O-gAKQZic-RP4 z{v%$odQ@Z}s_Bw~Wrn5wieu5W16%-tGiF60xSYx8GMTTpI}{jY3?OGm8a6s<512S; zC5~;h?$aGFu~%yZ$LB0K{aLo(Rb|=u)_fW%-AR)=0613Fpnn#@a_wz$lhylnQVTH+ zM(82ziTB^J@*{=nib|i;y$j2p)&%i$zNwe%*FnL|FNBw!rXeWUgLPS; zKjilFpZ)Wd2^lv}SxH;o`?Zs`XeeYSVtJ>K$$<^Pn)BEW8L36;)NPMVv)8WuvymPb zHxqTTXk}7Raer|asy|$2gRMJx4M%H$GBOAl< zMr+;XLD`{4h6;B$CU7|83?4-)$ctYNFbZsvYA0^a@F{O5{Hy)$PHe|t>MVXN89NE4 zsK_pv3PAG7)&4hLs>)j z!0K`&TfO)ZYSBKrC8@5x=Z(19E*`(l%NM?-5z3M7bc8qF`XdedL<(BZDo&zH&>ooU zHl{DNGk*;;l8kl(42l5gUFeo~4Um+Tuo=fkQf!{~NkzMndD$$>5Gv61WBCd{u$66b z={TdL!&Z1nFPw$ zRLta~il{dU)b2slm#|}+BdctI^P3p9y*J5jVSmBzaK@NEYCXNtUSfA@B;HjSZOM~L z?2PkOQi;koJr?9uoK)c_Z?)rTU;UY+iflH8{x0O-hOXL@2_9EhPJJbMQBom|U3VB24<*q2JJsrbCU=aLck!`7nKmvkl zPk(c59h0gmt2o=T`{D{{FArn`8`VPyWYss#!f95Evmz|RW#(wH6%R|}7IS?;bK$P6vQ3izJ|OWL zB{@r3!^LI|4e}e%xXa_Y7FSw8$DPDmBYz=~huZ=YN=ArexNrMI0nDw9rBZVOJ6) zn~h&$PS9Yc6eia(5yj3PIQJa$C24{gDfdPCmg}5y%{SG68oCaQ3Y#*=l zb#4m+tJea{r>2o%Dwri@mwTJm!(Bnef}9W5kpdmcs#)Z$7G%x;#O-aDOB9mMQlsc5 z;kC~nvG=BFaV{VKkWX&2Yc|9YOK&Uq<@nt;dT;G}oORul73q%rEz~O8Vt?t?HViCd z7YT}PYj%s#8MHGsyH^DV6%K>LW=5fGyR~y6_%Tf|d`V&WCQI_1`FCwN7%Z>1XN;Dq zC6L;;3Fyxy)!`5mz+g$I$(r2S;QyMIn|>v6y-%aMNq^phxAqg}Fkw6v zbvv)7Xg_rytFl-ntEadH|<>M+#S2W|T__RQ}8crKf zvs3bFRMQ{%COJfcQzRpt2+z`paz=tP8cljJ&Kdj4$A;#`6(wx4Bww}HQk#@luzhC1 zkqC}Nuu?2F&`qGk@L2@tgd7`M6|DGw;UZe)RL2cmWNDVA;QP$NBXL-%3!jlpixpfr zK`Xj0*eG?5=bnernyw3*kw%A94R9V0P68Ed4?H+J9t>iSMfU1eIZQde(hyT+@Iv!$ z`7*3cRw<#$pPTcImiezda-cvYB1ekQ9!u3ckqWNa&XaN^ax9IO-Zkw-*-QbZj! zrPR35G&mmYH26uLuJ_Ppz<<&Iwv;Y{SF$aurtI_6ss>xUG_%FV_2;O{?N+RPEKp&M zR;JqUv4)3aQyVfFBVoDfnlydSBKlsGFVPECibQYav5M}_FCHTHAleIr!RecZqzZvQ z<)rC*Fe|`XF{~9}t$=C;SS!GPS_v#GE&Y%&OnXcAxxtr=WZ2D*AdK}5Y6YOwvxN}W z%c!lS%yYVkriMPXdjnSE(kYL&M*T8@4My^ zul1KN#kO^SKY_5wBmL2 zTU3xE_`9Hk;gYtPR5#i->{))GW5X$MD|~gV$!e3-#3X6TAheEXKm-9ZLbCK{BQDVJ z(_wIbCz}=$m(&hznt{(@v9wxc`(!uk>Y<<^^)>s7_}8y9;7q3#+`Y=@u`)0gasf-N zUAbnf*4rk@4p7_^?8aB-3W?P04=#tgk&EGRe##m^xSId@AODNe{pWxDZ?|L{Nm%Sy z8>CSVb8;*1M!k(7p$QRXY{j1GG%1sa9Wnvale--+0-nl~>>Wja(ykKq%UsvE*TEh{ zB>PoM8GY+1qjQOJsHM7{@0CzA^DTs!q!%tywPiirEPWpc?&N_(F*!+|$qJU4gi~Mv zNF!#Jo&`p*$RxIs(7>S>k<;OFa1FRbrOMnriBm;3C&S#3F-?qhmO?X^_DijvxlhpP zXllHMis}ZDV|65dTJZwmXrRa+mw}DZVl(4x)itH!ebsW?yd)K&1FOwb))2my05#4f zsc9EmE#fCAd%AUtd5L4T3s?mmtyav6tA(bkMADSim?_iLS^*=FolnHH{H(LvIOgIj zdH$Z(1fvDk9_J^%0ezLMp0lbs(%eQ0`m2-%N!{Ee^;lVd?DDr&MW71;YnKK+7NcJ< z{wRqAe@^rIh3E|oxciQZkhCY8 zr-b&J_VucHs>@Sg#8 zSWIe0QGde}hvZZz!FGsZJTfGU#u-VkJj^4JJ-mbtYk*Z&e_z1`xO-(~q1M_21rt$! z`p^mckXoabf36AH9*x(YX>Y1YozbPIAu1vR`(E8)BIj-X%dn^Wa4 zt8PTYjybShZw3`+c$nzF{Mrw1e&Z2bB;Rc}TO36cE01FH-Dn)!F}IucG^1q+ZAv{4 z&o2s`Q?0*3aJYYKKq|s5^G~9zB{}S+XrA_TNV8>o8EFmQ9voR6-e`^}6q^O43k#jW zL<#N~BEbN2>n%gX8Kiv#o$_oqPFd|}4_MV8NZL)QFJWVj95TXJcX=gPMW43=5nDCc z(O(^oOedZ}fYei?@mZ?YO?0?^tVz|m+%)esFWCuG(vp8qgY|r{1X*>dIX5Myw#a4n zPG^M%XQH`j0cJRQ5bZ@Pqe~J9rT^rK@@|I{yFNP`LP2)X?yT6_1On*`hHgDxeAhEWm#1CD9j>aiMz;QFxiuD^u`Jj< zbYQFhh)vdxc75cB2sE<)!a%xR{&tcFOtRd+aJf#rHXKPEWJor?%b5+#_u|P5mIh~X zDx7JR^2U^NFdIbDDkz#Hv7<$)mI|@w0CmI^x9kScbSY9v8Jnj z`l-_MyOX#hP+l>qvyDNdmKDY?>_CAS7mJz6-yo7Jaja>37;pCJY%2A!e4}Bvi1QU9 zqEAOUik;&8l%&rI+tmusv}U_wrSRPlUd47Ug5iuuA3P&Ah3z4_<vNG@N7>S)_ z`gZ<7lNKdm0^Vkmc_lXpA$hJ5X$|o}lei^61&SX=wNU?eA?ucl|;2QnxLH=h?84W1AizNi*P|LQgumGJVL&9kR zc}kD-8^C|R{QB)&9?!CX0nbn5IfG4Ta6b5j%6NpU=viG3w>rW$Emn7{ku3o}2&#xt zMU09y1cHh+5hKl1#a$LYTiiPc_G9&wQGXGzHxd@o$J@78Y*#iUG8Cm_Q6FhV{hZZw z%xAaXN75#QI0pY~^!wF^`}6yY->)uc1^w;({_6b8)#qFK`~2&d&#Mo&7oRV$Za+mI zKi@}RZy&zieYpR2_3+^`y8L`|esvpt`fyA6udn{b*yfy+)vW>prRs8(H7jUH>yuiVSM>mlbtFyC=<;LPi3B>n>9rD;&L}~6T{!QtddPmSQ!8LoYh%VRtt?B zqJzGl*^~P!jsanliYtv3ImuB?cusO-F||;=K0b0Hdk0G_50fk`D-4fT^N1?-b=c>h zA61iMELQ9giU zHga(-YpuAU!u7;)a|gfRXAMth#q5QZSaChLgW+mSXAU=G(xfQH$qHc$Y+;Zizh9o; z-bStrIF_JgPL-OLdD4y@$!50uBu9d#MIv6WZN6NBWT_X=3G?UO$Q=_b9a#|zU8GmfDyk6IboJ}2nH1U=s4LdIHVWb+% zn$k|`A7vSfA!naZn6NmL-!n2BY?RM}O3$2{RQ_wL) z1~OJ94H0YJHS?4D7FH@4jcaF`Ws2dfhPMu597D8f?LnjHWO(kc>lDgNu|Q*SsqbQS z1&$$lfufdUcN-LcZR8Ra?FMT$DB8#cD%y3@EkJZ}3-q(?kXpg$V;5-2I|9qq6gY-x z4@Frc@2?+cpAYrWrfZ}8HR>VMF*?m`SBf3ihk;bAxLN{t)DQjdDYTe}sMnivf3R7K z9y83IjvW!H*NSJvPZpqCniZ$9TgskO(;7L9ISLwaw|Sg@R6E*1NK^LX?2uG7EtraX z*IjH~A{GM#*Gud#|2iaD70@wU61(934kUT<;hN9KRPf+syeVM_7$>`sxno}5BJN-g zf&B22)kMfTjaSA(mqUc`&-PfvbWl{66&;UNdw6~&ynkkeE{6!=ADEG6*FABzxTbP` zVI*-HzIf?>r6Xd6a2@}GQ(r~a1a}pGZ-h#_0%;3)gS&#IL~-25Rryk%)-OZRSU^EN zMul+BGsv*>5447J+U;@AU#B{Ipwh)dRft$8cilT>zB$`bGNTbVsE41ksyQZkMYd^8 znlqBn;1lQ2Z(N0Yvb_WqK2Q~wWm6^K`*`x5GOW^n3n;nP4&Bb-irI?s$kg3)mcYiz zb$_J?qsU+u45c^`tEPJpOlE(>+swtdQ3tfzV$u-9>iNki@7W99Gu(an|{{KG~MQ^>o_uLcUi5I$D01!G*&yp6xyrN&jtDy#fC653%Z5k zeB+~kL-pcm?Lg8pc6={9l5JU$lel3ZY9W6xM>N%GywFt95@vk+;eMiu>@03=UF?E^ zsD=Dt9TAt$wCKTgbNsV{XN*tioq72oK=J*Ul1)jW2?^}4Ud|Hc0QSS?bAb;p-a$v1 zt&WEFhf*#k7{0%3PKPdbj^e1OH7CB=FwnJs*rv4D@?OpoW_8Ygm?82sA@2u*;Ya&K74}@ zvE@PR#dgb*mazx$LdnWgXb%V}{;UBkui#z2LsHS))fjJ1TJD0eI0OuQr&!#*zSt9g z^S7_tGUD6pbeZd7>rL0WxP zevo8h0iDQ4Yaa3*u4+XiSWHk=Hd`A!aIdM7g9ImP@mX6l3-M{6S(Ze zW$$_*6%WeSQ~Q0&_J_71^cy}={Lx{HKEut;{q8sB_OFi1J4Qa({BACJXND`f$1Fw@tM)VD8_7C0UObCWeVDGC(Gj`XNz z1x)LYlV~_G0r!)dI3oni|9dQx!8jTcu>gDsI}7QQThd(6klzYO?K_k0IB@|JlV3S& z0V9*kIVb_Tlk_=u0@HJoY&tXpRKA>(qdK7+DJfF82{C!BX$!t-y1OexgtOv0A?HF! zcl*TelVCd{1lu3w$I|K?Pa3|LStTkq+AKHUOqbvp_EtSABW!{QlR-nFGzNq6lNv?QP`Bps`6cbqkY(mg?&u%qwaZii|4< z1jQzhRz%7vX&WO%`hQ?3-$H34jmlb2o_9Lg&YYWP&1YxKHM7$5Gv!vZed^bn4B}qeG+0UZrYB|xja#D9zOgp3Uk&5}i#oXSuHCWphiV!fuIl-y- zjpY(k_j@T~NO#$F!tp$%?1e7?v-OWakJ}A)`|sd`IR_QVo_|;@x6Zp|Dh9NjGv$_w zIiFpGb;s~PGlM(&Gw)|hy)u!jv#md;einrQ_3#^Y5BMaHw^g~1w^>d!zT?C| z34}{sH!E6Y=?6yqky3ul=2r7?aBUrUY_h!O$+RO)3^y3XX|`X>A71XU0`}svH4W|s zEsC|as@1=~pno+`z-ueR$*T*_yb5C#{Y4u?RIKLCyROeGHxvuB7~<$-m0ez)o}(&! ziK=5LBf(OIhm|Rk>LXAG(xv81Yd0j5W`G^Zy-ZWoT}zBoL0S`f8NzF)C3*`4Rk&6w z>w&84Al2qJ<`J~zEzHjeu^$Xe#Pd7JdIHVxbSQE<;(w*_ZUg6$T2_d-s?Ws3L*?*a&UXYX5{U^}V%)d$@|NC}z%{8)}%T68y9oh~h8_2wLQAvvtO z$SOLuz`wr~dWT};pJ?8I--?y z{qUh^8w2Ue>}UaRP!|NK#0=(G)m4@lVOA4hp-Y_mWMHw)@jpD$J;)OWPs%-O=lco)MI_rY)Jt2t`M*V(v*@;^U5Z)u< z%8Ax6mqKF&bOKL>D^7X~UCNdn5|XnRP27}+Y|{p2``sA}pnCiWlBr~&!o({o^a}(~ zOaEq>rHnAA@=?eFXf)ET7+S-f%y%J+q<`63w}NSphwr|)^V^Ig>!!*AJ5Z88HJ!;v zM|=k-o~S4rP4_O<*2wSS+!K{`vk$zBy*+do|MFjFwB8gHbyB1+v~R$|LAf8N$h*B7 z!V3>V4dw}O$>034t_|?}dG>_0gN-8=RT3@&-qGVHtTL=TvZ&^8l7Qp%*S8e58Wtsa zPVzPV5aDf50Jjb#{_DS83RCY|lfgh21>Rs|+dGrpKqCRilL0{#M-EPW#|*h(@y`?b zB>v?mmrn0ae$+HVLpq_e-Z2w?7%6}cc@jS!Yf_!wd0%T%MMyX&bJII!wy!m*P_x@GmwEiXL;wk8eayj;`G+~Rd#lm8s}+_&e%FZ2Vr+i-gG`_bW#98u?eIVk@iTC*$62CaZHc_ zEl3uruiS};VTu~uksQIq(Y^yb^TxmqqzGmqrMt*l;KO6I*y-S>&xuc<7Y8I=M06|<{cl-FT9c;Mohao5qr;>uaPGWj z2PYl>WMQV$qrDgt0zFa`W{ioj^K*(~4Q1-8Pl>U)_`gm8B$0FrSCvnQWgHNDhP> z)P2Qq&#CLvIZl#^j%epBCIs%$I3&$eoKs;eDBHs~e0n%z1``y%+TYRdYru5oKr)88 z2!6!P@sJZn%WVr_rCJysl!IU7oCLWjPRTYo=G;Ynk$@->sJBqRc^6x&l+jLuAGV*J6M`9tUgV*Sg ze_OI+7c8*-^7Td!2O2&cA3v6VqbV^nv6uE+(rli(4p_Zh|1qE&xt9%3NgZ!9+BZz{ zynG>P3~$yHb*7$rwKqqD?agUtB%k<9lP2Lk`C?D@u*O;GuBOu%CgHIjc5y@YmmQuH zwX*@+VmnhR?SkO4CM_$S1*pMaKwGd_!JKb2xT_?NUOO3R$$m5fw>U-X%3?a+I-dsdUsIv zr2aZ@p;L^cG{=v3f^C%y?azvTVIzY|vkKk94%n9w z&AR%y-KrNJml-Iq7w{sOS&@DDOt1|Us9szMM%4+vNet72$I+ZaRW>EZk9m>e=S(n#>Z}4E$%A zG;0MqVZ*5kFVQ1^2wmKC{>rEueEX}#xD~-3%CKbi*M_sJ!qxhir}jG8HWPbsTVxC~ zd_CQCs;`SvOur3ejHmHLl*KIIPz4hXhK9o^TB%e7-uZ=>e~)>?hLrKPD)-P^BOBUS zzImwcA!GqBp5cbVxWwFPNwwj?Gfn&sWL=By!{lRQbu40k8UUx{X=?!8hG$YWG)run zvn?GeHjC{vOYmhxuoniy?48d&p-pg&ZR{A{iwnWb;U~70BXl1ot&MZdWM64p8>5HO zcZ(bpnwC){6=#6031_gA!D7eKTfKQA<-`@XnmEZUC!)~XQcWCwS}0)^N1aW`fP?aAk9pl&_b#hTF? zu$stMPuSr~*->2IgIQC0s!bCoE3#sTp>>C4OX#R`cT+Og-Ei5QjpiL9SL|YP#heW} ziWOqey~~>T4d0>xx<4zXrF9+e2ps~Z&%w-pl|(G$$20sus%^|WqJiv|B`jmdvl^wd ztWnzBX|`cTvbQ2IKh6ADKK>EH;*o@9>^N4Z**-|c#c_R@aO@(_Nq$Tk62P<$+LI5- zz{PWDDmC>}YFaKd({Lx;)H*xNzqx?m_#rE{q>A6-=R>wB0@y#qU&4%I?;ef{GXJrE zYaP_vxEI(wm9Z26$#$ISHI>p9vu$Qp97|y;M*Pl}jk**Xj|42@#xeO&k@QTr9e07X zJ+?0+j*YZ)lEMrXY=(g~qCn=(ib24@aRSDuT%ojakZ$0!HW1eK{>bP6YMrDhTk1PI zBvnl-dlu?vnhDgI!^s@>G{p!-^Wcbo*%TF^6Sb`gS<%ogf3hL$_W0Mgv=5;H^xMmh zDe6@l^6mV?zy8~00?~E2{-+ro6QH>l7kmY_@)XJ%UHCDH9 zW*#@A(xfQhU?%)sV8j5%g2qwF?3vwg2Y;6+DZ^OMxG71hJuuQg3sjUqj1+F_J*7u( z8T+w7#SxGdm7A)(uS0qLds(wUR#a}Pil(CLaPlT?$+FHoE^mR9i#rxfh;|{)N?9mw zJ2&MGh3Ofz-8DAW*loG-MZ$D{3qmugH^9_j+HosPo@|pB8@4TsQL$czc4q`lx#xSPOJgLlFmv+Iwllz)Vd?{YV}6^7Bp6Bu|M%>YhRin0F#NKVWRY+xo=L#GNONmu-Hh41x@`M@)H2LtXwQRlg3)ns z7C6_8b#PMq$HGi)W-ms`=_g)rqqL`QmWGwOK%_DA`MV?hgMPwKBX?JXfG4M>)BzUoRPg%%VP zy+@F)4BnRTSJ{xNpB~MDcB$$Ws_K=4>Px*&2XL6W1Dp%7ul99__PBO$k87m%n61d^ zj<-7*(xW-hx~Sf*vGpbxX5)<>AC)5jr9mCzUP+L`(|sy4Ds<6+X{s0zpi*S{!fyj5I_o0JH9We`W< z5vDVTzcJ9oXZJ%rw$IlgSu_Le$L4R_(gDbve)=JZk;BE9X&pe}{F|jmMe(Jes)otT zf@RNoQyz5e;Kej`(PWyP z&YBs2HCF%!%i%37V1&9IERVZTmTi2kHNpBY`B>E^W*d+m_xrcSMz!w0qXE}6;F6(w z@SvH5SX1jYwyvk0VH|wHA9&+?8^w%6mipQ|4`xqlr#EBeQ(<5>>)Jk(jhM15eY$cI zOUhD(H4Xq=aDt6hmn?1+VqxqtDt^H@kW<$%#(P>c*{O8~e2+=Ee0qAT-NwXJ)M{(0W7q|(#qY*+1Pb4n+X+~F zb}C6=P$AzfvEHCQGt>JvqbgwN<$(PPM|3C@v5E@2YeoL|c4|VSje`+_`{jBocCTE5 zi>9z;kJfju+8>vzFvrEAGGp=oG@5Ns`0zskm`v9*-pmO|LsWv^nrKNkm?DlO=V`c% zR+%up;C6wo(cE`M_L?}Mk1zXz79okGgiODHdlQ{~^G>85g5G}O^=kgbtwx1pWPya+ zX%zWY;fAw&;zIPRL4tO_KYn~r*ED-dnyn6=gSbwwdHreuVQuxNfN?&wT2Bop8=D6+ zg`@vfHe#7h`6di1jDD(2>_N}z<@g7hT>l3-7O4la*w9j08gc*)hn7t$J{ ziiluYmtbD>;oU%Pk_8bLV4S&*&46M6d5o5))W}y^%0%YtthMc zJARk|HlgDd+FHJUOK3M4rQgd(c6Q{fRbpH^ox_qUB21BsZ|?ws&RL-JFiqIc z3B6%8q3x>@$Piv6#I8y!LtGp|VT*|QtrT-c&?up|MahCs;8vn>vd=EQ{HuTI4()#CLFDt+9A)s8`vexnFdHJq+Bwz$|`DT%cuu?gY6Q#9W^KH6&5uRg-PoZ|1uz7~uwbRI(KEi{NeD zsEDWiit%9Sni$%ZYbzYCmjDgk?&KW8y*+YRKs$^Kw-)BM`sPdrf1t2V%+IBN(%M-XXv@a9<&BS{K9ru?gPG(n2CLBZW`=A!1lR9^Rc>eQZ`&zQq{|H7~4UAg4v(7jdnnX z(L||k_w1B}NAg#CdE^N1sq_ro;00K)pdw{?5*>^>3s{fTKjf>)#>1uV@2x z$s;~W1y6MiH=VyUH5N?xMjN>LHhNOHp+rG0-%jAEOiH=z!`E=ggN}|m?t(wM5x1Pz z>U&Ae{t9OU*4WQmBV5R&gMuT!u)^&n#ec|HK|<>Eq;B{cRa-bOLmDzM`Or&JVn zP3^#-h1AQ;ae)KL%Tn4oHcN6rsj|X18t_ugmWuPH{*X(H=|;RugAtG96{hj1K^5R9 z@-wpx-)WqX+AU!bm*~=@3$VegZ^#Ia8vQ|uqm1RHS)+=9`n&1{VzT>%!{?c#1oP?Y zRpSzw{hn@N7snb@1Q%IPtit%kcpEp})Wiab(|v3CK*10%Ml9!;)hm$~=CyVNWho0l zU=PY*PR;j^gK7CrK*$mciHj7&L)K2dpMGg2GV)Pqq%eQGC|Nt56U>+RsDg15|4^#( zBP5IBpjL~C>kzfSd7XZx9f>kp))7>g-1GXDY@<1!LBQ9#4c>25lLXzXPrKdJlzg8zEs{S;N z0Z8WLa-K=U2u_`F;;lL`jA}d*Ky5OnkigbSzl;>XPik3t$0?>kjpqs`T$AprOChqN zaW_0`lC9iCeDV*IaApSPL0-Q@Re;^RPQ1`o*dw=SnE0DFU`pjHzY&N{Tizv1pwASl z%eioPqAsRA`>S&d_DY6RT|>LR#NxET*h1OgTFH%W=R*J7&~(GYD#_&@+0s_))bR;xGIi@aqy1E~wt>Woo#vvSb~4T0+K-VRo7kOMj95GzM-0YfwLe5a_`5al})q#FSdSUr=dF^jDyIuC0gxu?5rls`HReY0vpkRf+ zHKiU(9CaKC-t}oe)8e9R#_t?WDHB}JRdKW@y}eW4Pj<T< zZv{vv!RJ7n!Vd|;>W9@kAn6#6nl~xvnXh$54?D6&-Fev5dFr@pZ7efPTI64vs+Nsq znto8Vi6gYha2evfHNDS`?g5qbYju~}>{zqS(rrbG&wgt14yyJ~m&Q=GP@qz~1x=48 z)H6XhX=v^%i9zl2c^1NOKMkGrFXwTMdT7u1z%liv7!H^dz0j?@yOY)KO>z z^rl$+)n!ZB?M0zIGXZ$CSUj<+Dg~sw3!DCND+~NK+570iAEK=dnDR~9ZHT0!%WMnu zNU{F)Sl-TOw|WcUjDtT{(X@Pd;Y6tWi)(VPramGBH0Lygv8IA1g5Tg?S+pmFhSZeh4nFm89%GXW(OHCYZdL4q*c>(6q|;>8fGifW&a*G&S< zYYp7o6trY|3B!ML?11fzR2)>piE~7JGyKJ6F4zaJbbhn1^9tE<|6}{vHOLG4R9nT^ zPFYv}-5u4JjmC^+GFZ*3KI9&zG_)F30(Pt=6C5w%`_1^xD5N}V-AEnVSp1Kt8{j*P zddJ~L=9T1b!l7|xYKuWwvRZh##F@Wo#@h&=N~8amDm1zK|DX!N|BDs6gk=Ue%|_}4 zw2$Rc((gdi3nH31*(?_3qr4WUkm+g&Le{a>(O>9;3hDv0&Va#>0q^^*pzK>ThPbY8 zPMRyjD+;>470-C~z0c!Qjb_D*B96fT`$O~nNMysNKI&g5M95A*6C=8W`p8t+t<1-g~)EZ9j8vi}GXaghTlDIk4prD*WKf;4LQhvak>Fx6*F-H}6C*V%+9 zxym+2lDg9fN~|It8KrbZY8r0IpJn1hgSQ~X+mx;hoDx?*eZ=8|%RH!rW#B;}^{(b` z`!8426(aWX>aw)(4mYywNUEj!yc?&#(ByXi^8!>hFGBv2U!+w|jE)Z=Xly*bU^{0> zQw<5T)AzTU=6+nv$v`NX*Ls z@l!J{Py>fuaGW_%ACJhNw&~BZI;L6mIBeII{~_KaF7niRz2#i}F7?HX#BOW$g||88 z)ZEEUDr%;hLS2N8m-lzSN!07NSYX$|aj(~W#ov|Y^cd7pW_8#!hiuBN3FYNlZri!v z$hIG5pd)`+^zmJV4(tWhtOvJQdGs1BCB>xY4$z)N0j7ZEvxTm3x9>#t8ihNnEXjdt zJ}|d}_?8axRdlXUX;r5mL3>PH>OgXeq?Ct?5)Oll-&qN!|oo zosq*RHR^u8crVxI^MEju>NR>%_Noq)6~zAD8})X1#t=X+wA#}&)-aXwBL{bwRp9lF zfKF2s&3!}==`-F2&N9(jb1AJf&R~m~A{UH%e)D#DhHL!k+TucgLHG{LA=mjyZOaqR zg2BkdV9g)nv7RR1_N?#Y!4HEFp>}d^y^Dk6KMHO?(LfY-Bn2Dn#8&?-nLqrJ%lgCBw>6+H&It&_lBYl)~NJlB~u{+n=jHGLQr+eKmk1dhnA+XVQP=$^R zI;wJv_;M6*u2F8fx8=VljSntl2M{ONd^#^(_YJiHS7bSSXDIwK;mL+sMwTeD&v-r~ zy_=8-Jw!(7OmLlSJI+}hvkqd=)g^BXpZ8eI7Cm!v&5cb>cMl8IF`PE;S5Y#Gy1IE3lk zMmdi*S2FQsh;!kC-4lO~;4f8AMGFVBjqP3^fi(C9*haW4bzK5JQBBxLY%qUp-_6@d z7H2jENY{`pELR6ss9mBF(ZAbF@Rr>s6>fURt79p4E?^36)ZbeKNHbt#l2GP1E|>Uz zTk;B3a-_w!W>Kd}kf77bPS{uGZ;Qi2v|jxYtREcJf}ZMufz3m(#Tt7O(N}Y*`YOm2 z$5>?gVP(&fLr;Y_=_=gpr$VhdTqD2YFJ^y^tj`M$u2C_Iepu8Ap6)-7@`*WkA08Fu zpeP*?gC=QiXy9nm_v~tl_Kqr_TG*cICwkq=xSBvi?!7tPnpUwZS8nKuj#kJy5 zUiSu7U^1DuCulfhS0T=ND0JrL_D>LObsF8?q$mj%^XI8+rn!$g)zD)9#}@yYq`jkN zQlS+s4eNx&$$v=yrM>RiS>J_Qe(^s4llr72El&r9VUYqAGaA_bSW0M9`HE)6zJ+MH z;5xu-TjuL0LLchU{^sFL`W+Q~qLo)sud7^XA+s<>>UMSXdTCwX@vp5dixQjEcJM4G zyD@>)#OvtdvXxspMg2Ee+@vhiCykV4xNbuhWKDK`uf%0pYktnIB%Xgq$}k?8P9=+x z6%6Re@6;Ea%2S6&d-#Pq3rGGt>-(-P z0n+>t3P#3csnDcV#RiT+S9zA3^G0!IG{$`F?#*tF2F_3qC(~{*>2KO*P1Me2t}Q^$ zMsYTkC^2O|lRnio;z)nzL)||1)?+@aWtFt;%0-$;Qf$>bTBALHu}@i=0#x@UF~M{AE8@31m7HK=e!l+zMkknB32`0KC)9buvG$Y1I#fiwFm z=i!K^WJwEmm&nrEEb6fvjB%dg^=&W*->AcBQ}F7KsiAC8MhVLHwp=k3G<$brE;PZY zo4J(CLPjW@G%wnc!w{h3T6pZa&<+i3F)>^~ru!Kk>X^UJR*5}19M)7W^iGn7fS#?m zN?g6$M^mr79f^Qv(nmFmh<61z4AZCC_U|*wR*0yrb8VF1opc%uY0(F^WP%qv9m|?t zKAuj85fJybp|?TO-ki_ZTh7#WkQf`Er^liXVTgOa)GZfNLQ?R9SXPNAJvQZ(k-Y9IoVL&UFO**zva9;K%6slUw5A zy$}3&?pB{vFJtM|8tKZ%PqiiwJhus+5I_w8U3S0j(l{v$RcJ(HiP`a z^Nm4%fZNkVMKG4$wkimcl=WrvuPBXx4HV{(p?t4NuRoum7beAUwJ2wA=K zBc`w5I#}!Z1$0@jx9w`Gj431J20E!(noSwv;^|t9I7&R7^?pA3)_#d-I6qukuRu`Q zlO5sW)dS-rP+$ap2?NKW1+pLbf4i(DZ2VPVF05+ST$pITT)u*-Q0f7Q z17UOEm3Jcu;oB7{l^t4Mz&*lV9k41`qc;w1kbr0%2HR3eC_lfzFF^Hp?Nd^?LF}oF zPHI!fHCBz~zFezC{vw^!L5LU|s@gy)=jJBH5zlp57kzlETQUZ;jyUlBq9`4)403ZODI5w0qv~9qGNpN% zxF7nyH&Qd4FYnK>w)==(HShzYADgzkRn`_%Z_ax!z;Be3e-lMN-laUswJ(B6mX-kJ zjK54ls!Ks1HI%QNyiO>R%_ z^6^pP^3VgR@0%FDtb>gH=>RSbUElRs_}H3*iaB1r{rPyO4<`1@x&4_RC##Dj-8v0Q fy#WN$P%h(a6dXPZj>XB*-#p&sJW>}Fb65Wtc>0B% diff --git a/docs/build/html/py-modindex.html b/docs/build/html/py-modindex.html index d5343d43a..da4bbba3b 100644 --- a/docs/build/html/py-modindex.html +++ b/docs/build/html/py-modindex.html @@ -1,7 +1,7 @@ - + @@ -54,7 +54,6 @@ - @@ -310,13 +309,6 @@ Causal IRL - -
      • - - - Invariant Causal Mechanisms through Distribution Matching - -
      @@ -567,11 +559,6 @@

      Python Module Index

          domainlab.arg_parser - - -     - domainlab.cli -     @@ -827,16 +814,6 @@

      Python Module Index

          domainlab.exp.exp_cuda_seed - - -     - domainlab.exp.exp_main - - - -     - domainlab.exp.exp_utils -     @@ -847,16 +824,6 @@

      Python Module Index

          domainlab.exp_protocol.aggregate_results - - -     - domainlab.exp_protocol.run_experiment - - - -     - domainlab.mk_exp -     @@ -1117,7 +1084,7 @@

      Python Module Index

      Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/readme_link.html b/docs/build/html/readme_link.html index 14b2fa138..061936023 100644 --- a/docs/build/html/readme_link.html +++ b/docs/build/html/readme_link.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -310,13 +309,6 @@ Causal IRL - -
    • - - - Invariant Causal Mechanisms through Distribution Matching - -
    @@ -378,20 +370,20 @@
    -

    Introduction

    +

    Introduction

    -

    DomainLab: modular python package for training domain invariant neural networks

    +

    DomainLab: modular python package for training domain invariant neural networks

    GH Actions CI codecov Codacy Badge Documentation pages-build-deployment

    -

    Distribution shifts, domain generalization and DomainLab

    +

    Distribution shifts, domain generalization and DomainLab

    Neural networks trained using data from a specific distribution (domain) usually fail to generalize to novel distributions (domains). Domain generalization aims at learning domain invariant features by utilizing data from multiple domains (data sites, cohorts, batches, vendors) so the learned feature can be generalized to new unseen domains (distributions).

    DomainLab is a software platform with state-of-the-art domain generalization algorithms implemented and designed by maximal decoupling of different software components thus enhancing maximal code reuse.

    -

    DomainLab

    +

    DomainLab

    DomainLab decouples the following concepts or objects:

    • task \(M\): In DomainLab, a task is a container for datasets from different domains. (e.g. from distribution \(D_1\) and \(D_2\)). The task offers a static protocol to evaluate the generalization performance of a neural network: which dataset(s) is used for training, and which dataset(s) is used for testing.

    • @@ -427,21 +419,21 @@

      DomainLab -

      Getting started

      +

      Getting started

      -

      Installation

      +

      Installation

      For the development version in Github, see Installation and Dependencies handling

      We also offer a PyPI version here https://pypi.org/project/domainlab/ which one could install via pip install domainlab and it is recommended to create a virtual environment for it.

      -

      Task specification

      +

      Task specification

      We offer various ways for the user to specify a scenario to evaluate the generalization performance via training on a limited number of datasets. See detail in Task Specification

      -

      Example and usage

      +

      Example and usage

      -
      Available arguments for commandline
      +
      Available arguments for commandline

      The following command tells which arguments/hyperparameters/multipliers are available to be set by the user and which model they are associated with

      python main_out.py --help
       
      @@ -452,7 +444,7 @@
      Available arguments for commandline
      -
      Command line configuration file
      +
      Command line configuration file

      domainlab -c ./examples/conf/vlcs_diva_mldg_dial.yaml (if you install via pip)

      or if you clone this the code repository for DomainLab

      python main_out.py -c ./examples/conf/vlcs_diva_mldg_dial.yaml

      @@ -472,12 +464,12 @@
      Command line configuration fileSee details in Command line usage

      -
      or Programm against DomainLab API
      +
      or Programm against DomainLab API

      See example here: Transformer as feature extractor, decorate JIGEN with DANN, training using MLDG decorated by DIAL

      -

      Benchmark different methods

      +

      Benchmark different methods

      DomainLab provides a powerful benchmark functionality. To benchmark several algorithms(combination of neural networks, models, trainers and associated hyperparameters), a single line command along with a benchmark configuration files is sufficient. See details in benchmarks documentation and tutorial

      One could simply run @@ -493,7 +485,7 @@

      Benchmark different methods -

      Citation

      +

      Citation

      Source: https://arxiv.org/pdf/2403.14356.pdf

      @misc{sun2024domainlab,
         title={DomainLab: A modular Python package for domain generalization in deep learning},
      @@ -531,7 +523,7 @@ 

      CitationSphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/requirements.html b/docs/build/html/requirements.html index 98334bc5c..2c9bf40e2 100644 --- a/docs/build/html/requirements.html +++ b/docs/build/html/requirements.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
    • - - - Invariant Causal Mechanisms through Distribution Matching - -
    @@ -347,7 +339,7 @@ sphinx_material commonmark

    -

    recommonmark==0.5.0.dev0

    +

    recommonmark==0.5.0.dev0

    git+https://github.com/rtfd/recommonmark

    @@ -374,7 +366,7 @@

    recommonmark==0.5.0.dev0 Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/search.html b/docs/build/html/search.html index 5a2256db5..acb1c74b5 100644 --- a/docs/build/html/search.html +++ b/docs/build/html/search.html @@ -1,7 +1,7 @@ - + @@ -55,7 +55,6 @@ - @@ -313,13 +312,6 @@ Causal IRL -

  • -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -378,7 +370,7 @@

    Search

    Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx diff --git a/docs/build/html/searchindex.js b/docs/build/html/searchindex.js index cc844d206..63392d5bd 100644 --- a/docs/build/html/searchindex.js +++ b/docs/build/html/searchindex.js @@ -1 +1 @@ -Search.setIndex({"docnames": ["docDIAL", "docFishr", "docHDUVA", "docIRM", "docJiGen", "docMA", "docMatchDG", "doc_MNIST_classification", "doc_benchmark", "doc_benchmark_further_explanation", "doc_benchmark_pacs", "doc_benchmark_yaml", "doc_coral", "doc_custom_nn", "doc_dann", "doc_diva", "doc_extend_contribute", "doc_install", "doc_irl", "doc_mldg", "doc_model", "doc_output", "doc_tasks", "doc_trainer", "doc_usage_cmd", "domainlab", "domainlab.algos", "domainlab.algos.msels", "domainlab.algos.observers", "domainlab.algos.trainers", "domainlab.algos.trainers.compos", "domainlab.compos", "domainlab.compos.nn_zoo", "domainlab.compos.pcr", "domainlab.compos.vae", "domainlab.compos.vae.compos", "domainlab.dsets", "domainlab.exp", "domainlab.exp_protocol", "domainlab.models", "domainlab.tasks", "domainlab.utils", "index", "modules", "readme_link", "requirements", "tips"], "filenames": ["docDIAL.md", "docFishr.md", "docHDUVA.md", "docIRM.md", "docJiGen.md", "docMA.md", "docMatchDG.md", "doc_MNIST_classification.md", "doc_benchmark.md", "doc_benchmark_further_explanation.md", "doc_benchmark_pacs.md", "doc_benchmark_yaml.md", "doc_coral.md", "doc_custom_nn.md", "doc_dann.md", "doc_diva.md", "doc_extend_contribute.md", "doc_install.md", "doc_irl.md", "doc_mldg.md", "doc_model.md", "doc_output.md", "doc_tasks.md", "doc_trainer.md", "doc_usage_cmd.md", "domainlab.rst", "domainlab.algos.rst", "domainlab.algos.msels.rst", "domainlab.algos.observers.rst", "domainlab.algos.trainers.rst", "domainlab.algos.trainers.compos.rst", "domainlab.compos.rst", "domainlab.compos.nn_zoo.rst", "domainlab.compos.pcr.rst", "domainlab.compos.vae.rst", "domainlab.compos.vae.compos.rst", "domainlab.dsets.rst", "domainlab.exp.rst", "domainlab.exp_protocol.rst", "domainlab.models.rst", "domainlab.tasks.rst", "domainlab.utils.rst", "index.rst", "modules.rst", "readme_link.rst", "requirements.txt", "tips.txt"], "titles": ["Trainer DIAL", "Trainer Fishr", "Model HDUVA", "Trainer IRM (Invariant Risk Minimization)", "Model JiGen", "Simple Moving Average", "Trainer MatchDG", "colored MNIST classification", "Benchmarking with DomainLab", "Further explanations to Benchmark Setup", "Let DomainLab know where your PACS data were located", "Benchmark yaml files", "Deep CORAL", "Specify neural network in command line", "Model DANN", "Model DIVA", "Software Architecture and Design", "Installation of DomainLab", "Causal IRL", "Trainer MLDG", "Model Specification", "Output structure", "Task Specification", "Model Specification", "DomainLab Usage Guide", "domainlab package", "domainlab.algos package", "domainlab.algos.msels package", "domainlab.algos.observers package", "domainlab.algos.trainers package", "domainlab.algos.trainers.compos package", "domainlab.compos package", "domainlab.compos.nn_zoo package", "domainlab.compos.pcr package", "domainlab.compos.vae package", "domainlab.compos.vae.compos package", "domainlab.dsets package", "domainlab.exp package", "domainlab.exp_protocol package", "domainlab.models package", "domainlab.tasks package", "domainlab.utils package", "Welcome to domainlab\u2019s documentation!", "domainlab", "Introduction", "recommonmark==0.5.0.dev0", "<no title>"], "terms": {"The": [0, 1, 2, 3, 4, 6, 7, 8, 9, 11, 13, 14, 15, 20, 21, 22, 25, 26, 29, 32, 34, 39, 40, 41, 44], "algorithm": [0, 6, 7, 11, 20, 21, 24, 26, 29, 31, 37, 38, 39, 44], "introduc": [0, 2, 6, 23, 39], "http": [0, 1, 2, 6, 12, 13, 15, 18, 32, 35, 36, 37, 39, 44, 45], "arxiv": [0, 1, 2, 6, 12, 15, 18, 39, 44], "org": [0, 1, 2, 6, 12, 15, 18, 32, 36, 37, 44], "pdf": [0, 2, 6, 12, 15, 18, 44], "2104": 0, "00322": 0, "us": [0, 1, 3, 4, 7, 8, 9, 11, 15, 17, 20, 23, 24, 25, 27, 28, 29, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 44], "tackl": 0, "task": [0, 1, 2, 3, 4, 5, 6, 7, 9, 12, 13, 15, 16, 18, 19, 20, 21, 25, 26, 28, 29, 30, 33, 36, 37, 38, 39, 41, 43], "therefor": [0, 6, 11, 15], "sourc": [0, 1, 2, 6, 18, 20, 22, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 44], "natur": 0, "dataset": [0, 10, 11, 24, 25, 29, 30, 36, 39, 40, 41, 44], "while": [0, 6, 7, 23, 32, 34], "target": [0, 29, 39, 40], "attack": 0, "imag": [0, 1, 2, 4, 6, 8, 21, 22, 24, 29, 31, 32, 35, 36, 39, 40, 41, 42], "demonstr": 0, "figur": [0, 1, 2, 6, 7, 37], "1": [0, 1, 2, 3, 4, 5, 6, 7, 8, 11, 12, 14, 15, 19, 24, 26, 29, 31, 32, 33, 34, 35, 36, 38, 39, 40, 41, 44], "find": [0, 1, 2, 6, 8, 22, 24, 29], "an": [0, 1, 2, 3, 4, 6, 7, 8, 9, 15, 20, 21, 22, 24, 25, 26, 29, 32, 33, 36, 39, 40, 41, 44], "x": [0, 2, 3, 6, 29, 32, 34, 35, 39, 40, 44], "small": [0, 1, 6, 29], "output": [0, 9, 11, 24, 29, 31, 32, 37, 38, 39, 44], "classif": [0, 3, 4, 6, 15, 20, 22, 24, 39, 40, 41, 42, 44], "phi": [0, 3, 6, 29, 44], "fulfil": [0, 41], "big": 0, "In": [0, 2, 3, 6, 8, 9, 11, 15, 20, 22, 24, 31, 32, 44], "you": [0, 2, 8, 11, 17, 22, 23, 24, 32, 33, 44], "can": [0, 1, 2, 3, 4, 7, 8, 9, 11, 13, 15, 20, 21, 22, 23, 24, 25, 26, 29, 31, 32, 33, 34, 39, 40, 41, 44], "see": [0, 1, 3, 8, 9, 10, 13, 20, 22, 24, 29, 36, 39, 44], "differ": [0, 1, 2, 6, 7, 8, 9, 11, 15, 21, 22, 24, 26, 30, 31, 37, 38, 39, 40, 41], "between": [0, 3, 6, 9, 21, 22, 24, 37, 39, 41, 42], "left": [0, 6, 46], "right": [0, 1, 6, 23, 39], "panda": [0, 38], "unobserv": [0, 2, 24], "classifi": [0, 3, 4, 6, 7, 15, 22, 24, 32, 34, 39], "doe": [0, 4, 6, 8, 11, 15, 22, 27, 30, 33, 40, 41], "still": [0, 8, 33], "them": [0, 32, 34, 41], "domainlab": [0, 1, 2, 3, 7, 11, 13, 16, 20], "ar": [0, 1, 2, 4, 6, 7, 8, 9, 11, 15, 20, 21, 22, 24, 25, 29, 32, 37, 39, 40, 41, 44], "creat": [0, 4, 8, 9, 11, 22, 25, 29, 30, 33, 39, 40, 41, 44], "start": [0, 6, 8, 11, 22, 23, 24, 29, 31, 34, 38, 41], "from": [0, 2, 4, 5, 6, 7, 9, 11, 13, 15, 20, 21, 23, 24, 25, 26, 27, 29, 30, 32, 33, 34, 35, 37, 38, 39, 40, 41, 44], "random": [0, 8, 9, 11, 24, 29, 36, 37, 38, 40, 44], "perturb": [0, 1], "_0": 0, "sigma": 0, "tild": 0, "sim": [0, 3, 6], "mathcal": [0, 1, 3, 6], "n": [0, 4, 6, 8, 29, 39], "0": [0, 1, 2, 3, 5, 6, 7, 8, 9, 11, 12, 15, 18, 22, 24, 30, 32, 35, 39, 40, 41, 44], "step": [0, 9, 11, 24, 29, 41], "descend": 0, "size": [0, 2, 6, 7, 9, 11, 15, 24, 25, 29, 35, 39, 40, 44], "tau": [0, 6, 11, 24, 30], "maxim": [0, 7, 11, 39, 44], "machin": [0, 1, 14, 29, 39], "dure": [0, 2, 8, 15, 29, 32, 39, 42], "process": [0, 6, 8, 11, 23], "make": [0, 8, 10, 23, 25, 29, 33, 40, 41, 44], "more": [0, 1, 2, 3, 7, 8, 9, 11, 22, 24, 25, 29, 31, 39], "robust": 0, "explain": [0, 4, 8, 14, 22], "har": 0, "ab": [0, 1], "1412": 0, "6572": 0, "consist": [0, 40], "three": [0, 2, 6, 15, 39], "part": [0, 2, 29, 41], "At": [0, 11], "first": [0, 1, 4, 6, 8, 17, 29, 32, 39, 40, 41], "featur": [0, 3, 4, 6, 9, 15, 22, 24, 26, 31, 32, 35, 39, 40, 41, 44], "extractor": [0, 3, 26, 31, 35, 39, 44], "which": [0, 1, 2, 3, 4, 6, 7, 8, 9, 10, 11, 15, 20, 21, 22, 24, 27, 29, 32, 33, 34, 35, 39, 40, 41, 44], "extract": [0, 3, 4, 7, 9, 24, 31, 32, 35, 39, 41, 44], "main": [0, 29], "characterist": 0, "thi": [0, 1, 2, 3, 4, 6, 7, 8, 9, 11, 13, 15, 17, 20, 21, 22, 23, 24, 26, 29, 32, 33, 34, 35, 37, 38, 39, 40, 41, 44], "input": [0, 2, 15, 25, 29, 30, 32, 33, 34, 39, 40, 41, 44], "label": [0, 2, 4, 6, 15, 22, 29, 34, 35, 39, 40, 41], "optim": [0, 1, 3, 29, 32, 33], "have": [0, 6, 7, 8, 9, 11, 17, 22, 23, 30, 32, 39, 40, 41, 44], "low": 0, "error": [0, 9, 41], "ensur": [0, 7, 23, 31, 33, 34, 37], "intern": [0, 1, 29], "represent": [0, 2, 3, 6, 35, 39, 40], "cannot": [0, 11], "discrimin": 0, "goal": [0, 1, 4, 6, 29], "archiv": 0, "special": [0, 23], "combin": [0, 5, 7, 26, 29, 36, 37, 38, 41, 44], "2": [0, 1, 3, 5, 6, 7, 8, 11, 13, 15, 24, 29, 32, 33, 34, 35, 36, 39, 40, 41, 42, 44], "packag": [0, 1, 7, 8, 21, 22, 42, 43], "one": [0, 1, 2, 3, 4, 6, 8, 10, 11, 20, 22, 29, 30, 32, 33, 34, 35, 36, 38, 39, 40, 41, 42, 44], "describ": 0, "paper": [0, 2, 5, 6, 15, 29, 36], "It": [0, 8, 17, 20, 22, 32], "standard": [0, 9, 11, 36, 39], "cross": [0, 3, 4, 20, 29, 39, 44], "entropi": [0, 3, 4, 20, 39, 44], "predict": [0, 4, 6, 15, 21, 28, 35, 39, 40, 41], "probabl": [0, 4, 39, 41], "actual": [0, 15, 41], "ce_": 0, "nat": 0, "adv": 0, "weight": [0, 1, 2, 4, 5, 6, 24, 29, 31, 33, 34, 39], "gamma_": [0, 6], "text": [0, 1, 6, 21, 22, 37], "reg": [0, 6, 24, 41], "procedur": [0, 6, 11], "yield": [0, 6], "follow": [0, 1, 2, 4, 6, 7, 8, 9, 11, 13, 14, 15, 20, 22, 32, 44], "avail": [0, 6, 7, 9, 24, 33, 40], "hyperparamet": [0, 3, 4, 6, 7, 15, 38, 41, 42, 44], "dial_steps_perturb": [0, 24], "how": [0, 7, 11, 24, 27, 33, 44], "mani": [0, 11, 32], "go": [0, 41], "dial_noise_scal": [0, 24], "varianc": [0, 2, 15, 29, 35, 39, 42], "gaussian": [0, 29], "nois": 0, "inject": 0, "pure": 0, "dial_lr": [0, 24], "rate": [0, 7, 9, 11, 23, 24], "dial_epsilon": [0, 24], "pixel": [0, 35], "wise": [0, 21, 23, 29, 35], "threshold": [0, 27], "gamma_reg": [0, 4, 6, 11, 24], "epsilon": [0, 1, 29], "lr": [0, 7, 9, 11, 24], "alpha": [0, 2, 32, 39], "python": [0, 1, 2, 3, 4, 5, 6, 7, 8, 11, 12, 13, 15, 17, 18, 19, 20, 24, 26, 31, 40, 41], "main_out": [0, 1, 2, 3, 4, 5, 6, 7, 8, 12, 13, 15, 18, 19, 22, 24, 44], "py": [0, 1, 2, 3, 4, 5, 6, 7, 8, 11, 12, 13, 15, 16, 17, 18, 19, 20, 22, 23, 24, 25, 35, 36, 39, 44, 45, 46], "te_d": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 13, 15, 18, 19, 22, 24, 37, 41, 44], "mnistcolor10": [0, 2, 3, 5, 6, 7, 11, 12, 15, 18, 22, 24], "erm": [0, 1, 3, 5, 6, 12, 18, 19, 22, 24, 26, 29, 32, 39], "nname": [0, 1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 13, 15, 18, 19, 22, 24, 31], "conv_bn_pool_2": [0, 2, 3, 5, 6, 7, 11, 12, 15, 18, 22, 24], "keep_model": [0, 15, 21, 24], "tr_d": [0, 2, 5, 6, 7, 9, 11, 12, 15, 18, 22, 24], "3": [0, 2, 3, 4, 5, 6, 7, 8, 11, 12, 13, 15, 17, 18, 26, 29, 31, 32, 33, 35, 36, 37, 39, 40, 41], "7": [0, 6, 7, 8, 12, 15, 18, 22, 32], "nname_dom": [0, 7, 11, 15, 22, 24], "gamma_i": [0, 2, 6, 7, 11, 15, 22, 24, 39, 44], "7e5": [0, 2, 6, 15, 22, 24], "gamma_d": [0, 7, 11, 15, 22, 24, 39, 44], "1e5": [0, 7, 11, 15, 22, 24, 38], "regular": [1, 6, 20, 24, 29, 39, 41], "techniqu": [1, 9], "local": [1, 8, 29, 36, 40], "align": [1, 29, 42, 44], "level": [1, 2, 3, 6, 11, 24, 27, 29, 35, 39, 41], "loss": [1, 3, 4, 6, 20, 24, 27, 29, 39, 42, 44], "landscap": [1, 29], "around": [1, 29, 40], "final": [1, 2, 6, 27, 29, 41], "minim": [1, 6, 8, 22, 24, 29, 39, 42], "possibl": [1, 11, 20, 31, 41, 44], "done": [1, 6, 7, 8, 11, 20, 28, 29], "consid": [1, 8, 9, 29, 36, 41], "second": [1, 6, 7, 8, 29, 40], "order": [1, 15, 20, 22, 34, 35, 40], "term": [1, 2, 4, 6, 39], "across": [1, 3, 6, 22, 27, 29, 40], "intuit": 1, "two": [1, 2, 3, 4, 6, 8, 9, 11, 15, 20, 29, 39], "highli": 1, "affect": [1, 6, 27, 33], "its": [1, 2, 6, 7, 8, 11, 24, 29, 33, 39], "onli": [1, 4, 6, 7, 8, 9, 11, 13, 15, 20, 21, 24, 29, 33, 34, 35, 36, 39, 40, 41], "other": [1, 2, 4, 6, 7, 8, 9, 11, 20, 23, 24, 29, 32, 33, 36, 39, 40, 41, 44], "under": [1, 3, 20, 41], "certain": 1, "assumpt": [1, 6], "most": [1, 22, 29, 34], "importantli": [1, 34], "hessian": 1, "being": [1, 4, 9, 20, 22, 39], "posit": [1, 6], "definit": [1, 6, 25, 31], "measur": [1, 24], "A": [1, 7, 22, 33, 39, 41, 44], "b": [1, 6, 44], "score": 1, "i": [1, 3, 4, 6, 8, 11, 29, 35, 36, 39], "theta": [1, 2, 29], "max": [1, 9, 11, 41], "_": [1, 3], "e": [1, 3, 4, 8, 9, 11, 24, 29, 31, 32, 33, 39, 41, 44], "biggl": 1, "r": [1, 3, 8, 17, 29, 32, 44], "frac": [1, 2, 6], "t": [1, 6, 22, 27, 29, 32, 34, 36, 37], "h_a": 1, "leq": 1, "h_b": 1, "wherebi": [1, 4], "denot": [1, 2, 3, 4, 6, 39], "set": [1, 2, 4, 6, 7, 9, 11, 15, 23, 24, 25, 26, 29, 30, 37, 39, 40, 41, 42, 44], "h_e": 1, "network": [1, 7, 9, 11, 12, 24, 26, 29, 31, 35, 39, 41, 42], "paramet": [1, 2, 6, 8, 9, 11, 20, 22, 25, 26, 29, 31, 32, 34, 35, 36, 37, 38, 39, 40, 41, 42], "_e": 1, "object": [1, 6, 26, 27, 28, 29, 30, 31, 33, 34, 35, 37, 39, 40, 41, 44], "method": [1, 4, 7, 8, 11, 23, 26, 29, 33, 34, 39, 40, 41], "forc": 1, "both": [1, 4, 6, 7, 8, 20, 33, 34, 36, 39], "hand": 1, "side": 1, "becom": [1, 8], "repres": [1, 2, 9, 28, 29, 32, 39, 41, 44], "risk": [1, 22, 29, 39, 42, 44], "implicitli": 1, "appli": [1, 9, 11, 20, 23, 24, 25, 32, 41], "For": [1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 20, 22, 23, 24, 25, 28, 29, 32, 33, 34, 39, 40, 44], "suffic": 1, "diagon": [1, 29], "approxim": 1, "let": [1, 3, 29], "space": [1, 2, 6, 15, 24, 39, 41, 44], "all": [1, 3, 6, 7, 8, 9, 11, 15, 20, 22, 24, 29, 32, 33, 36, 37, 38, 39, 40, 41, 44], "function": [1, 3, 6, 8, 20, 22, 24, 25, 29, 31, 32, 33, 34, 36, 37, 38, 39, 40, 41, 42, 44], "l": [1, 6, 8], "sum_": [1, 3, 6], "lambda": [1, 3], "v_e": 1, "v": [1, 29, 39], "2_2": 1, "averag": [1, 29, 42], "within": [1, 2, 8, 9, 32, 39, 40], "each": [1, 3, 5, 6, 8, 9, 11, 15, 21, 23, 28, 29, 30, 31, 32, 34, 35, 36, 38, 39, 40, 41, 44], "comput": [1, 4, 20, 29, 32, 35, 39], "backpack": [1, 29], "dangel": 1, "felix": 1, "frederik": 1, "kunstner": 1, "philipp": 1, "hennig": 1, "pack": 1, "backprop": 1, "1912": 1, "10985": 1, "further": [1, 44], "we": [1, 2, 3, 6, 7, 8, 9, 10, 11, 20, 22, 27, 28, 29, 30, 34, 38, 40, 41, 44], "textnorm": 1, "var": [1, 29, 35], "g": [1, 3, 8, 9, 24, 31, 32, 33, 39, 41, 44], "approx": 1, "diag": 1, "h": [1, 6, 40], "fisher": 1, "inform": [1, 2, 6, 7, 8, 11, 15, 22, 27, 29, 39, 40, 41], "matrix": [1, 6, 24, 37], "again": 1, "empir": [1, 3, 39, 44], "estim": [1, 3, 29, 41], "effici": 1, "detail": [1, 3, 9, 11, 14, 23, 24, 29, 39, 44], "refer": [1, 4, 8, 9, 11, 15, 25, 41], "below": [1, 3, 8, 9, 13, 15, 21, 22, 27, 44], "code": [1, 8, 10, 13, 20, 22, 26, 31, 33, 44], "mini_vlc": [1, 2, 4, 6, 13, 15, 19, 22], "model": [1, 3, 5, 6, 7, 9, 11, 12, 16, 18, 19, 21, 22, 25, 26, 27, 28, 29, 31, 32, 34, 35, 36, 37, 40, 41, 42, 43, 44], "alexnet": [1, 2, 4, 6, 13, 15, 19, 22, 24, 32], "bs": [1, 2, 4, 5, 6, 7, 9, 11, 12, 13, 15, 18, 19, 22, 24, 40, 44], "nocu": [1, 24, 25], "rame": 1, "alexandr": [1, 29], "corentin": [1, 29], "dancett": [1, 29], "matthieu": [1, 29], "cord": [1, 29], "confer": [1, 4, 19, 29, 39], "learn": [1, 2, 6, 7, 9, 11, 14, 15, 23, 24, 29, 39, 42, 44], "pmlr": [1, 15, 29, 39], "2022": [1, 29], "build": [2, 23, 26, 34, 39, 40, 41], "approach": [2, 6, 20, 39], "framework": [2, 20, 23, 39], "autoencod": [2, 39, 42], "facilit": [2, 39], "new": [2, 15, 20, 23, 26, 29, 39, 40, 44], "without": [2, 11, 24, 32, 35, 39, 40, 44], "supervis": [2, 15, 39], "disentangl": [2, 15, 39], "specif": [2, 7, 8, 9, 15, 39, 40, 41], "class": [2, 6, 15, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 39, 40, 41, 44], "even": [2, 11, 31, 33, 34, 39], "complex": [2, 22, 39], "where": [2, 3, 6, 7, 8, 13, 15, 22, 23, 24, 25, 29, 30, 35, 39, 40, 41, 44], "structur": [2, 6, 7, 22, 24, 25, 29, 36, 39, 40, 41, 42, 44], "observ": [2, 6, 16, 20, 25, 26, 27, 29, 37, 39, 41], "base": [2, 6, 8, 20, 23, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 39, 40, 41], "latent": [2, 15, 24, 35, 39], "variabl": [2, 6, 15, 29, 35, 39, 44], "distinct": 2, "z_y": [2, 34], "z_d": [2, 34], "z_x": 2, "residu": [2, 15, 39], "addit": [2, 15, 23, 24, 38, 39], "continu": [2, 11], "s": [2, 22, 23, 24, 35, 36, 39, 40, 44], "potenti": [2, 21], "mean": [2, 9, 11, 21, 22, 29, 32, 35, 41], "encourag": [2, 3, 15], "through": [2, 11, 24, 35, 41, 42], "condit": [2, 6, 15, 34], "prior": [2, 34, 35], "need": [2, 6, 7, 11, 15, 20, 22, 23, 29, 32, 34, 39, 40, 41], "hot": [2, 39, 40, 41], "along": [2, 25, 39, 41, 44], "shown": [2, 7, 20, 22], "probabilist": 2, "graphic": [2, 6, 8], "note": [2, 7, 8, 9, 26, 32, 36, 37, 39, 41, 44], "concatent": 2, "dash": [2, 6], "arrow": [2, 6], "requir": [2, 4, 10, 17, 25, 32, 40], "respec": 2, "elbo": [2, 24, 39], "decompos": [2, 3, 39], "4": [2, 7, 8, 11, 22, 32, 33, 36, 40], "likelihood": [2, 35], "e_": 2, "q": 2, "log": [2, 6, 8, 23, 24, 29, 41], "p_": 2, "kl": 2, "diverg": 2, "beta": 2, "vae": [2, 16, 25, 31, 32, 39], "beta_x": [2, 24, 39], "q_": 2, "phi_x": 2, "theta_x": 2, "beta_i": [2, 24, 39], "phi_i": 2, "theta_i": 2, "y": [2, 3, 6, 15, 29, 34, 39, 40, 44], "beta_d": [2, 24, 39], "phi_": [2, 6], "phi_d": 2, "theta_d": 2, "beta_t": [2, 24, 39], "theta_": 2, "construct": [2, 6, 11, 31, 33, 34, 39, 40], "ad": [2, 26, 29, 31, 39, 40], "auxiliari": [2, 44], "classsifi": 2, "fit": 2, "specifi": [2, 4, 7, 8, 9, 11, 15, 20, 21, 24, 25, 26, 31, 39, 40, 41, 44], "relat": [2, 22], "well": [2, 7, 8, 22, 24, 42], "zd_dim": [2, 11, 15, 24, 34, 35, 39], "zx_dim": [2, 11, 15, 24, 34, 35, 39], "zy_dim": [2, 11, 15, 24, 34, 35, 39], "topic_dim": [2, 24, 34, 35, 39], "dirichlet": [2, 35], "distribut": [2, 3, 9, 24, 29, 35, 39, 40, 41, 42], "user": [2, 4, 8, 9, 11, 13, 15, 21, 24, 25, 26, 29, 31, 36, 39, 40, 41, 44], "least": 2, "via": [2, 3, 4, 8, 13, 22, 27, 28, 29, 39, 40, 44], "npath_encoder_x2topic_h": [2, 24], "file": [2, 13, 17, 20, 21, 24, 26, 31, 33, 34, 36, 37, 38, 39, 40, 41], "path": [2, 8, 9, 11, 13, 22, 24, 31, 32, 35, 36, 38, 40, 41, 44], "map": [2, 22, 39, 40, 41, 44], "modal": 2, "data": [2, 4, 6, 7, 15, 23, 24, 25, 29, 32, 34, 36, 39, 40, 41, 44], "dimension": 2, "hidden": [2, 32, 35], "serv": [2, 3, 33, 39], "h_t": 2, "d": [2, 3, 6, 15, 29, 34, 35, 39, 40, 44], "layer": [2, 3, 12, 26, 29, 31, 32, 35, 42], "concentr": 2, "npath_encoder_sandwich_x2h4zd": [2, 24], "same": [2, 6, 7, 8, 11, 20, 21, 22, 30, 31, 32, 37, 39, 40, 44], "infer": [2, 15, 22, 32, 40], "posterior": 2, "h_d": 2, "zd_mean": 2, "zd_scale": 2, "altern": [2, 8], "could": [2, 3, 8, 10, 11, 13, 20, 22, 39, 44], "exist": [2, 15, 21, 22, 31, 39], "instead": [2, 6, 28, 32, 33, 40, 41], "npath": [2, 6, 9, 11, 13, 15, 22, 24, 31, 44], "nname_encoder_x2topic_h": [2, 6, 11, 24], "nname_encoder_sandwich_x2h4zd": [2, 6, 11, 24], "number": [2, 4, 6, 7, 8, 9, 11, 24, 29, 32, 38, 39, 40, 41, 44], "epoch": [2, 5, 6, 7, 9, 11, 23, 24, 27, 28, 29, 44], "hyper": [2, 24, 29, 35, 38, 39, 41, 42], "warm": [2, 24], "up": [2, 11, 23, 24, 28, 29, 41], "argument": [2, 8, 9, 13, 21, 22, 25, 26, 31, 32, 33, 34, 38, 39, 40, 41], "pleas": [2, 8, 9], "cite": 2, "our": [2, 7, 8, 22, 41, 44], "inproceed": 2, "sun2021hierarch": 2, "titl": [2, 41, 44], "author": [2, 6, 44], "sun": [2, 39, 44], "xudong": [2, 29, 39, 44], "buettner": [2, 39], "florian": [2, 39], "booktitl": 2, "iclr": 2, "2021": [2, 39], "robustml": 2, "workshop": 2, "2101": [2, 39], "09436": [2, 39], "year": [2, 44], "caltech": [2, 4, 6, 13, 15, 19, 22, 24, 44], "resnet": [2, 6, 13, 15, 22, 44], "cdot": 3, "classificaiton": [3, 39], "w": [3, 24, 29, 32, 40], "ell": [3, 29, 44], "circ": [3, 6], "mathbb": 3, "_d": 3, "domain": [3, 4, 5, 9, 11, 20, 24, 26, 27, 28, 29, 30, 31, 34, 36, 38, 39, 40, 41, 42], "idea": [3, 8], "choos": [3, 7, 11, 13, 24, 41, 42], "intersect": 3, "arg": [3, 6, 25, 26, 28, 29, 30, 31, 32, 33, 35, 36, 37, 39, 40, 41], "min": [3, 6, 9, 11, 41], "bar": [3, 44, 46], "quad": [3, 6], "foral": 3, "regardless": 3, "constraint": [3, 11, 29, 41], "choic": [3, 8, 24, 25], "get": [3, 8, 22, 25, 26, 27, 29, 31, 35, 37, 39, 40, 41], "thu": [3, 7, 9, 39, 41, 44], "form": [3, 6, 8, 36, 41, 44], "bi": 3, "jointli": 3, "hard": [3, 7, 21, 22], "solv": [3, 4, 39], "so": [3, 6, 8, 20, 21, 22, 29, 31, 32, 33, 34, 37, 40, 42, 44], "practic": 3, "write": [3, 37], "result": [3, 4, 9, 11, 21, 24, 33, 37, 38, 41, 44], "optmiz": 3, "min_": 3, "nabla_": [3, 29], "control": [3, 29], "trade": 3, "off": 3, "penalti": [3, 44], "One": [3, 8, 11, 20, 39, 44], "interpret": [3, 8], "orthogon": 3, "gradient": [3, 23, 29, 32, 42], "simpli": [3, 44], "divid": [3, 35], "mini": [3, 29, 39], "batch": [3, 6, 7, 9, 11, 23, 24, 25, 29, 32, 35, 39, 40, 41, 44], "subset": 3, "j": [3, 6, 29, 35], "index": [3, 24, 29, 36, 37, 38, 39, 40, 41, 42, 44, 46], "multipli": [3, 15, 24, 39, 44], "unbias": [3, 29], "l2": 3, "norm": [3, 29, 32], "squar": [3, 29, 36], "inner": [3, 29, 39], "product": [3, 29, 30, 39, 41], "dimens": [3, 24, 29, 30, 31, 32, 34, 39, 40, 41], "dim": [3, 29, 36, 39, 40, 41], "grad": [3, 29, 32], "section": [3, 9, 11, 29], "appendix": [3, 29], "arjovski": [3, 29], "et": [3, 4, 14, 19, 29, 39], "al": [3, 4, 14, 19, 29, 39], "extend": [4, 24, 25, 26, 29, 32, 39], "understand": [4, 24], "concept": [4, 7, 44], "spatial": 4, "correl": [4, 6, 29, 42], "train": [4, 7, 9, 11, 14, 21, 23, 24, 26, 27, 28, 29, 32, 34, 35, 36, 37, 39, 40, 41, 42], "also": [4, 7, 8, 9, 11, 15, 20, 22, 24, 26, 32, 33, 36, 39, 41, 42, 44], "jigsaw": [4, 39], "puzzl": [4, 39], "To": [4, 8, 9, 10, 13, 20, 22, 24, 33, 39, 44], "split": [4, 21, 24, 26, 29, 40], "time": [4, 6, 11, 22, 34, 38, 40, 41], "patch": [4, 39], "permut": [4, 24, 29, 36, 39], "correct": [4, 7, 20, 22, 40, 41], "problem": [4, 15], "parallel": [4, 35], "origin": [4, 29, 35, 39], "fed": [4, 15, 39], "convolut": [4, 29, 31, 32, 35], "given": [4, 6, 7, 8, 24, 28, 32, 35, 36, 40, 41], "total": [4, 11, 24], "plu": 4, "anoth": [4, 6, 9, 11, 22, 29, 32, 35, 39, 40], "instanc": [4, 21, 24, 28, 29, 32, 39, 41], "rel": [4, 6, 10, 21, 24, 40], "ratio": [4, 40], "advantag": [4, 15], "despit": 4, "miss": [4, 15], "nperm": [4, 11, 24, 39], "pperm": [4, 11, 24], "abov": [4, 8, 10, 11, 20, 27, 44], "furthermor": [4, 9, 15], "custom": [4, 9, 23, 24, 25, 26, 32, 39, 42], "grid": [4, 9, 11, 24, 39, 41], "length": [4, 24, 39], "grid_len": [4, 24, 36], "carlucci": [4, 39], "fabio": [4, 39], "m": [4, 6, 39, 44], "gener": [4, 5, 7, 20, 21, 22, 24, 28, 29, 34, 36, 37, 38, 39, 40, 41, 42], "proceed": [4, 19, 39], "ieee": [4, 39], "cvf": [4, 39], "vision": [4, 36, 39], "pattern": [4, 26, 27, 28, 29, 31, 34, 35, 39, 44], "recognit": [4, 39], "2019": [4, 39], "debug": [4, 6, 13, 15, 18, 19, 22, 24], "8": [4, 7, 11, 13, 19, 29, 32, 36, 41], "100": [4, 11, 36], "sketch": [4, 22], "tpath": [4, 9, 11, 22, 24, 44], "demo_task_path_list_smal": [4, 22], "san_check": [4, 7, 9, 22, 24], "convex": [5, 29], "layei": 5, "ensembl": [5, 29], "improv": [5, 15, 29], "select": [5, 6, 8, 11, 24, 26, 27, 28, 29, 31, 34, 39, 40], "boost": [5, 29, 44], "perform": [5, 6, 7, 8, 15, 23, 24, 27, 28, 29, 32, 37, 39, 41, 44], "devansh": [5, 29], "arpit": [5, 29], "huan": [5, 29], "wang": [5, 29], "yingbo": [5, 29], "zhou": [5, 29], "caim": [5, 29], "xiong": [5, 29], "salesforc": [5, 29], "research": [5, 14, 29, 39], "usa": [5, 29], "exampl": [5, 7, 8, 9, 10, 11, 20, 25, 29, 33, 37, 39, 42], "5": [5, 7, 8, 11, 22, 29, 31, 32, 40, 41], "epo": [5, 6, 7, 8, 9, 11, 12, 13, 24, 41, 44], "trainer": [5, 9, 11, 12, 16, 18, 20, 23, 24, 25, 26, 27, 28, 39, 40, 42, 44], "ma": [5, 29], "2006": 6, "07500": 6, "try": [6, 8], "enforc": [6, 9, 32], "includ": [6, 7, 9, 11, 23, 24, 36, 37], "look": [6, 11], "underli": [6, 7], "scm": 6, "o": 6, "directli": [6, 11, 32, 35], "influenc": 6, "true": [6, 9, 28, 29, 32, 33, 34, 36, 40, 41], "y_": 6, "do": [6, 8, 11, 20, 22, 23, 25, 27, 31, 33, 34, 40, 41], "addition": [6, 15, 27, 37, 39, 40], "togeth": [6, 7, 39, 40], "shall": [6, 11, 22, 41], "neuron": [6, 7], "contribut": [6, 24], "provid": [6, 7, 8, 22, 25, 36, 40, 41, 44], "high": [6, 7], "x_c": 6, "common": [6, 9, 22, 29, 34, 39, 40], "ani": [6, 29, 32, 38, 44], "kei": [6, 9, 20, 22, 23, 25, 29, 30, 39, 40, 41], "sublimin": 6, "depend": [6, 9, 11, 24, 37, 44], "x_a": 6, "respect": [6, 8, 9, 10, 15, 22, 23, 39, 41], "shade": 6, "node": [6, 26, 29, 31, 33, 34, 40, 41], "mai": [6, 8, 11], "befor": [6, 7, 8, 11, 23, 24, 29], "defin": [6, 8, 9, 10, 11, 20, 23, 24, 32, 33, 34, 37, 40, 41], "c": [6, 8, 35, 39, 40, 44], "henc": [6, 11], "rightarrow": 6, "sinc": [6, 13, 15, 32, 33, 39, 40, 41], "unknown": [6, 25], "By": [6, 15, 21], "x_j": 6, "x_k": 6, "dist": 6, "associ": [6, 44], "omega": 6, "begin": [6, 23], "case": [6, 7, 8, 9, 22, 40, 41], "correspond": [6, 7, 8, 21, 22, 29, 32, 38, 40, 44], "otherwis": [6, 29, 40, 41], "end": [6, 7, 24, 35, 36], "substack": 6, "neq": 6, "desir": 6, "f": [6, 11, 23, 35, 44], "initialis": [6, 11], "pair": [6, 7, 40], "point": [6, 7, 9, 11, 41], "randomli": [6, 8, 9, 11, 29], "chosen": [6, 11, 36], "highest": 6, "sampl": [6, 15, 38, 40, 41], "lead": [6, 11, 15, 44], "k": [6, 40], "sum": [6, 27, 29], "over": [6, 11, 41], "ob": 6, "phase": 6, "adapt": [6, 29, 33, 42], "contrast": [6, 29], "everi": [6, 24, 32, 39], "exp": [6, 16, 20, 25, 26, 43], "limits_": 6, "y_i": 6, "y_j": 6, "x_b": 6, "cosin": [6, 24], "similar": [6, 8, 21, 22, 24, 39], "after": [6, 8, 20, 21, 23, 28, 29, 37, 39, 41], "epos_per_match_upd": [6, 24], "tensor": [6, 24, 25, 29, 30, 32, 35, 39, 41], "updat": [6, 8, 23, 24, 26, 27, 28, 29], "nearest": [6, 41], "neighbor": 6, "among": [6, 11], "l_2": 6, "distanc": [6, 29, 30], "underset": 6, "n_d": 6, "x_i": 6, "scratch": [6, 20], "05": [6, 24, 30], "experi": [6, 8, 9, 11, 24, 25, 26, 33, 37, 38, 44], "epochs_ctr": [6, 13, 24], "6": [6, 7, 8, 13, 32, 37], "net": [6, 11, 13, 15, 22, 26, 39, 42, 44], "verion": 7, "digit": [7, 41], "semant": [7, 39], "want": [7, 30, 34, 38, 41], "10": [7, 9, 32, 40, 41], "version": [7, 8, 34, 42, 44], "9": [7, 17, 32], "background": [7, 36], "zero": [7, 40], "nine": 7, "simplic": 7, "show": [7, 20, 32, 33], "red": 7, "blue": 7, "exact": [7, 41], "revers": [7, 42], "seen": 7, "come": [7, 24, 29, 30, 33], "particular": [7, 29], "few": 7, "test": [7, 8, 9, 11, 21, 24, 27, 28, 39, 40, 41, 42, 44], "lot": 7, "here": [7, 8, 10, 11, 20, 22, 23, 24, 34, 36, 39, 44], "appear": [7, 11], "far": 7, "ones": [7, 41], "like": [7, 8, 11, 23, 26, 36, 39, 41, 44], "compar": [7, 21, 24], "wa": [7, 11, 27, 41], "command": [7, 8, 20, 21, 22, 25, 31, 33, 37], "500": 7, "16": [7, 11], "1e": 7, "seed": [7, 8, 11, 24, 37, 38, 41, 44], "san_num": [7, 22, 24], "earli": [7, 9, 11, 24, 27], "stop": [7, 9, 11, 24, 27, 28], "criterion": 7, "summar": 7, "tabl": 7, "acc": [7, 8, 27, 39, 41], "precis": [7, 8, 41], "recal": [7, 8, 41], "f1": [7, 8, 41], "auroc": [7, 8, 41], "798": 7, "858": 7, "800": 7, "978": 7, "797": 7, "832": 7, "959": 7, "961": 7, "958": 7, "995": 7, "999": 7, "program": 7, "descript": [7, 41], "contain": [7, 8, 10, 15, 21, 22, 29, 36, 37, 40, 41, 44], "about": [7, 11, 24, 27, 41], "save": [7, 8, 11, 24, 27, 32, 37, 39, 40, 41], "transform": [7, 12, 22, 36, 40, 44], "should": [7, 9, 22, 27, 28, 29, 31, 32, 33, 35, 37, 38, 39, 41, 44], "found": [7, 8, 9, 13, 22, 25, 41], "doc_task": 7, "md": [7, 8], "valu": [7, 8, 9, 11, 20, 22, 24, 25, 29, 32, 35, 39, 40, 41], "suffici": [7, 44], "termin": 7, "full": [7, 41], "converg": 7, "t0": 7, "basi": [7, 24], "alreadi": [7, 11, 22, 36], "implement": [7, 26, 33, 34, 35, 36, 40, 41, 42, 44], "nn": [7, 25, 30, 31], "sequenti": [7, 32], "conv2d": [7, 32], "in_channel": 7, "num_of_input_channel": 7, "out_channel": 7, "32": [7, 8, 11, 12, 18, 19, 22, 29, 36], "kernel_s": [7, 32], "stride": [7, 32], "bia": [7, 32], "fals": [7, 9, 13, 25, 27, 28, 29, 31, 32, 33, 34, 36, 38, 40, 41], "batchnorm2d": 7, "num_featur": 7, "relu": [7, 32, 35], "maxpool2d": [7, 32], "64": [7, 11, 29, 32], "give": [7, 22, 24], "plot": [7, 24, 36, 38, 41, 44], "check": [7, 23, 24, 27, 29, 33, 36, 37, 39, 41, 42], "document": [8, 9, 22, 25, 44], "markdown": 8, "offer": [8, 9, 22, 44], "abil": 8, "against": [8, 41], "collect": [8, 29], "prepar": [8, 15, 22, 23], "chart": 8, "aspect": 8, "stochast": [8, 38, 41], "variat": [8, 38, 39, 41, 42], "sensit": 8, "investig": 8, "reli": 8, "snakemak": [8, 38], "pulp": 8, "due": [8, 29], "upgrd": 8, "unstabl": 8, "recom": 8, "pip": [8, 17, 25, 44], "current": [8, 9, 11, 21, 26, 29, 32, 35, 38, 39, 40, 41], "datri": 8, "howev": [8, 15], "mamba": 8, "bioconda": 8, "conda": [8, 17], "forg": 8, "work": [8, 11, 21, 31, 33, 34, 37, 42, 44], "configur": [8, 17, 21, 23, 34, 37, 40, 41], "yaml": [8, 24, 38, 41, 44], "doc_benchmark_yaml": 8, "execut": [8, 11, 37, 40, 41], "script": [8, 16, 24], "repositori": [8, 10, 13, 17, 20, 22, 24, 44], "run_benchmark_standalon": 8, "sh": [8, 44], "launch": [8, 44], "login": 8, "dispatch": [8, 28, 40], "run_benchmark_slurm": [8, 44], "insid": [8, 9, 21, 29, 44], "assum": [8, 22, 30, 36, 40], "core": 8, "ha": [8, 13, 22, 29, 30, 32, 33, 34, 36, 39, 41, 44], "been": 8, "linux": 8, "system": 8, "demo_benchmark": 8, "mandatori": [8, 9], "third": 8, "cuda": [8, 24, 37], "option": [8, 9, 11, 23, 38, 39, 40, 41], "fourth": 8, "default": [8, 9, 11, 21, 24, 25, 39, 40], "your": [8, 11, 23], "last": [8, 26, 27, 32, 35, 41], "cpu": [8, 39], "rm": 8, "If": [8, 9, 11, 20, 32, 41], "access": 8, "support": [8, 9, 15], "submiss": 8, "clone": [8, 17, 20, 44], "cd": 8, "sure": [8, 23, 33], "tool": [8, 11, 25], "nohup": 8, "tmux": 8, "keep": [8, 39], "activ": [8, 12, 17, 32], "good": 8, "submit": 8, "pytorch": [8, 20, 21, 22, 26, 32, 35, 36, 37, 38, 39, 40], "help": [8, 24, 25, 44], "out": [8, 22, 24, 27, 29, 32, 38, 40, 42], "fail": [8, 33, 44], "messag": [8, 33], "direct": [8, 29, 32], "bash": [8, 44], "sh_list_error": 8, "zoutput": [8, 11, 21, 36], "sepcif": 8, "slurm_log": 8, "suppos": [8, 17, 22, 34], "14144163": 8, "grep": 8, "run_experi": [8, 16, 25, 43], "41": 8, "err": 8, "name": [8, 9, 11, 17, 20, 22, 24, 26, 29, 31, 33, 34, 36, 37, 39, 40, 41, 44], "directori": [8, 9, 10, 11, 17, 20, 21, 22, 24, 36, 38, 40], "translat": 8, "config": [8, 24, 34, 38, 41], "txt": [8, 17, 21], "commit": [8, 21, 37], "format": [8, 25, 36, 41], "reproduc": [8, 9, 24, 37], "line": [8, 20, 21, 22, 24, 25, 31, 33, 37, 38, 40, 41], "rule_result": 8, "tree": 8, "someth": [8, 21, 22], "diva_fbopt_ful": 8, "radar_dist": 8, "png": [8, 40], "radar": 8, "scatterpl": 8, "sp_matrix_dist": 8, "sp_matrix_dist_reg": 8, "sp_matrix": 8, "sp_matrix_reg": 8, "variational_plot": 8, "json": 8, "visual": 8, "variou": [8, 22, 24, 44], "specificli": 8, "stochastic_vari": 8, "aggreg": [8, 21, 24, 25, 37, 38, 39, 41], "chang": [8, 10, 17, 29, 32, 33, 34, 37, 39, 41, 46], "anyth": [8, 23, 41], "moreov": 8, "subdirectori": 8, "entir": 8, "complet": [8, 21, 23], "yet": [8, 41], "some": [8, 11, 15, 22, 24, 29, 31, 33, 34], "brokenpip": 8, "multiprocess": 8, "pil": 8, "read": [8, 37], "agg_partial_bm": [8, 24], "output_dir": [8, 9, 11, 41], "cat": 8, "clean": [8, 28], "extra": [8, 21, 29, 32, 39, 44], "head": 8, "might": [8, 11, 21, 22, 31, 33, 38], "manual": [8, 37], "must": [8, 9, 11, 20, 32, 39, 40, 41], "gen_plot": [8, 24, 41], "csv_file": 8, "outp_dir": [8, 24], "cv": 8, "fulli": [8, 15, 31, 32], "param_index": [8, 38, 41], "algo": [8, 16, 25, 41, 43], "param": [8, 9, 11, 25, 26, 29, 31, 32, 33, 34, 35, 36, 37, 38, 40, 41], "param1": 8, "p1": [8, 41], "p2": [8, 41], "incorrect": 9, "next": [9, 11, 26], "share": [9, 30, 34, 39, 41, 44], "list": [9, 11, 22, 25, 29, 34, 36, 38, 39, 40, 41, 46], "dial": [9, 11, 42, 44], "mldg": [9, 11, 42, 44], "declar": 9, "domainlab_arg": [9, 11], "address": [9, 15], "int": [9, 11, 25, 35, 38, 39, 40, 41], "float": [9, 11, 39, 41], "es": [9, 11, 24], "neural": [9, 12, 24, 26, 29, 31, 35, 39, 41, 42], "dmem": [9, 24], "startse": [9, 11, 38], "endse": [9, 11, 38], "mode": [9, 11, 22, 23, 24, 29, 32], "search": [9, 11, 42], "num_shared_param_sampl": [9, 11], "sampling_se": [9, 11, 41], "outsid": 9, "num_param_sampl": [9, 11], "multipl": [9, 11, 24, 33, 44], "apart": 9, "real": [9, 29, 31, 33, 34], "num": [9, 11, 41], "ly": 9, "demo_hyperparameter_sampl": 9, "yml": 9, "categor": [9, 11, 41], "valid": [9, 22, 23, 24, 27, 28, 36, 40], "datatyp": [9, 11, 41], "uniform": [9, 41], "loguniform": [9, 41], "normal": [9, 22, 35, 36, 41], "std": [9, 11, 41], "deviat": [9, 11], "geq": 9, "lognorm": [9, 41], "lower": [9, 42], "bound": [9, 42], "upper": 9, "reject": 9, "prevent": 9, "contradictori": 9, "amongst": 9, "concret": [9, 11, 25, 33, 34, 39], "abort": 9, "000": 9, "row": [9, 36, 38, 41], "equal": 9, "p4": 9, "task1": 9, "demo_hypeparameter_sampl": 9, "referenc": [9, 41], "undefin": 9, "behaviour": [9, 41], "diretori": 10, "avoid": [10, 31, 33, 34, 40], "alwai": [10, 15, 29, 31, 33, 34, 37, 40, 41], "discourag": 10, "symbol": 10, "link": 10, "power": [11, 29, 44], "own": [11, 22], "gridsearch": [11, 41], "veri": [11, 44], "fix": [11, 40, 41], "top": 11, "decid": [11, 28], "whether": [11, 28, 32, 44], "switch": 11, "sould": 11, "when": [11, 23, 27, 29, 33, 40, 41], "rang": [11, 36], "mnist_benchmark": 11, "test_domain": [11, 25, 38], "exactli": [11, 30], "per": [11, 29], "run": [11, 20, 21, 24, 25, 32, 38, 41, 44], "happen": [11, 29], "though": [11, 32], "were": [11, 32], "pass": [11, 29, 32, 41], "take": [11, 32, 34, 35], "preced": 11, "pac": [11, 44], "task_pacs_path_list": 11, "001": 11, "50": 11, "iter": [11, 40], "resnet50": 11, "resnet50domainb": 11, "npath_": 11, "pick": 11, "crash": 11, "add": [11, 22, 31, 40, 41], "uniqu": [11, 22, 40, 44], "task_diva_di": 11, "skip": [11, 41], "diva": [11, 22, 24, 26, 31, 39, 42, 44], "adam": 11, "theire": 11, "express": 11, "task_jigen": 11, "task_dann": 11, "There": [11, 20], "wai": [11, 22, 37, 44], "rand": 11, "decis": [11, 44], "previou": 11, "either": [11, 24, 29, 32, 34, 40, 41], "fill": [11, 41], "interv": [11, 24], "01": 11, "useful": 11, "span": 11, "magnitud": 11, "1e4": 11, "1e6": 11, "2e4": 11, "predefin": [11, 24], "30": 11, "31": [11, 39], "taken": 11, "formular": 11, "evalu": [11, 28, 38, 39, 40, 44], "round": [11, 41], "integ": [11, 41], "mention": [11, 20], "othervis": 11, "thei": [11, 23, 32, 41, 44], "96": 11, "dimes": 11, "thep": 11, "subgrid": 11, "84": 11, "80": 11, "sort": 11, "task_nam": [11, 40, 41], "hyperparamt": [11, 41], "standalon": 11, "No": [11, 19], "indic": [11, 22], "nonlinear": [12, 29], "1607": 12, "01719": 12, "signatur": [13, 22], "def": [13, 22, 23, 41], "build_feat_extract_net": 13, "dim_i": [13, 30, 32, 39, 40, 41], "remove_last_lay": [13, 31, 41], "ignor": [13, 25, 32, 36], "fair": [13, 32], "benchmark": [13, 16, 24, 38, 41], "comparison": [13, 32], "folder": [13, 21, 36, 40, 41], "github": [13, 35, 36, 39, 44, 45], "com": [13, 35, 36, 39, 45], "marrlab": [13, 39, 44], "blob": [13, 35, 36, 39], "master": [13, 35, 36, 39], "public": 14, "ganin": [14, 39], "yaroslav": [14, 39], "journal": [14, 39, 44], "17": [14, 39], "2016": [14, 39], "2096": [14, 39], "2030": [14, 39], "independ": [15, 20, 35, 39], "encod": [15, 31, 32, 34, 39, 42], "store": [15, 22, 24, 29, 32, 41], "obtain": 15, "margin": 15, "densiti": 15, "These": 15, "parameter": 15, "learnabl": 15, "singl": [15, 24, 29, 36, 38, 42, 44], "decod": [15, 34, 35, 39], "reconstruct": [15, 34, 39], "overal": [15, 39, 42], "larg": 15, "accuraci": 15, "semi": 15, "turn": 15, "notabl": 15, "better": [15, 40], "close": 15, "setup": [15, 17, 41], "allow": [15, 22, 23, 36, 40], "unsupervis": [15, 39, 42], "clear": [15, 29], "decreas": 15, "alpha_i": 15, "eq": 15, "alpha_d": 15, "npath_dom": [15, 22, 24, 44], "10e5": 15, "gen": [15, 21, 24, 41], "1905": 15, "10427": 15, "medic": [15, 39], "deep": [15, 29, 39, 42, 44], "2020": [15, 39], "a_algo_build": [16, 25, 43], "builder_api_model": [16, 25, 43], "builder_custom": [16, 25, 43], "builder_dann": [16, 25, 43], "builder_diva": [16, 25, 43], "builder_erm": [16, 25, 43], "builder_hduva": [16, 25, 43], "builder_jigen1": [16, 25, 43], "__init__": [16, 33], "msel": [16, 25, 26], "util": [16, 23, 25, 36, 40, 43, 44], "zoo_algo": [16, 25, 43], "arg_pars": [16, 20, 42, 43], "cli": [16, 42, 43], "compo": [16, 25, 26, 29, 39, 43], "a_nn_build": [16, 25, 43], "builder_nn_alex": [16, 25, 43], "builder_nn_conv_bn_relu_2": [16, 25, 43], "builder_nn_external_from_fil": [16, 25, 43], "nn_zoo": [16, 25, 31], "pcr": [16, 25, 31], "utils_conv_get_flat_dim": [16, 25, 43], "zoo_nn": [16, 25, 43], "dset": [16, 25, 40, 43], "a_dset_mnist_color_rgb_solo": [16, 25, 43], "dset_img_path_list": [16, 25, 43], "dset_mnist_color_solo_default": [16, 25, 43], "dset_poly_domains_mnist_color_default": [16, 25, 43], "dset_subfold": [16, 25, 43], "utils_color_palett": [16, 25, 43], "utils_data": [16, 25, 43], "utils_wrapdset_patch": [16, 25, 43], "exp_cuda_se": [16, 25, 43], "exp_main": [16, 25, 43], "exp_util": [16, 25, 43], "exp_protocol": [16, 25, 43], "aggregate_result": [16, 25, 43], "smk": 16, "mk_exp": [16, 42, 43], "a_model_classif": [16, 25, 43], "a_model": [16, 25, 43], "args_jigen": [16, 25, 43], "args_va": [16, 25, 43], "interface_vae_xyd": [16, 25, 43], "model_custom": [16, 25, 43], "model_dann": [16, 25, 43], "model_diva": [16, 25, 43], "model_erm": [16, 25, 43], "model_hduva": [16, 25, 43], "model_jigen": [16, 25, 43], "model_vae_xyd_classif": [16, 25, 43], "a_task_classif": [16, 25, 43], "a_task": [16, 25, 43], "b_task_classif": [16, 25, 43], "b_task": [16, 25, 43], "task_dset": [16, 25, 43], "task_folder_mk": [16, 25, 43], "task_fold": [16, 25, 43], "task_mini_vlc": [16, 25, 43], "task_mnist_color": [16, 25, 43], "task_pathlist": [16, 25, 43], "task_util": [16, 25, 43], "utils_task_dset": [16, 25, 43], "utils_task": [16, 25, 43], "zoo_task": [16, 25, 43], "flows_gen_img_model": [16, 25, 43], "generate_benchmark_plot": [16, 25, 43], "get_git_tag": [16, 25, 43], "hyperparameter_gridsearch": [16, 25, 43], "hyperparameter_sampl": [16, 25, 43], "logger": [16, 24, 25, 43], "override_interfac": [16, 25, 43], "perf_metr": [16, 25, 43], "perf": [16, 25, 43], "sanity_check": [16, 25, 43], "test_img": [16, 25, 43], "u_import_net_modul": [16, 25, 43], "u_import": [16, 25, 43], "utils_classif": [16, 25, 43], "utils_class": [16, 25, 43], "utils_cuda": [16, 25, 43], "utils_img_sav": [16, 25, 43], "zdata": 16, "mixed_codec": 16, "pacs_mini_10": 16, "pacs_split": 16, "patches_permutation4jigsaw": 16, "vlcs_mini": [16, 24], "ztest_fil": 16, "domainlab_py39": 17, "poetri": 17, "pyproject": 17, "toml": 17, "2206": 18, "11646": 18, "causalirl": 18, "li": 19, "da": 19, "aaai": 19, "artifici": 19, "intellig": 19, "vol": 19, "2018": 19, "built": 20, "easili": 20, "guid": [20, 44], "outlin": 20, "necessari": [20, 23, 24, 39], "amodelcustom": [20, 39], "becaus": 20, "amodelclassif": [20, 39], "dict_net_module_na2arg_na": [20, 39], "return": [20, 22, 25, 26, 27, 28, 29, 31, 32, 33, 34, 35, 36, 39, 40, 41], "dictionari": [20, 22, 24, 26, 29, 30, 34, 38, 39, 40, 41], "modul": [20, 42, 43], "call": [20, 22, 23, 28, 32, 33, 35, 37, 39, 41], "get_node_na": 20, "extern": [20, 26, 41], "amodel": [20, 39], "cal_task_loss": [20, 39], "self": [20, 23, 29, 32, 33, 40, 41], "tensor_x": [20, 23, 29, 32, 39], "tensor_i": [20, 23, 29, 39], "primari": 20, "_cal_reg_loss": 20, "tensor_d": [20, 23, 29, 39], "none": [20, 22, 25, 26, 27, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41], "calcul": [20, 28, 29, 35, 39], "pars": [20, 25], "root": [20, 22, 36, 40], "retriev": [20, 32, 41], "nodealgobuild": [20, 26], "inherit": [20, 23, 29, 33, 37, 39, 41], "init_busi": [20, 26, 29, 31, 33, 34, 40], "devic": [20, 24, 28, 29, 34, 35, 39, 41], "chain": [20, 22, 26, 29, 31, 33, 34, 40], "alter": 21, "comment": [21, 22], "aggrst": [21, 37], "task1_test_domain1_tagnam": 21, "csv": [21, 38, 41], "task2_test_domain3_tagnam": 21, "counterfactu": [21, 34], "task1_test_domain1": 21, "saved_model": [21, 37], "persist": [21, 28, 37, 39, 41], "task1_algo1_git": 21, "hashtag1_seed_1_instance_wise_predict": 21, "hashtag1_seed_1": 21, "sever": [21, 23, 29, 34, 36, 38, 44], "ident": [21, 32, 38, 40], "append": [21, 29, 40, 44], "conveni": 21, "upon": [21, 23, 29, 35, 36], "disk": [21, 28, 37, 39, 40], "delet": [21, 24, 32], "accord": [22, 34, 41], "thress": 22, "cover": [22, 38], "feed": [22, 35], "inlin": 22, "what": [22, 23, 44], "straightforward": 22, "emper": [22, 39], "domain1": [22, 40], "domain3": 22, "task_dset_custom": 22, "batchsiz": [22, 25, 29, 31, 36], "nicknam": 22, "locat": [22, 35], "plain": 22, "subfold": [22, 36], "car": 22, "dog": 22, "human": 22, "chair": 22, "bird": 22, "folder_a": 22, "folder_b": 22, "folder_c": 22, "sub": [22, 33, 40], "3rd_domain": 22, "3rd": 22, "1st": 22, "hund": 22, "2nd": 22, "huski": 22, "dict_domain_folder_name2class": [22, 40], "2nd_domain": 22, "don": [22, 27], "entri": [22, 29, 39], "1st_domain": 22, "auto": [22, 39, 42], "mensch": 22, "stuhl": 22, "vogel": 22, "drive": 22, "sit": 22, "fly": 22, "sapien": 22, "sofa": 22, "vehicl": 22, "expect": [22, 41], "separ": [22, 34, 35], "get_task": [22, 40], "na": 22, "import": [22, 25, 40, 41], "os": 22, "torchvis": [22, 31, 32, 36, 40, 41], "mk_task_fold": [22, 40], "imsiz": [22, 40], "trans_tr": 22, "compos": 22, "resiz": [22, 36], "224": [22, 40], "randomresizedcrop": 22, "scale": [22, 32, 35, 41], "randomhorizontalflip": 22, "colorjitt": 22, "randomgrayscal": 22, "totensor": 22, "485": 22, "456": 22, "406": 22, "229": 22, "225": 22, "trans_t": 22, "extens": [22, 36, 40], "jpg": [22, 40], "list_str_i": [22, 39, 40], "dict_domain_img_tran": [22, 40], "img_trans_t": [22, 40], "isiz": [22, 32, 35, 40], "dict_domain2imgroot": [22, 40], "join": [22, 38], "taskna": [22, 40], "task_demo": 22, "print": [22, 23, 29, 32, 39, 40], "__doc__": 22, "matter": [22, 40], "carri": [22, 40], "composit": [22, 40], "knowledg": [22, 40], "channel": [22, 29, 32, 40], "height": [22, 32, 40], "width": [22, 32, 40], "absolut": [22, 40], "task_vlc": [22, 24, 44], "apath": [22, 24, 26], "demo_custom_model": 22, "nname_argna2v": [22, 24], "my_custom_arg_nam": 22, "npath_argna2v": [22, 24], "develop": [23, 44], "establish": 23, "routin": 23, "behavior": [23, 33], "tailor": [23, 37], "overrid": [23, 29, 31, 33, 34, 41], "brief": 23, "overview": [23, 42], "before_tr": [23, 29], "state": [23, 29, 44], "initi": [23, 26, 29, 31, 33, 34, 38, 39, 41], "tr_epoch": [23, 29], "logic": 23, "bulk": 23, "before_epoch": [23, 29], "reset": [23, 26, 27, 28, 29, 39], "counter": 23, "schedul": [23, 29], "after_epoch": [23, 29], "typic": 23, "tr_batch": [23, 29], "ind_batch": [23, 29], "handl": [23, 33, 36, 44], "forward": [23, 32, 35, 39, 40], "backward": [23, 32], "propag": 23, "before_batch": [23, 29], "after_batch": [23, 29], "action": [23, 25, 39], "regist": [23, 31, 32], "zoo_train": [23, 25, 26], "simpl": [23, 29, 42], "mycustomtrain": 23, "super": [23, 39], "session": 23, "enumer": 23, "loader_tr": [23, 40], "minimum": 24, "eiter": 24, "predfin": 24, "python3": 24, "individu": [24, 41], "dann": [24, 39, 42, 44], "gamma": [24, 41], "preserv": 24, "epos_min": 24, "maximum": [24, 27, 29, 39, 41], "epo_t": 24, "warmup": [24, 29, 42], "enabl": [24, 40], "verbos": 24, "memori": [24, 36], "suppress": 24, "no_dump": 24, "confus": [24, 37], "dpath": 24, "download": [24, 44], "tag": [24, 37], "exptag": 24, "aggtag": 24, "track": 24, "partial": [24, 38], "param_idx": [24, 41], "loader": [24, 28, 29, 36, 39, 40, 41], "mix": [24, 36, 40], "proport": 24, "saniti": [24, 39, 41], "count": [24, 40], "loglevel": [24, 41], "shuffl": [24, 36, 39, 40, 42], "shuffling_off": 24, "disabl": 24, "dataload": 24, "topic": [24, 32, 35, 39, 42], "hduva": [24, 26, 39, 42], "zd": [24, 34, 35], "compon": [24, 26, 29, 32, 39, 44], "factor": [24, 34, 44], "magnifi": 24, "match": [24, 29, 30, 40, 42], "frequenc": 24, "ctr": 24, "jigen_ppath": 24, "tile": [24, 36, 39, 42], "unit": 24, "adversari": [24, 26, 29, 39, 42], "comprehens": 24, "toi": 24, "subsampl": [24, 41], "vlc": 24, "invari": [24, 29, 39, 42], "demo_config_single_run_diva": 24, "restrain": 24, "prefix": [24, 29, 39, 41], "a_model_sel": [25, 26], "c_msel_oracl": [25, 26], "c_msel_tr_loss": [25, 26], "c_msel_val": [25, 26], "a_observ": [25, 26], "b_obvisitor": [25, 26], "c_obvisitor_cleanup": [25, 26], "c_obvisitor_gen": [25, 26], "a_train": [25, 26], "args_dial": [25, 26], "hyper_schedul": [25, 26], "mmd_base": [25, 26], "train_bas": [25, 26], "train_causirl": [25, 26], "train_cor": [25, 26], "train_dial": [25, 26], "train_ema": [25, 26], "train_fishr": [25, 26], "train_hyper_schedul": [25, 26], "train_irm": [25, 26], "train_matchdg": [25, 26], "train_mldg": [25, 26], "net_adversari": [25, 31], "net_classif": [25, 31], "net_conv_conv_bn_pool_2": [25, 31], "net_gat": [25, 31], "nn_alex": [25, 31], "nn_torchvis": [25, 31], "p_chain_handl": [25, 31], "request": [25, 26, 29, 31, 34, 39, 40, 41], "a_model_build": [25, 31], "a_vae_build": [25, 31], "c_vae_adaptor_model_recon": [25, 31], "c_vae_builder_classif": [25, 31], "c_vae_recon": [25, 31], "utils_request_chain_build": [25, 31, 39], "zoo_vae_builders_classif": [25, 31], "zoo_vae_builders_classif_top": [25, 31], "hyperparameter_retriev": [25, 43], "parsevaluesorkeyvaluepair": 25, "option_str": 25, "dest": [25, 41], "narg": 25, "const": 25, "type": [25, 32, 39, 40], "metavar": 25, "apply_dict_to_arg": 25, "dict": [25, 30, 38, 41], "tri": [25, 41], "silent": [25, 32], "long": 25, "mk_parser_main": 25, "parse_cmd_arg": 25, "interfac": [25, 28, 33, 35, 39, 40, 41], "domainlab_cli": 25, "instal": 25, "str": [25, 36, 38, 40, 41], "mk_task_dset": [25, 40], "explan": 25, "nameofmodel": 25, "mk_": 25, "model_": 25, "string": [25, 26, 28, 29, 31, 33, 34, 36, 39, 40, 41], "global": 25, "whole": [25, 33], "g_list_model_penalized_reg_agg": 25, "list_penalized_reg": 25, "diminish": 25, "matchdg_arg": [26, 29], "matchdg_match": [26, 29], "matchdg_util": [26, 29], "parent": [26, 29, 33, 35, 39, 40], "comb": 26, "success_nod": [26, 33, 34], "abstractchainnodehandl": [26, 29, 31, 33, 34, 40], "builder": [26, 31, 34], "next_model": 26, "attribut": [26, 32, 34, 37, 41], "abstract": [26, 27, 28, 29, 33, 34, 36, 39, 40], "init_next_model": 26, "decor": [26, 28, 29, 36, 37, 39, 40, 41, 44], "is_myjob": [26, 29, 31, 33, 34, 40], "na_prefix": 26, "properti": [26, 27, 28, 29, 36, 39, 40, 41], "api": [26, 29], "backbon": 26, "nodealgobuilderapimodel": 26, "make_basic_train": 26, "class_name_model": 26, "accept": [26, 27, 28, 32, 33, 41], "scheme": 26, "nodealgobuilderdann": 26, "reset_aux_net": [26, 39], "net_encod": [26, 39], "auxilliari": 26, "extract_semantic_feat": [26, 39], "nodealgobuilderdiva": 26, "get_train": 26, "respons": [26, 28, 29, 31, 33, 34, 40], "fetch": [26, 32, 35, 36], "nodealgobuildererm": 26, "cmd": 26, "nodealgobuilderhduva": 26, "jigen": [26, 29, 36, 39, 42, 44], "nodealgobuilderjigen": 26, "split_net_feat_last": 26, "algobuilderchainnodegett": 26, "hardcod": [26, 31, 34, 40], "register_external_nod": 26, "amsel": 27, "val_threshold": 27, "observer4msel": 27, "visitor": [27, 28, 37, 39], "tr_observ": 27, "base_upd": 27, "clear_count": 27, "best": 27, "boolean": [27, 32], "best_te_metr": 27, "decorate": [27, 29, 39], "metric": [27, 28, 29, 39, 41], "best_val_acc": 27, "val": [27, 30, 41], "early_stop": 27, "if_stop": 27, "acc_val": 27, "max_": 27, "model_selection_epoch": 27, "selector": [27, 28], "sel_model_te_acc": 27, "accuaraci": 27, "treshold": 27, "decoupl": [27, 33, 34, 44], "mseloraclevisitor": 27, "oracl": [27, 28], "interven": 27, "innermost": 27, "oracle_last_setpoint_sel_te_acc": 27, "setpoint": 27, "mseltrloss": 27, "mselvalperf": 27, "aobvisitor": 28, "invit": 28, "after_al": 28, "clean_up": [28, 37], "set_task": 28, "coupl": [28, 33, 34, 37], "bool": [28, 32, 35, 36, 38, 39, 41], "obvisitor": 28, "model_sel": 28, "cal_oracle_perf": 28, "dump_predict": 28, "model_ld": 28, "metric_t": [28, 39], "load": [28, 29, 37, 38, 39, 40, 41], "str_metric4msel": [28, 29], "obvisitorcleanup": 28, "els": [28, 41], "obvisitorgen": 28, "abstracttrain": 29, "successor_nod": [29, 31, 34], "director": 29, "flow": [29, 40, 44], "probe": 29, "cal_reg_loss": [29, 39], "_model": 29, "cal_reg_loss_over_task_loss_ratio": 29, "dset_decoration_args_algo": [29, 39], "ddset": [29, 39], "item": [29, 36, 39], "deleg": [29, 32, 39], "get_model": 29, "recurs": [29, 41], "aconf": 29, "flag_accept": 29, "list_tr_domain_s": [29, 30], "p_na_prefix": [29, 39], "post_tr": 29, "print_paramet": [29, 39], "child": [29, 34, 39, 40, 41], "mk_opt": 29, "parmaet": 29, "add_args2parser_di": 29, "parser": [29, 30, 39, 45], "argpars": 29, "hyperschedulerwarmupexponenti": 29, "kwarg": [29, 32, 33, 34, 39], "hyperschedulerwarmuplinear": 29, "hyperschedul": 29, "exponenti": 29, "par_setpoint": 29, "par": 29, "ramp": 29, "steadi": 29, "total_step": 29, "set_step": 29, "gradual": 29, "alexej": [29, 44], "trainermmdbas": 29, "trainerbas": 29, "causal": [29, 42], "gaussian_kernel": 29, "kernel": 29, "mmd": 29, "discrep": 29, "my_cdist": 29, "x1": [29, 30], "x2": [29, 30], "basic": [29, 40], "member": [29, 33], "cal_loss": [29, 39], "log_loss": 29, "list_b_reg_loss": 29, "loss_task": 29, "just": [29, 32], "epo_reg_loss_tr": 29, "list_divid": 29, "list_val": 29, "scalar": [29, 41], "alex": [29, 42], "trainercausalirl": 29, "coral": [29, 42], "au": 29, "trainercor": 29, "cross_domain_mmd": 29, "tuple_data_domains_batch": 29, "pairwis": [29, 30], "trainerdi": 29, "gen_adversari": 29, "img_natur": 29, "vec_i": [29, 34, 39, 41], "naiv": 29, "trim": 29, "img": [29, 35, 36, 41], "necessarili": 29, "hoc": 29, "move": [29, 39, 42], "ma_trainer2_trainer3": 29, "outer": 29, "trainerma": 29, "goe": 29, "block": 29, "break": [29, 33, 37], "move_averag": 29, "dict_data": 29, "trainerfishr": 29, "reduc": 29, "inconsist": [29, 42], "ram\u00e9": 29, "fishr": [29, 42], "cal_dict_variance_grad": 29, "question": 29, "v_i": 29, "d_i": 29, "vector": [29, 35, 39, 40, 41], "layer1": 29, "11": [29, 32], "cal_mean_across_dict": 29, "list_dict": 29, "cal_power_single_dict": 29, "mdict": 29, "element": [29, 39], "var_grads_and_loss": 29, "tupel": 29, "variance_between_dict": 29, "list_dict_var_paragrad": 29, "nsum_i": 29, "interest": 29, "nsum_": 29, "sum_i": 29, "layer2": 29, "minibatch": [29, 39], "trainerhyperschedul": 29, "num_batch": 29, "set_schedul": 29, "flag_update_epoch": 29, "flag_update_batch": 29, "strategi": 29, "wheter": 29, "trainerirm": 29, "irmv1": [29, 42], "half": 29, "delta_": 29, "wdot": 29, "matchdg": [29, 30, 42], "trainermatchdg": 29, "flag_erm": 29, "mk_match_tensor": 29, "batch_idx": 29, "x_e": 29, "y_e": 29, "d_e": 29, "match_tensor_reshap": 29, "batch_tensor_ref_domain2each": 29, "ref_domain": 29, "img_h": 29, "img_w": 29, "easier": 29, "accomod": 29, "meta": [29, 42], "trainermldg": 29, "prepare_ziped_load": 29, "virtual": [29, 44], "trainerchainnodegett": 29, "str_trainer": 29, "xxx": 29, "add_args2parser_matchdg": 30, "matchpair": 30, "i_c": [30, 31, 33, 35, 40], "i_h": [30, 31, 33, 35, 40, 41], "i_w": [30, 31, 33, 35, 40], "bs_match": 30, "virtual_ref_dset_s": 30, "num_domains_tr": 30, "matchdictinit": 30, "creator": 30, "get_num_row": 30, "matchdictnumdomain2sizedomain": 30, "kth": 30, "size_domain_k": 30, "matchdictvirtualrefdset2eachdomain": 30, "dist_cosine_agg": 30, "torch": [30, 37], "cosinesimilar": 30, "dist_pairwise_cosin": 30, "necesarilli": 30, "shape": [30, 39], "cartesian": [30, 41], "fun_tensor_norm": 30, "tensor_batch_x": 30, "get_base_domain_size4match_dg": 30, "nomin": [30, 36], "decoder_concat_vec_reshape_conv": [31, 34], "decoder_concat_vec_reshape_conv_gated_conv": [31, 34], "decoder_cond_prior": [31, 34], "decoder_loss": [31, 34], "encoder_dirichlet": [31, 34], "encoder_domain_top": [31, 34], "encoder_domain_topic_img2top": [31, 34], "encoder_domain_topic_img_topic2zd": [31, 34], "encoder_xyd_parallel": [31, 34], "encoder_xydt_elev": [31, 34], "encoder_zi": [31, 34], "integr": [31, 34], "abstractfeatextractnnbuilderchainnod": 31, "heavi": [31, 33, 34], "busi": [31, 33, 34], "light": [31, 33, 34], "dim_out": 31, "flag_pretrain": [31, 32, 35], "job": [31, 33, 34, 38], "subclass": [31, 32, 33, 34, 39], "servic": [31, 33, 34], "args_nnam": 31, "mknodefeatextractnnbuildernamealex": 31, "arg_name4net": 31, "arg_val": 31, "mknodefeatextractnnbuildernameconvbnrelu2": 31, "conv_strid": [31, 32, 35], "reus": [31, 34, 44], "famili": [31, 39], "28": 31, "mknodefeatextractnnbuilderexternfromfil": 31, "arg_name_net_path": 31, "get_flat_dim": 31, "i_channel": [31, 32, 35], "flat": 31, "connect": [31, 32, 35], "featextractnnbuilderchainnodegett": 31, "arg_name_of_net": 31, "arg_path_of_net": 31, "autogradfunmultipli": 32, "static": [32, 39, 40, 41, 44], "ctx": 32, "grad_output": 32, "formula": 32, "differenti": 32, "oper": [32, 33, 39, 40], "automat": 32, "alia": 32, "vjp": 32, "overridden": 32, "context": 32, "non": 32, "needs_input_grad": 32, "tupl": [32, 36, 38, 40, 41], "arbitrari": [32, 39], "compat": [32, 39], "save_for_backward": 32, "intend": 32, "equival": 32, "save_for_forward": 32, "jvp": 32, "autogradfunreversemultipli": 32, "doc": [32, 37], "stabl": [32, 36, 37], "autograd": 32, "html": [32, 36, 37], "flatten": [32, 39], "although": 32, "recip": 32, "afterward": [32, 41], "former": 32, "care": 32, "hook": 32, "latter": 32, "classifdropoutrelulinear": 32, "z_dim": [32, 35], "target_dim": [32, 41], "dropout": 32, "linearli": 32, "z_vec": 32, "netconvbnrelupool2l": 32, "dim_out_h": 32, "netconvdens": 32, "dense_lay": 32, "until": 32, "replac": [32, 41], "mk_conv_bn_relu_pool": 32, "max_pool_strid": 32, "maxpool_2d": 32, "input_channel": 32, "output_channel": 32, "pad": 32, "dilat": 32, "gatedconv2d": 32, "gateddens": 32, "input_s": 32, "output_s": 32, "densenet": [32, 35], "input_flat_s": 32, "out_hidden_s": 32, "1024": 32, "last_feat_dim": 32, "4096": 32, "p_dropout": 32, "linear": 32, "layerid": 32, "alex4deepal": 32, "alexnetbas": 32, "nettorchvisionbas": 32, "inplac": 32, "ceil_mod": 32, "192": 32, "384": 32, "256": [32, 35], "12": 32, "avgpool": 32, "adaptiveavgpool2d": 32, "p": [32, 34, 35, 39], "in_featur": 32, "9216": 32, "out_featur": 32, "fetch_net": 32, "alexnetnolastlay": 32, "depth": 32, "net_torchvis": 32, "successfulli": 33, "design": [33, 44], "constructor": [33, 34, 35, 37, 39], "handler": 33, "factori": [33, 41], "redund": [33, 34], "safest": 33, "solut": 33, "karg": [33, 34], "chainnod": 33, "onc": 33, "init": [33, 39], "invok": 33, "print_opt": 33, "set_par": 33, "dummybusi": 33, "dummi": 33, "dummychainnodehandlerbeav": 33, "dummychainnodehandlerlazi": 33, "request4chain": 33, "field": 33, "safeti": 33, "convert": [33, 34, 38, 39, 40], "obj": [33, 41], "pre": [33, 37, 44], "requestargs2expcmd": 33, "isol": [33, 41], "requesttask": 33, "requestvaebuilderchw": 33, "requestvaebuildernn": 33, "net_class_d": [33, 35], "net_x": [33, 35], "net_class_i": [33, 35], "test_exp": 33, "patter": 34, "abstractmodelbuilderchainnod": 34, "abstractvaebuilderchainnod": 34, "build_decod": 34, "build_encod": 34, "adaptor": 34, "intens": [34, 35], "refactor": 34, "old": 34, "adaptorreconvaexyd": 34, "cal_lat": 34, "won": 34, "retrain": 34, "cal_prior_zd": 34, "vec_d": [34, 41], "cal_prior_zi": 34, "recon_ydx": 34, "zy": [34, 35], "zx": [34, 35], "concatn": 34, "plai": 34, "vital": 34, "role": 34, "chainnodevaebuilderclassifcondprior": 34, "bridg": [34, 35], "construct_classifi": 34, "input_dim": 34, "output_dim": 34, "construct_cond_prior": 34, "reconvaexyd": 34, "na_adaptor": 34, "recon": 34, "sample_p_zi": 34, "sample_p_zd": 34, "scalar_zx2fil": 34, "recon_cf": 34, "na_cf": 34, "dim_cf": 34, "zx2fill": 34, "countefactu": 34, "vaechainnodegett": [34, 39], "scenario": [34, 44], "vabuild": 34, "chainnodevaebuilderclassifcondpriorbas": 34, "ae": 34, "config_img": 34, "flag": 34, "nodevaebuilderarg": 34, "commandlin": [34, 38, 39], "nodevaebuilderimgalex": 34, "nodevaebuilderimgconvbnpool": 34, "nodevaebuilderus": 34, "test_mk_exp": 34, "nodevaebuilderimgtop": 34, "concaten": 35, "decoderconcatlatentfcreshapeconv": 35, "cls_fun_nll_p_x": 35, "net_fc_z2flat_img": 35, "net_conv": 35, "net_p_x_mean": 35, "net_p_x_log_var": 35, "re": [35, 36, 40], "arrang": 35, "textur": 35, "cal_p_x_pars_loc_scal": 35, "vec_z": 35, "z": 35, "concat_ydx": 35, "concat_ytdx": 35, "decoderconcatlatentfcreshapeconvgatedconv": 35, "lscondpriorlinearbnrelulinearsoftplu": 35, "hyper_prior_dim": 35, "hidden_dim": 35, "hyper_prior": 35, "nllpixellogistic256": 35, "reduce_dim": 35, "bin_siz": 35, "00390625": 35, "neg": 35, "bin": 35, "x_": 35, "openai": 35, "iaf": 35, "tf_util": 35, "l29": 35, "softplu": 35, "lsencoderconvbnrelupool": 35, "pool": [35, 39], "lsencoderlinear": 35, "dim_input": 35, "encoderh2dirichlet": 35, "dim_top": 35, "encoderimg2topicdirzd": 35, "num_top": 35, "encoderimg2topicdistri": 35, "encodersandwichtopicimg2zd": 35, "img_h_dim": 35, "sandwich": [35, 42], "vec_top": 35, "xydencoderparallel": 35, "net_infer_zd": 35, "net_infer_zx": 35, "net_infer_zi": 35, "q_zd": 35, "zd_q": 35, "q_zx": 35, "zx_q": 35, "q_zy": 35, "zy_q": 35, "infer_zy_loc": 35, "xydencoderparallelalex": 35, "reimplement": 35, "contructor": 35, "xydencoderparallelconvbnrelupool": 35, "xydencoderparallelextern": 35, "xydencoderparallelus": 35, "xydtencoderarg": 35, "xydtencoderelev": 35, "net_infer_zd_top": 35, "encoderconnectlastfeatlayer2z": 35, "arg_nam": 35, "arg_path_nam": 35, "transpar": 35, "color": [36, 40, 42], "mnist": [36, 40, 42], "adsetmnistcolorrgbsolo": 36, "ind_color": 36, "subset_step": 36, "color_schem": 36, "label_transform": 36, "mk_fun_label2onehot": 36, "fun_label2onehot": 36, "list_transform": 36, "raw_split": 36, "flag_rand_color": 36, "palett": [36, 40], "spectrum": 36, "subdomain": 36, "foreground": 36, "get_background_color": 36, "ind": [36, 40], "get_foreground_color": 36, "get_num_color": 36, "dsetimpathlist": 36, "root_img": 36, "path2filelist": 36, "trans_img": 36, "trans_target": 36, "get_list_tuple_img_label": 36, "dsetmnistcolorsolodefault": 36, "merg": 36, "solo": 36, "dsetmnistcolormix": 36, "n_domain": 36, "dsetmnistcolormixnodomainlabel": 36, "bb5af1d77658133af8be8c9b1a13139722315c3a": 36, "l93": 36, "_modul": 36, "datasetfold": 36, "fetch_img_path": 36, "dsetsubfold": 36, "list_class_dir": 36, "target_transform": 36, "is_valid_fil": 36, "path_dir": 36, "class_to_idx": 36, "len": 36, "fstring": 36, "list_tuple_path_cls_ind": 36, "has_file_allowed_extens": 36, "filenam": [36, 39, 41], "lowercas": 36, "dsetinmemdecor": 36, "fun_img_path_loader_default": 36, "discuss": [36, 37], "rgba": 36, "88428": 36, "onehot": 36, "plot_d": 36, "f_name": 36, "batch_siz": [36, 39, 40, 41], "plot_ds_list": 36, "ds_list": 36, "fname": 36, "crope": 36, "flip": [36, 44], "fmcarlucci": 36, "jigendg": 36, "jigsawload": 36, "wrapdsetpatch": 36, "num_perms2classifi": 36, "prob_no_perm": 36, "ppath": 36, "flag_do_not_weave_til": 36, "permuat": 36, "weav": 36, "get_til": 36, "ind_til": 36, "13848": 37, "set_se": 37, "aggwrit": 37, "num_epoch": 37, "expprotocolaggwrit": 37, "expmodelpersistvisitor": 37, "host": 37, "confmat_to_fil": 37, "confmat": 37, "confmat_filenam": 37, "first_lin": 37, "dict_col": 37, "header": [37, 38], "get_col": 37, "column": [37, 38, 41], "get_fpath": 37, "dirnam": 37, "futur": 37, "softwar": [37, 41, 44], "to_fil": 37, "str_line": 37, "suffix": [37, 39], "mk_model_na": 37, "dd_cut": 37, "19": 37, "git": [37, 45], "hash": 37, "model_dir": 37, "model_suffix": 37, "remov": [37, 41], "experiment": 37, "protocol": [37, 44], "filepath": 37, "agg_from_directori": 38, "input_dir": 38, "output_fil": 38, "agg_main": 38, "bm_dir": 38, "skip_plot": 38, "agg_result": [38, 41], "input_fil": 38, "aggregr": 38, "leav": [38, 42], "convert_dict2float": 38, "dict_in": 38, "scientif": 38, "notat": 38, "10000": 38, "load_paramet": 38, "param_fil": 38, "out_fil": 38, "start_se": 38, "misc": [38, 44], "num_gpu": 38, "datafram": [38, 41], "fixm": 38, "kind": [39, 41, 44], "segment": 39, "seq2seq": 39, "than": [39, 40], "reshap": 39, "list_inner_product": 39, "list_loss": 39, "list_multipli": 39, "metric4msel": 39, "multiplier4task_loss": 39, "except": [39, 40], "net_invar_feat": 39, "usag": [39, 41], "reset_feature_extractor": 39, "set_sav": 39, "claasif": 39, "cal_logit_i": 39, "logit": [39, 41], "softmax": 39, "cal_loss_gen_adv": 39, "x_natur": 39, "x_adv": 39, "cal_perf_metr": 39, "modifi": 39, "create_perf_obj": 39, "quieri": 39, "embed": [39, 41], "loader_t": [39, 40, 41], "cal": 39, "infer_y_vpicn": 39, "vpicn": 39, "confid": 39, "match_feat_fun_na": 39, "net_classifi": 39, "pred2fil": 39, "spliter": 39, "dump": [39, 41], "read_prediction_fil": 39, "written": 39, "fiel": 39, "add_args2parser_jigen": 39, "add_args2parser_va": 39, "xyd": 39, "interfacevaexyd": 39, "init_p_zx4batch": 39, "mk_dann": 39, "parent_class": 39, "instanti": [39, 41], "dan": 39, "determin": [39, 41], "modeldan": 39, "list_d_tr": 39, "total_loss": 39, "task_loss": 39, "domain_classification_loss": 39, "net_discrimin": 39, "test_mk_exp_dann": 39, "mk_diva": 39, "vaexydclassif": 39, "subspac": 39, "ils": 39, "maximilian": 39, "modeldiva": 39, "chain_node_build": 39, "enter": 39, "test_mk_exp_diva": 39, "mk_erm": 39, "deepal": 39, "fact": 39, "stem": 39, "neglect": 39, "modelerm": 39, "test_mk_exp_erm": 39, "hierarch": [39, 41, 42], "mk_hduva": 39, "benefit": 39, "incomplet": 39, "preprint": 39, "modelhduva": 39, "gpu": [39, 41], "mk_jigen": 39, "recov": 39, "modeljigen": 39, "list_str_d": 39, "net_classifier_class": 39, "net_classifier_permut": 39, "coeff_reg": 39, "img_class_loss": 39, "perm_task_loss": 39, "prob_permut": 39, "test_mk_exp_jigen": 39, "nodetaskdg": 40, "succ": 40, "conf_without_arg": 40, "get_list_domain": 40, "get_list_domains_tr_t": 40, "tr_id": 40, "te_id": 40, "dg": 40, "id": 40, "complement": 40, "get_na": 40, "na_tr": 40, "na_t": 40, "getter": 40, "list_domain_tr": 40, "loader_v": 40, "set_list_domain": 40, "list_domain": 40, "setter": 40, "_list_domain": 40, "taskclassif": 40, "nodetaskdgclassif": 40, "sample_sav": 40, "subfolder_na": [40, 41], "task_sampl": 40, "nodetaskdict": 40, "decorate_dset": 40, "organ": 40, "get_dset_by_domain": 40, "na_domain": 40, "nodetaskdictclassif": 40, "count_domain_class": 40, "doubl": 40, "class1": 40, "class2": 40, "task_custom": 40, "nodetaskfold": 40, "xx": 40, "yy": 40, "zz": 40, "nodetaskfolderclassnamismatch": 40, "jpeg": 40, "addtask2chain": 40, "nodetaskmnistcolor10": 40, "deafult": 40, "statement": 40, "taskpathlist": 40, "slot": 40, "know": 40, "numer": [40, 41], "nodetaskpathlistdummi": 40, "typedef": 40, "isinst": 40, "mk_node_task_path_list": 40, "img_trans_tr": 40, "dict_class_label_ind2nam": 40, "dict_d2filepath_list_img_tr": 40, "dict_d2filepath_list_img_v": 40, "dict_d2filepath_list_img_t": 40, "parse_domain_id": 40, "list_domain_id": 40, "wrap": 40, "dsetclassvecdecor": 40, "dict_folder_name2class_glob": 40, "fun_class_local_na2vec_new": 40, "straight": 40, "dsetclassvecdecoratorimgpath": 40, "dsetdomainvecdecor": 40, "vec_domain": 40, "dsetdomainvecdecoratorimgpath": 40, "mostli": 40, "loaderdomainlabel": 40, "dim_d": 40, "img_loader2dir": 40, "list_domain_na": 40, "list_class_na": 40, "specul": 40, "assumend": 40, "mk_loader": 40, "bsize": 40, "drop_last": 40, "num_work": 40, "mk_onehot": 40, "tensor1hot2ind": 40, "tensor_label": 40, "dsetinddecorator4xyd": 40, "wih": 40, "dsetzip": 40, "dset1": 40, "dset2": 40, "zip": 40, "getitem": 40, "x_1": 40, "y_1": 40, "x_2": 40, "y_2": 40, "idx": 40, "__get__item": 40, "taskchainnodegett": 40, "flowgenimg": 41, "gen_img_load": 41, "gen_img_xyd": 41, "folder_na": 41, "fun_gen": 41, "output_folder_na": 41, "gen_bencmark_plot": 41, "boxplot": 41, "dataframe_in": 41, "obj1": 41, "obj2": 41, "foldernam": 41, "boxplot_stochast": 41, "boxplot_systemat": 41, "ssystemt": 41, "gen_benchmark_plot": 41, "use_param_index": 41, "restult": 41, "intrepret": 41, "skip_gen": 41, "speed": 41, "max_0_x": 41, "x_arg": 41, "radar_plot": 41, "distinguish_hyperparam": 41, "distinguish_param_setup": 41, "distinguish": 41, "round_vals_in_dict": 41, "df_column_in": 41, "signific": 41, "hyperparam": 41, "scatterplot": 41, "kde": 41, "scatterplot_matrix": 41, "regress": 41, "print_diff": 41, "add_next_param_from_list": 41, "resurs": 41, "grid_task": 41, "param_grid": 41, "grid_df": 41, "recours": 41, "fassion": 41, "itself": 41, "finish": 41, "g_model_na": 41, "add_references_and_check_constraint": 41, "grid_df_prior": 41, "referenced_param": 41, "gridpoint": 41, "satisfi": 41, "add_shared_params_to_param_grid": 41, "shared_df": 41, "dict_param_grid": 41, "build_param_grid_of_shared_param": 41, "back": 41, "frame": 41, "lognormal_grid": 41, "param_config": 41, "loguniform_grid": 41, "normal_grid": 41, "rais_error_if_num_not_specifi": 41, "param_nam": 41, "rais": 41, "consider": 41, "round_to_discreate_grid_norm": 41, "round_to_discreate_grid_uniform": 41, "sample_grid": 41, "sample_gridsearch": 41, "explicitli": 41, "trust": 41, "uniform_grid": 41, "get_gamma_reg": 41, "component_nam": 41, "categoricalhyperparamet": 41, "get_val": 41, "referencehyperparamet": 41, "sampledhyperparamet": 41, "check_constraint": 41, "create_samples_from_shared_sampl": 41, "shared_sampl": 41, "constraind": 41, "pd": 41, "g_method_na": 41, "get_hyperparamet": 41, "get_shared_sampl": 41, "shared_samples_ful": 41, "shared_config_ful": 41, "task_config": 41, "is_dict_with_kei": 41, "input_dict": 41, "sample_hyperparamet": 41, "sample_paramet": 41, "init_param": 41, "shared_config": 41, "10_0000": 41, "attempt": 41, "compli": 41, "sample_task": 41, "num_sampl": 41, "conf_samp": 41, "shared_conf_samp": 41, "sample_task_only_shar": 41, "sample_df": 41, "get_logg": 41, "logger_nam": 41, "logger_185015": 41, "info": 41, "anymor": 41, "interface_class": 41, "method2overrid": 41, "baseclass": 41, "fun": 41, "perfclassif": 41, "classmethod": 41, "cal_acc": 41, "gen_fun_acc": 41, "dim_target": 41, "get_list_pred_target": 41, "model_loc": 41, "went": 41, "perfmetricclassif": 41, "num_class": 41, "agg_precision_recall_f1": 41, "macro": 41, "cal_metr": 41, "sanitycheck": 41, "dataset_sanity_check": 41, "save_san_check_for_domain": 41, "sample_num": 41, "folder_nam": 41, "d_dataset": 41, "destin": 41, "mk_img": 41, "i_ch": 41, "mk_rand_label_onehot": 41, "mk_rand_xyd": 41, "im": 41, "y_dim": 41, "d_dim": 41, "import_path": 41, "build_external_obj_net_module_feat_extract": 41, "mpath": 41, "fine": 41, "problemat": 41, "architectur": 41, "store_arg": 41, "get_label_na": 41, "tensor_ind": 41, "list_str_na": 41, "logit2preds_vp": 41, "mk_dummy_label_list_str": 41, "get_devic": 41, "mk_fun_sav_img": 41, "nrow": 41, "saver": 41, "sav_add_titl": 41, "grid_img": 41, "matplotlib": 41, "pyplot": 41, "subpackag": [42, 43], "submodul": [42, 43], "FOR": 42, "evid": 42, "sanniti": 42, "motiv": 42, "theori": 42, "quantifi": 42, "irm": 42, "irl": 42, "mechan": 42, "page": [42, 44], "content": 43, "usual": 44, "novel": 44, "aim": 44, "site": 44, "cohort": 44, "vendor": 44, "unseen": 44, "platform": 44, "art": 44, "enhanc": 44, "d_1": 44, "d_2": 44, "varphi": 44, "hat": 44, "mu": 44, "inter": 44, "io": 44, "mu_a": 44, "r_a": 44, "mu_b": 44, "r_b": 44, "mu_c": 44, "r_c": 44, "mu_d": 44, "r_d": 44, "pypi": 44, "project": 44, "recommend": 44, "environ": 44, "limit": 44, "tell": 44, "conf": [44, 46], "vlcs_diva_mldg_di": 44, "dann_diva": 44, "mldg_dial": 44, "700000": 44, "100000": 44, "tutori": 44, "your_benchmark_configur": 44, "configuraiton": 44, "augment": 44, "reader": 44, "preprocessinga": 44, "rectangl": 44, "dot": 44, "2403": 44, "14356": 44, "sun2024domainlab": 44, "feistner": 44, "carla": 44, "gossmann": 44, "schwarz": 44, "georg": 44, "umer": 44, "rao": 44, "muhammad": 44, "beer": 44, "lisa": 44, "rockenschaub": 44, "patrick": 44, "shrestha": 44, "rahul": 44, "babu": 44, "gruber": 44, "armin": 44, "chen": 44, "nutan": 44, "2024": 44, "linkifi": 45, "m2r2": 45, "myst": 45, "rinohtyp": 45, "nbsphinx": 45, "nbsphinx_link": 45, "sphinx_materi": 45, "commonmark": 45, "rtfd": 45, "tab": 46, "websit": 46, "rst": 46, "navig": 46, "algoritm": 46}, "objects": {"": [[25, 0, 0, "-", "domainlab"]], "domainlab": [[26, 0, 0, "-", "algos"], [25, 0, 0, "-", "arg_parser"], [25, 0, 0, "-", "cli"], [31, 0, 0, "-", "compos"], [36, 0, 0, "-", "dsets"], [37, 0, 0, "-", "exp"], [38, 0, 0, "-", "exp_protocol"], [25, 5, 1, "", "g_list_model_penalized_reg_agg"], [25, 0, 0, "-", "mk_exp"], [39, 0, 0, "-", "models"], [40, 0, 0, "-", "tasks"], [41, 0, 0, "-", "utils"]], "domainlab.algos": [[26, 0, 0, "-", "a_algo_builder"], [26, 0, 0, "-", "builder_api_model"], [26, 0, 0, "-", "builder_custom"], [26, 0, 0, "-", "builder_dann"], [26, 0, 0, "-", "builder_diva"], [26, 0, 0, "-", "builder_erm"], [26, 0, 0, "-", "builder_hduva"], [26, 0, 0, "-", "builder_jigen1"], [27, 0, 0, "-", "msels"], [28, 0, 0, "-", "observers"], [29, 0, 0, "-", "trainers"], [26, 0, 0, "-", "utils"], [26, 0, 0, "-", "zoo_algos"]], "domainlab.algos.a_algo_builder": [[26, 1, 1, "", "NodeAlgoBuilder"]], "domainlab.algos.a_algo_builder.NodeAlgoBuilder": [[26, 2, 1, "", "extend"], [26, 2, 1, "", "init_business"], [26, 2, 1, "", "init_next_model"], [26, 2, 1, "", "is_myjob"], [26, 3, 1, "", "na_prefix"], [26, 4, 1, "", "name"]], "domainlab.algos.builder_api_model": [[26, 1, 1, "", "NodeAlgoBuilderAPIModel"]], "domainlab.algos.builder_api_model.NodeAlgoBuilderAPIModel": [[26, 2, 1, "", "init_business"]], "domainlab.algos.builder_custom": [[26, 5, 1, "", "make_basic_trainer"]], "domainlab.algos.builder_dann": [[26, 1, 1, "", "NodeAlgoBuilderDANN"]], "domainlab.algos.builder_dann.NodeAlgoBuilderDANN": [[26, 2, 1, "", "init_business"], [26, 2, 1, "", "reset_aux_net"]], "domainlab.algos.builder_diva": [[26, 1, 1, "", "NodeAlgoBuilderDIVA"]], "domainlab.algos.builder_diva.NodeAlgoBuilderDIVA": [[26, 2, 1, "", "get_trainer"], [26, 2, 1, "", "init_business"]], "domainlab.algos.builder_erm": [[26, 1, 1, "", "NodeAlgoBuilderERM"]], "domainlab.algos.builder_erm.NodeAlgoBuilderERM": [[26, 2, 1, "", "init_business"]], "domainlab.algos.builder_hduva": [[26, 1, 1, "", "NodeAlgoBuilderHDUVA"]], "domainlab.algos.builder_hduva.NodeAlgoBuilderHDUVA": [[26, 2, 1, "", "init_business"]], "domainlab.algos.builder_jigen1": [[26, 1, 1, "", "NodeAlgoBuilderJiGen"]], "domainlab.algos.builder_jigen1.NodeAlgoBuilderJiGen": [[26, 2, 1, "", "init_business"]], "domainlab.algos.msels": [[27, 0, 0, "-", "a_model_sel"], [27, 0, 0, "-", "c_msel_oracle"], [27, 0, 0, "-", "c_msel_tr_loss"], [27, 0, 0, "-", "c_msel_val"]], "domainlab.algos.msels.a_model_sel": [[27, 1, 1, "", "AMSel"]], "domainlab.algos.msels.a_model_sel.AMSel": [[27, 2, 1, "", "accept"], [27, 2, 1, "", "base_update"], [27, 4, 1, "", "best_te_metric"], [27, 4, 1, "", "best_val_acc"], [27, 2, 1, "", "early_stop"], [27, 2, 1, "", "if_stop"], [27, 4, 1, "", "max_es"], [27, 4, 1, "", "model_selection_epoch"], [27, 4, 1, "", "observer4msel"], [27, 2, 1, "", "reset"], [27, 4, 1, "", "sel_model_te_acc"], [27, 2, 1, "", "update"], [27, 4, 1, "", "val_threshold"]], "domainlab.algos.msels.c_msel_oracle": [[27, 1, 1, "", "MSelOracleVisitor"]], "domainlab.algos.msels.c_msel_oracle.MSelOracleVisitor": [[27, 2, 1, "", "accept"], [27, 2, 1, "", "base_update"], [27, 2, 1, "", "early_stop"], [27, 4, 1, "", "oracle_last_setpoint_sel_te_acc"]], "domainlab.algos.msels.c_msel_tr_loss": [[27, 1, 1, "", "MSelTrLoss"]], "domainlab.algos.msels.c_msel_tr_loss.MSelTrLoss": [[27, 2, 1, "", "base_update"], [27, 2, 1, "", "early_stop"], [27, 4, 1, "", "max_es"], [27, 2, 1, "", "reset"]], "domainlab.algos.msels.c_msel_val": [[27, 1, 1, "", "MSelValPerf"]], "domainlab.algos.msels.c_msel_val.MSelValPerf": [[27, 2, 1, "", "base_update"], [27, 4, 1, "", "best_te_metric"], [27, 4, 1, "", "best_val_acc"], [27, 2, 1, "", "reset"], [27, 4, 1, "", "sel_model_te_acc"]], "domainlab.algos.observers": [[28, 0, 0, "-", "a_observer"], [28, 0, 0, "-", "b_obvisitor"], [28, 0, 0, "-", "c_obvisitor_cleanup"], [28, 0, 0, "-", "c_obvisitor_gen"]], "domainlab.algos.observers.a_observer": [[28, 1, 1, "", "AObVisitor"]], "domainlab.algos.observers.a_observer.AObVisitor": [[28, 2, 1, "", "accept"], [28, 2, 1, "", "after_all"], [28, 2, 1, "", "clean_up"], [28, 2, 1, "", "reset"], [28, 2, 1, "", "set_task"], [28, 2, 1, "", "update"]], "domainlab.algos.observers.b_obvisitor": [[28, 1, 1, "", "ObVisitor"]], "domainlab.algos.observers.b_obvisitor.ObVisitor": [[28, 2, 1, "", "accept"], [28, 2, 1, "", "after_all"], [28, 2, 1, "", "cal_oracle_perf"], [28, 2, 1, "", "clean_up"], [28, 2, 1, "", "dump_prediction"], [28, 4, 1, "", "str_metric4msel"], [28, 2, 1, "", "update"]], "domainlab.algos.observers.c_obvisitor_cleanup": [[28, 1, 1, "", "ObVisitorCleanUp"]], "domainlab.algos.observers.c_obvisitor_cleanup.ObVisitorCleanUp": [[28, 2, 1, "", "accept"], [28, 2, 1, "", "after_all"], [28, 2, 1, "", "clean_up"], [28, 4, 1, "", "model_sel"], [28, 2, 1, "", "update"]], "domainlab.algos.observers.c_obvisitor_gen": [[28, 1, 1, "", "ObVisitorGen"]], "domainlab.algos.observers.c_obvisitor_gen.ObVisitorGen": [[28, 2, 1, "", "after_all"]], "domainlab.algos.trainers": [[29, 0, 0, "-", "a_trainer"], [29, 0, 0, "-", "args_dial"], [30, 0, 0, "-", "compos"], [29, 0, 0, "-", "hyper_scheduler"], [29, 0, 0, "-", "mmd_base"], [29, 0, 0, "-", "train_basic"], [29, 0, 0, "-", "train_causIRL"], [29, 0, 0, "-", "train_coral"], [29, 0, 0, "-", "train_dial"], [29, 0, 0, "-", "train_ema"], [29, 0, 0, "-", "train_fishr"], [29, 0, 0, "-", "train_hyper_scheduler"], [29, 0, 0, "-", "train_irm"], [29, 0, 0, "-", "train_matchdg"], [29, 0, 0, "-", "train_mldg"], [29, 0, 0, "-", "zoo_trainer"]], "domainlab.algos.trainers.a_trainer": [[29, 1, 1, "", "AbstractTrainer"], [29, 5, 1, "", "mk_opt"]], "domainlab.algos.trainers.a_trainer.AbstractTrainer": [[29, 2, 1, "", "after_batch"], [29, 2, 1, "", "before_batch"], [29, 2, 1, "", "before_tr"], [29, 2, 1, "", "cal_reg_loss"], [29, 2, 1, "", "cal_reg_loss_over_task_loss_ratio"], [29, 4, 1, "", "decoratee"], [29, 2, 1, "", "dset_decoration_args_algo"], [29, 2, 1, "", "extend"], [29, 2, 1, "", "get_model"], [29, 2, 1, "", "init_business"], [29, 2, 1, "", "is_myjob"], [29, 4, 1, "", "list_tr_domain_size"], [29, 4, 1, "", "model"], [29, 4, 1, "", "name"], [29, 4, 1, "", "p_na_prefix"], [29, 2, 1, "", "post_tr"], [29, 2, 1, "", "print_parameters"], [29, 2, 1, "", "reset"], [29, 4, 1, "", "str_metric4msel"], [29, 2, 1, "", "tr_epoch"]], "domainlab.algos.trainers.args_dial": [[29, 5, 1, "", "add_args2parser_dial"]], "domainlab.algos.trainers.compos": [[30, 0, 0, "-", "matchdg_args"], [30, 0, 0, "-", "matchdg_match"], [30, 0, 0, "-", "matchdg_utils"]], "domainlab.algos.trainers.compos.matchdg_args": [[30, 5, 1, "", "add_args2parser_matchdg"]], "domainlab.algos.trainers.compos.matchdg_match": [[30, 1, 1, "", "MatchPair"]], "domainlab.algos.trainers.compos.matchdg_utils": [[30, 1, 1, "", "MatchDictInit"], [30, 1, 1, "", "MatchDictNumDomain2SizeDomain"], [30, 1, 1, "", "MatchDictVirtualRefDset2EachDomain"], [30, 5, 1, "", "dist_cosine_agg"], [30, 5, 1, "", "dist_pairwise_cosine"], [30, 5, 1, "", "fun_tensor_normalize"], [30, 5, 1, "", "get_base_domain_size4match_dg"]], "domainlab.algos.trainers.compos.matchdg_utils.MatchDictInit": [[30, 2, 1, "", "get_num_rows"]], "domainlab.algos.trainers.compos.matchdg_utils.MatchDictNumDomain2SizeDomain": [[30, 2, 1, "", "get_num_rows"]], "domainlab.algos.trainers.compos.matchdg_utils.MatchDictVirtualRefDset2EachDomain": [[30, 2, 1, "", "get_num_rows"]], "domainlab.algos.trainers.hyper_scheduler": [[29, 1, 1, "", "HyperSchedulerWarmupExponential"], [29, 1, 1, "", "HyperSchedulerWarmupLinear"]], "domainlab.algos.trainers.hyper_scheduler.HyperSchedulerWarmupExponential": [[29, 2, 1, "", "warmup"]], "domainlab.algos.trainers.hyper_scheduler.HyperSchedulerWarmupLinear": [[29, 2, 1, "", "set_steps"], [29, 2, 1, "", "warmup"]], "domainlab.algos.trainers.mmd_base": [[29, 1, 1, "", "TrainerMMDBase"]], "domainlab.algos.trainers.mmd_base.TrainerMMDBase": [[29, 2, 1, "", "gaussian_kernel"], [29, 2, 1, "", "mmd"], [29, 2, 1, "", "my_cdist"]], "domainlab.algos.trainers.train_basic": [[29, 1, 1, "", "TrainerBasic"], [29, 5, 1, "", "list_divide"]], "domainlab.algos.trainers.train_basic.TrainerBasic": [[29, 2, 1, "", "after_epoch"], [29, 2, 1, "", "before_epoch"], [29, 2, 1, "", "before_tr"], [29, 2, 1, "", "cal_loss"], [29, 2, 1, "", "log_loss"], [29, 2, 1, "", "tr_batch"], [29, 2, 1, "", "tr_epoch"]], "domainlab.algos.trainers.train_causIRL": [[29, 1, 1, "", "TrainerCausalIRL"]], "domainlab.algos.trainers.train_causIRL.TrainerCausalIRL": [[29, 2, 1, "", "gaussian_kernel"], [29, 2, 1, "", "mmd"], [29, 2, 1, "", "my_cdist"], [29, 2, 1, "", "tr_batch"]], "domainlab.algos.trainers.train_coral": [[29, 1, 1, "", "TrainerCoral"]], "domainlab.algos.trainers.train_coral.TrainerCoral": [[29, 2, 1, "", "cross_domain_mmd"], [29, 2, 1, "", "tr_epoch"]], "domainlab.algos.trainers.train_dial": [[29, 1, 1, "", "TrainerDIAL"]], "domainlab.algos.trainers.train_dial.TrainerDIAL": [[29, 2, 1, "", "gen_adversarial"]], "domainlab.algos.trainers.train_ema": [[29, 1, 1, "", "TrainerMA"]], "domainlab.algos.trainers.train_ema.TrainerMA": [[29, 2, 1, "", "after_epoch"], [29, 2, 1, "", "move_average"]], "domainlab.algos.trainers.train_fishr": [[29, 1, 1, "", "TrainerFishr"]], "domainlab.algos.trainers.train_fishr.TrainerFishr": [[29, 2, 1, "", "cal_dict_variance_grads"], [29, 2, 1, "", "cal_mean_across_dict"], [29, 2, 1, "", "cal_power_single_dict"], [29, 2, 1, "", "tr_epoch"], [29, 2, 1, "", "var_grads_and_loss"], [29, 2, 1, "", "variance_between_dict"]], "domainlab.algos.trainers.train_hyper_scheduler": [[29, 1, 1, "", "TrainerHyperScheduler"]], "domainlab.algos.trainers.train_hyper_scheduler.TrainerHyperScheduler": [[29, 2, 1, "", "before_batch"], [29, 2, 1, "", "before_tr"], [29, 2, 1, "", "set_scheduler"], [29, 2, 1, "", "tr_epoch"]], "domainlab.algos.trainers.train_irm": [[29, 1, 1, "", "TrainerIRM"]], "domainlab.algos.trainers.train_irm.TrainerIRM": [[29, 2, 1, "", "tr_epoch"]], "domainlab.algos.trainers.train_matchdg": [[29, 1, 1, "", "TrainerMatchDG"], [29, 5, 1, "", "match_tensor_reshape"]], "domainlab.algos.trainers.train_matchdg.TrainerMatchDG": [[29, 2, 1, "", "before_tr"], [29, 2, 1, "", "dset_decoration_args_algo"], [29, 2, 1, "", "init_business"], [29, 2, 1, "", "mk_match_tensor"], [29, 2, 1, "", "tr_batch"], [29, 2, 1, "", "tr_epoch"]], "domainlab.algos.trainers.train_mldg": [[29, 1, 1, "", "TrainerMLDG"]], "domainlab.algos.trainers.train_mldg.TrainerMLDG": [[29, 2, 1, "", "before_tr"], [29, 2, 1, "", "prepare_ziped_loader"], [29, 2, 1, "", "tr_epoch"]], "domainlab.algos.trainers.zoo_trainer": [[29, 1, 1, "", "TrainerChainNodeGetter"]], "domainlab.algos.utils": [[26, 5, 1, "", "split_net_feat_last"]], "domainlab.algos.zoo_algos": [[26, 1, 1, "", "AlgoBuilderChainNodeGetter"]], "domainlab.algos.zoo_algos.AlgoBuilderChainNodeGetter": [[26, 2, 1, "", "register_external_node"]], "domainlab.arg_parser": [[25, 1, 1, "", "ParseValuesOrKeyValuePairs"], [25, 5, 1, "", "apply_dict_to_args"], [25, 5, 1, "", "mk_parser_main"], [25, 5, 1, "", "parse_cmd_args"]], "domainlab.cli": [[25, 5, 1, "", "domainlab_cli"]], "domainlab.compos": [[31, 0, 0, "-", "a_nn_builder"], [31, 0, 0, "-", "builder_nn_alex"], [31, 0, 0, "-", "builder_nn_conv_bn_relu_2"], [31, 0, 0, "-", "builder_nn_external_from_file"], [32, 0, 0, "-", "nn_zoo"], [33, 0, 0, "-", "pcr"], [31, 0, 0, "-", "utils_conv_get_flat_dim"], [34, 0, 0, "-", "vae"], [31, 0, 0, "-", "zoo_nn"]], "domainlab.compos.a_nn_builder": [[31, 1, 1, "", "AbstractFeatExtractNNBuilderChainNode"]], "domainlab.compos.a_nn_builder.AbstractFeatExtractNNBuilderChainNode": [[31, 2, 1, "", "init_business"], [31, 2, 1, "", "is_myjob"]], "domainlab.compos.builder_nn_alex": [[31, 5, 1, "", "mkNodeFeatExtractNNBuilderNameAlex"]], "domainlab.compos.builder_nn_conv_bn_relu_2": [[31, 5, 1, "", "mkNodeFeatExtractNNBuilderNameConvBnRelu2"]], "domainlab.compos.builder_nn_external_from_file": [[31, 5, 1, "", "mkNodeFeatExtractNNBuilderExternFromFile"]], "domainlab.compos.nn_zoo": [[32, 0, 0, "-", "net_adversarial"], [32, 0, 0, "-", "net_classif"], [32, 0, 0, "-", "net_conv_conv_bn_pool_2"], [32, 0, 0, "-", "net_gated"], [32, 0, 0, "-", "nn"], [32, 0, 0, "-", "nn_alex"], [32, 0, 0, "-", "nn_torchvision"]], "domainlab.compos.nn_zoo.net_adversarial": [[32, 1, 1, "", "AutoGradFunMultiply"], [32, 1, 1, "", "AutoGradFunReverseMultiply"], [32, 1, 1, "", "Flatten"]], "domainlab.compos.nn_zoo.net_adversarial.AutoGradFunMultiply": [[32, 2, 1, "", "backward"], [32, 2, 1, "", "forward"]], "domainlab.compos.nn_zoo.net_adversarial.AutoGradFunReverseMultiply": [[32, 2, 1, "", "backward"], [32, 2, 1, "", "forward"]], "domainlab.compos.nn_zoo.net_adversarial.Flatten": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.net_classif": [[32, 1, 1, "", "ClassifDropoutReluLinear"]], "domainlab.compos.nn_zoo.net_classif.ClassifDropoutReluLinear": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2": [[32, 1, 1, "", "NetConvBnReluPool2L"], [32, 1, 1, "", "NetConvDense"], [32, 5, 1, "", "mk_conv_bn_relu_pool"]], "domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2.NetConvBnReluPool2L": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2.NetConvDense": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.net_gated": [[32, 1, 1, "", "Conv2d"], [32, 1, 1, "", "GatedConv2d"], [32, 1, 1, "", "GatedDense"]], "domainlab.compos.nn_zoo.net_gated.Conv2d": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.net_gated.GatedConv2d": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.net_gated.GatedDense": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.nn": [[32, 1, 1, "", "DenseNet"], [32, 1, 1, "", "LayerId"]], "domainlab.compos.nn_zoo.nn.DenseNet": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.nn.LayerId": [[32, 2, 1, "", "forward"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.nn_alex": [[32, 1, 1, "", "Alex4DeepAll"], [32, 1, 1, "", "AlexNetBase"], [32, 1, 1, "", "AlexNetNoLastLayer"]], "domainlab.compos.nn_zoo.nn_alex.Alex4DeepAll": [[32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.nn_alex.AlexNetBase": [[32, 2, 1, "", "fetch_net"], [32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.nn_alex.AlexNetNoLastLayer": [[32, 3, 1, "", "training"]], "domainlab.compos.nn_zoo.nn_torchvision": [[32, 1, 1, "", "NetTorchVisionBase"]], "domainlab.compos.nn_zoo.nn_torchvision.NetTorchVisionBase": [[32, 2, 1, "", "fetch_net"], [32, 2, 1, "", "forward"], [32, 2, 1, "", "show"], [32, 3, 1, "", "training"]], "domainlab.compos.pcr": [[33, 0, 0, "-", "p_chain_handler"], [33, 0, 0, "-", "request"]], "domainlab.compos.pcr.p_chain_handler": [[33, 1, 1, "", "AbstractChainNodeHandler"], [33, 1, 1, "", "DummyBusiness"], [33, 1, 1, "", "DummyChainNodeHandlerBeaver"], [33, 1, 1, "", "DummyChainNodeHandlerLazy"], [33, 1, 1, "", "Request4Chain"]], "domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler": [[33, 2, 1, "", "handle"], [33, 2, 1, "", "init_business"], [33, 2, 1, "", "is_myjob"], [33, 2, 1, "", "print_options"], [33, 2, 1, "", "set_parent"]], "domainlab.compos.pcr.p_chain_handler.DummyBusiness": [[33, 3, 1, "", "message"]], "domainlab.compos.pcr.p_chain_handler.DummyChainNodeHandlerBeaver": [[33, 2, 1, "", "init_business"], [33, 2, 1, "", "is_myjob"]], "domainlab.compos.pcr.p_chain_handler.DummyChainNodeHandlerLazy": [[33, 2, 1, "", "init_business"], [33, 2, 1, "", "is_myjob"]], "domainlab.compos.pcr.p_chain_handler.Request4Chain": [[33, 2, 1, "", "convert"]], "domainlab.compos.pcr.request": [[33, 1, 1, "", "RequestArgs2ExpCmd"], [33, 1, 1, "", "RequestTask"], [33, 1, 1, "", "RequestVAEBuilderCHW"], [33, 1, 1, "", "RequestVAEBuilderNN"]], "domainlab.compos.utils_conv_get_flat_dim": [[31, 5, 1, "", "get_flat_dim"]], "domainlab.compos.vae": [[34, 0, 0, "-", "a_model_builder"], [34, 0, 0, "-", "a_vae_builder"], [34, 0, 0, "-", "c_vae_adaptor_model_recon"], [34, 0, 0, "-", "c_vae_builder_classif"], [34, 0, 0, "-", "c_vae_recon"], [35, 0, 0, "-", "compos"], [34, 0, 0, "-", "utils_request_chain_builder"], [34, 0, 0, "-", "zoo_vae_builders_classif"], [34, 0, 0, "-", "zoo_vae_builders_classif_topic"]], "domainlab.compos.vae.a_model_builder": [[34, 1, 1, "", "AbstractModelBuilderChainNode"]], "domainlab.compos.vae.a_model_builder.AbstractModelBuilderChainNode": [[34, 2, 1, "", "config"]], "domainlab.compos.vae.a_vae_builder": [[34, 1, 1, "", "AbstractVAEBuilderChainNode"]], "domainlab.compos.vae.a_vae_builder.AbstractVAEBuilderChainNode": [[34, 2, 1, "", "build_decoder"], [34, 2, 1, "", "build_encoder"], [34, 2, 1, "", "init_business"]], "domainlab.compos.vae.c_vae_adaptor_model_recon": [[34, 1, 1, "", "AdaptorReconVAEXYD"]], "domainlab.compos.vae.c_vae_adaptor_model_recon.AdaptorReconVAEXYD": [[34, 2, 1, "", "cal_latent"], [34, 2, 1, "", "cal_prior_zd"], [34, 2, 1, "", "cal_prior_zy"], [34, 2, 1, "", "recon_ydx"]], "domainlab.compos.vae.c_vae_builder_classif": [[34, 1, 1, "", "ChainNodeVAEBuilderClassifCondPrior"]], "domainlab.compos.vae.c_vae_builder_classif.ChainNodeVAEBuilderClassifCondPrior": [[34, 2, 1, "", "build_decoder"], [34, 2, 1, "", "build_encoder"], [34, 2, 1, "", "construct_classifier"], [34, 2, 1, "", "construct_cond_prior"], [34, 2, 1, "", "is_myjob"]], "domainlab.compos.vae.c_vae_recon": [[34, 1, 1, "", "ReconVAEXYD"]], "domainlab.compos.vae.c_vae_recon.ReconVAEXYD": [[34, 2, 1, "", "recon"], [34, 2, 1, "", "recon_cf"]], "domainlab.compos.vae.compos": [[35, 0, 0, "-", "decoder_concat_vec_reshape_conv"], [35, 0, 0, "-", "decoder_concat_vec_reshape_conv_gated_conv"], [35, 0, 0, "-", "decoder_cond_prior"], [35, 0, 0, "-", "decoder_losses"], [35, 0, 0, "-", "encoder"], [35, 0, 0, "-", "encoder_dirichlet"], [35, 0, 0, "-", "encoder_domain_topic"], [35, 0, 0, "-", "encoder_domain_topic_img2topic"], [35, 0, 0, "-", "encoder_domain_topic_img_topic2zd"], [35, 0, 0, "-", "encoder_xyd_parallel"], [35, 0, 0, "-", "encoder_xydt_elevator"], [35, 0, 0, "-", "encoder_zy"]], "domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv": [[35, 1, 1, "", "DecoderConcatLatentFcReshapeConv"]], "domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv.DecoderConcatLatentFcReshapeConv": [[35, 2, 1, "", "cal_p_x_pars_loc_scale"], [35, 2, 1, "", "concat_ydx"], [35, 2, 1, "", "concat_ytdx"], [35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv_gated_conv": [[35, 1, 1, "", "DecoderConcatLatentFCReshapeConvGatedConv"]], "domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv_gated_conv.DecoderConcatLatentFCReshapeConvGatedConv": [[35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.decoder_cond_prior": [[35, 1, 1, "", "LSCondPriorLinearBnReluLinearSoftPlus"]], "domainlab.compos.vae.compos.decoder_cond_prior.LSCondPriorLinearBnReluLinearSoftPlus": [[35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.decoder_losses": [[35, 1, 1, "", "NLLPixelLogistic256"]], "domainlab.compos.vae.compos.encoder": [[35, 1, 1, "", "LSEncoderConvBnReluPool"], [35, 1, 1, "", "LSEncoderLinear"]], "domainlab.compos.vae.compos.encoder.LSEncoderConvBnReluPool": [[35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder.LSEncoderLinear": [[35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_dirichlet": [[35, 1, 1, "", "EncoderH2Dirichlet"]], "domainlab.compos.vae.compos.encoder_dirichlet.EncoderH2Dirichlet": [[35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_domain_topic": [[35, 1, 1, "", "EncoderImg2TopicDirZd"]], "domainlab.compos.vae.compos.encoder_domain_topic.EncoderImg2TopicDirZd": [[35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_domain_topic_img2topic": [[35, 1, 1, "", "EncoderImg2TopicDistri"]], "domainlab.compos.vae.compos.encoder_domain_topic_img2topic.EncoderImg2TopicDistri": [[35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_domain_topic_img_topic2zd": [[35, 1, 1, "", "EncoderSandwichTopicImg2Zd"]], "domainlab.compos.vae.compos.encoder_domain_topic_img_topic2zd.EncoderSandwichTopicImg2Zd": [[35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_xyd_parallel": [[35, 1, 1, "", "XYDEncoderParallel"], [35, 1, 1, "", "XYDEncoderParallelAlex"], [35, 1, 1, "", "XYDEncoderParallelConvBnReluPool"], [35, 1, 1, "", "XYDEncoderParallelExtern"], [35, 1, 1, "", "XYDEncoderParallelUser"]], "domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallel": [[35, 2, 1, "", "forward"], [35, 2, 1, "", "infer_zy_loc"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallelAlex": [[35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallelConvBnReluPool": [[35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallelExtern": [[35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallelUser": [[35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_xydt_elevator": [[35, 1, 1, "", "XYDTEncoderArg"], [35, 1, 1, "", "XYDTEncoderElevator"]], "domainlab.compos.vae.compos.encoder_xydt_elevator.XYDTEncoderArg": [[35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_xydt_elevator.XYDTEncoderElevator": [[35, 2, 1, "", "forward"], [35, 2, 1, "", "infer_zy_loc"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.compos.encoder_zy": [[35, 1, 1, "", "EncoderConnectLastFeatLayer2Z"]], "domainlab.compos.vae.compos.encoder_zy.EncoderConnectLastFeatLayer2Z": [[35, 2, 1, "", "forward"], [35, 3, 1, "", "training"]], "domainlab.compos.vae.utils_request_chain_builder": [[34, 1, 1, "", "VAEChainNodeGetter"]], "domainlab.compos.vae.zoo_vae_builders_classif": [[34, 1, 1, "", "ChainNodeVAEBuilderClassifCondPriorBase"], [34, 1, 1, "", "NodeVAEBuilderArg"], [34, 1, 1, "", "NodeVAEBuilderImgAlex"], [34, 1, 1, "", "NodeVAEBuilderImgConvBnPool"], [34, 1, 1, "", "NodeVAEBuilderUser"]], "domainlab.compos.vae.zoo_vae_builders_classif.ChainNodeVAEBuilderClassifCondPriorBase": [[34, 2, 1, "", "build_decoder"], [34, 2, 1, "", "build_encoder"], [34, 2, 1, "", "config_img"], [34, 2, 1, "", "is_myjob"]], "domainlab.compos.vae.zoo_vae_builders_classif.NodeVAEBuilderArg": [[34, 2, 1, "", "build_encoder"], [34, 2, 1, "", "is_myjob"]], "domainlab.compos.vae.zoo_vae_builders_classif.NodeVAEBuilderImgAlex": [[34, 2, 1, "", "build_encoder"], [34, 2, 1, "", "is_myjob"]], "domainlab.compos.vae.zoo_vae_builders_classif.NodeVAEBuilderImgConvBnPool": [[34, 2, 1, "", "build_encoder"], [34, 2, 1, "", "is_myjob"]], "domainlab.compos.vae.zoo_vae_builders_classif.NodeVAEBuilderUser": [[34, 2, 1, "", "build_encoder"], [34, 2, 1, "", "is_myjob"]], "domainlab.compos.vae.zoo_vae_builders_classif_topic": [[34, 1, 1, "", "NodeVAEBuilderImgTopic"]], "domainlab.compos.vae.zoo_vae_builders_classif_topic.NodeVAEBuilderImgTopic": [[34, 2, 1, "", "build_decoder"], [34, 2, 1, "", "build_encoder"], [34, 2, 1, "", "is_myjob"]], "domainlab.compos.zoo_nn": [[31, 1, 1, "", "FeatExtractNNBuilderChainNodeGetter"]], "domainlab.dsets": [[36, 0, 0, "-", "a_dset_mnist_color_rgb_solo"], [36, 0, 0, "-", "dset_img_path_list"], [36, 0, 0, "-", "dset_mnist_color_solo_default"], [36, 0, 0, "-", "dset_poly_domains_mnist_color_default"], [36, 0, 0, "-", "dset_subfolder"], [36, 0, 0, "-", "utils_color_palette"], [36, 0, 0, "-", "utils_data"], [36, 0, 0, "-", "utils_wrapdset_patches"]], "domainlab.dsets.a_dset_mnist_color_rgb_solo": [[36, 1, 1, "", "ADsetMNISTColorRGBSolo"]], "domainlab.dsets.a_dset_mnist_color_rgb_solo.ADsetMNISTColorRGBSolo": [[36, 2, 1, "", "get_background_color"], [36, 2, 1, "", "get_foreground_color"], [36, 2, 1, "", "get_num_colors"]], "domainlab.dsets.dset_img_path_list": [[36, 1, 1, "", "DsetImPathList"]], "domainlab.dsets.dset_img_path_list.DsetImPathList": [[36, 2, 1, "", "get_list_tuple_img_label"]], "domainlab.dsets.dset_mnist_color_solo_default": [[36, 1, 1, "", "DsetMNISTColorSoloDefault"]], "domainlab.dsets.dset_mnist_color_solo_default.DsetMNISTColorSoloDefault": [[36, 2, 1, "", "get_background_color"], [36, 2, 1, "", "get_foreground_color"], [36, 2, 1, "", "get_num_colors"], [36, 4, 1, "", "palette"]], "domainlab.dsets.dset_poly_domains_mnist_color_default": [[36, 1, 1, "", "DsetMNISTColorMix"], [36, 1, 1, "", "DsetMNISTColorMixNoDomainLabel"]], "domainlab.dsets.dset_subfolder": [[36, 1, 1, "", "DsetSubFolder"], [36, 5, 1, "", "fetch_img_paths"], [36, 5, 1, "", "has_file_allowed_extension"]], "domainlab.dsets.utils_data": [[36, 1, 1, "", "DsetInMemDecorator"], [36, 5, 1, "", "fun_img_path_loader_default"], [36, 5, 1, "", "mk_fun_label2onehot"], [36, 5, 1, "", "plot_ds"], [36, 5, 1, "", "plot_ds_list"]], "domainlab.dsets.utils_wrapdset_patches": [[36, 1, 1, "", "WrapDsetPatches"]], "domainlab.dsets.utils_wrapdset_patches.WrapDsetPatches": [[36, 2, 1, "", "get_tile"]], "domainlab.exp": [[37, 0, 0, "-", "exp_cuda_seed"], [37, 0, 0, "-", "exp_main"], [37, 0, 0, "-", "exp_utils"]], "domainlab.exp.exp_cuda_seed": [[37, 5, 1, "", "set_seed"]], "domainlab.exp.exp_main": [[37, 1, 1, "", "Exp"]], "domainlab.exp.exp_main.Exp": [[37, 2, 1, "", "clean_up"], [37, 2, 1, "", "execute"]], "domainlab.exp.exp_utils": [[37, 1, 1, "", "AggWriter"], [37, 1, 1, "", "ExpModelPersistVisitor"], [37, 1, 1, "", "ExpProtocolAggWriter"]], "domainlab.exp.exp_utils.AggWriter": [[37, 2, 1, "", "confmat_to_file"], [37, 2, 1, "", "first_line"], [37, 2, 1, "", "get_cols"], [37, 2, 1, "", "get_fpath"], [37, 2, 1, "", "to_file"]], "domainlab.exp.exp_utils.ExpModelPersistVisitor": [[37, 2, 1, "", "clean_up"], [37, 2, 1, "", "load"], [37, 2, 1, "", "mk_model_na"], [37, 3, 1, "", "model_dir"], [37, 3, 1, "", "model_suffix"], [37, 2, 1, "", "remove"], [37, 2, 1, "", "save"]], "domainlab.exp.exp_utils.ExpProtocolAggWriter": [[37, 2, 1, "", "confmat_to_file"], [37, 2, 1, "", "get_cols"], [37, 2, 1, "", "get_fpath"]], "domainlab.exp_protocol": [[38, 0, 0, "-", "aggregate_results"], [38, 0, 0, "-", "run_experiment"]], "domainlab.exp_protocol.aggregate_results": [[38, 5, 1, "", "agg_from_directory"], [38, 5, 1, "", "agg_main"], [38, 5, 1, "", "agg_results"]], "domainlab.exp_protocol.run_experiment": [[38, 5, 1, "", "convert_dict2float"], [38, 5, 1, "", "load_parameters"], [38, 5, 1, "", "run_experiment"]], "domainlab.mk_exp": [[25, 5, 1, "", "mk_exp"]], "domainlab.models": [[39, 0, 0, "-", "a_model"], [39, 0, 0, "-", "a_model_classif"], [39, 0, 0, "-", "args_jigen"], [39, 0, 0, "-", "args_vae"], [39, 0, 0, "-", "interface_vae_xyd"], [39, 0, 0, "-", "model_custom"], [39, 0, 0, "-", "model_dann"], [39, 0, 0, "-", "model_diva"], [39, 0, 0, "-", "model_erm"], [39, 0, 0, "-", "model_hduva"], [39, 0, 0, "-", "model_jigen"], [39, 0, 0, "-", "model_vae_xyd_classif"]], "domainlab.models.a_model": [[39, 1, 1, "", "AModel"]], "domainlab.models.a_model.AModel": [[39, 2, 1, "", "cal_loss"], [39, 2, 1, "", "cal_reg_loss"], [39, 2, 1, "", "cal_task_loss"], [39, 2, 1, "", "dset_decoration_args_algo"], [39, 2, 1, "", "extend"], [39, 2, 1, "", "extract_semantic_feat"], [39, 2, 1, "", "forward"], [39, 2, 1, "", "list_inner_product"], [39, 2, 1, "", "load"], [39, 4, 1, "", "metric4msel"], [39, 4, 1, "", "multiplier4task_loss"], [39, 4, 1, "", "name"], [39, 4, 1, "", "net_invar_feat"], [39, 4, 1, "", "p_na_prefix"], [39, 2, 1, "", "print_parameters"], [39, 2, 1, "", "reset_aux_net"], [39, 2, 1, "", "reset_feature_extractor"], [39, 2, 1, "", "save"], [39, 2, 1, "", "set_saver"], [39, 3, 1, "", "training"]], "domainlab.models.a_model_classif": [[39, 1, 1, "", "AModelClassif"]], "domainlab.models.a_model_classif.AModelClassif": [[39, 2, 1, "", "cal_logit_y"], [39, 2, 1, "", "cal_loss_gen_adv"], [39, 2, 1, "", "cal_perf_metric"], [39, 2, 1, "", "cal_task_loss"], [39, 2, 1, "", "create_perf_obj"], [39, 4, 1, "", "dim_y"], [39, 2, 1, "", "evaluate"], [39, 2, 1, "", "extend"], [39, 2, 1, "", "extract_semantic_feat"], [39, 2, 1, "", "infer_y_vpicn"], [39, 3, 1, "", "match_feat_fun_na"], [39, 4, 1, "", "metric4msel"], [39, 4, 1, "", "net_classifier"], [39, 2, 1, "", "pred2file"], [39, 2, 1, "", "read_prediction_file"], [39, 3, 1, "", "training"]], "domainlab.models.args_jigen": [[39, 5, 1, "", "add_args2parser_jigen"]], "domainlab.models.args_vae": [[39, 5, 1, "", "add_args2parser_vae"]], "domainlab.models.interface_vae_xyd": [[39, 1, 1, "", "InterfaceVAEXYD"]], "domainlab.models.interface_vae_xyd.InterfaceVAEXYD": [[39, 2, 1, "", "init"], [39, 2, 1, "", "init_p_zx4batch"]], "domainlab.models.model_custom": [[39, 1, 1, "", "AModelCustom"]], "domainlab.models.model_custom.AModelCustom": [[39, 2, 1, "", "dict_net_module_na2arg_na"], [39, 3, 1, "", "training"]], "domainlab.models.model_dann": [[39, 5, 1, "", "mk_dann"]], "domainlab.models.model_diva": [[39, 5, 1, "", "mk_diva"]], "domainlab.models.model_erm": [[39, 5, 1, "", "mk_erm"]], "domainlab.models.model_hduva": [[39, 5, 1, "", "mk_hduva"]], "domainlab.models.model_jigen": [[39, 5, 1, "", "mk_jigen"]], "domainlab.models.model_vae_xyd_classif": [[39, 1, 1, "", "VAEXYDClassif"]], "domainlab.models.model_vae_xyd_classif.VAEXYDClassif": [[39, 2, 1, "", "extract_semantic_feat"], [39, 4, 1, "", "multiplier4task_loss"], [39, 3, 1, "", "training"]], "domainlab.tasks": [[40, 0, 0, "-", "a_task"], [40, 0, 0, "-", "a_task_classif"], [40, 0, 0, "-", "b_task"], [40, 0, 0, "-", "b_task_classif"], [40, 0, 0, "-", "task_dset"], [40, 0, 0, "-", "task_folder"], [40, 0, 0, "-", "task_folder_mk"], [40, 0, 0, "-", "task_mini_vlcs"], [40, 0, 0, "-", "task_mnist_color"], [40, 0, 0, "-", "task_pathlist"], [40, 0, 0, "-", "task_utils"], [40, 0, 0, "-", "utils_task"], [40, 0, 0, "-", "utils_task_dset"], [40, 0, 0, "-", "zoo_tasks"]], "domainlab.tasks.a_task": [[40, 1, 1, "", "NodeTaskDG"]], "domainlab.tasks.a_task.NodeTaskDG": [[40, 2, 1, "", "conf_without_args"], [40, 2, 1, "", "get_list_domains"], [40, 2, 1, "", "get_list_domains_tr_te"], [40, 2, 1, "", "get_na"], [40, 2, 1, "", "init_business"], [40, 2, 1, "", "is_myjob"], [40, 4, 1, "", "isize"], [40, 4, 1, "", "list_domain_tr"], [40, 4, 1, "", "loader_te"], [40, 4, 1, "", "loader_tr"], [40, 4, 1, "", "loader_val"], [40, 2, 1, "", "set_list_domains"], [40, 4, 1, "", "task_name"]], "domainlab.tasks.a_task_classif": [[40, 1, 1, "", "NodeTaskDGClassif"]], "domainlab.tasks.a_task_classif.NodeTaskDGClassif": [[40, 4, 1, "", "dim_y"], [40, 4, 1, "", "list_str_y"], [40, 2, 1, "", "sample_sav"]], "domainlab.tasks.b_task": [[40, 1, 1, "", "NodeTaskDict"]], "domainlab.tasks.b_task.NodeTaskDict": [[40, 2, 1, "", "decorate_dset"], [40, 2, 1, "", "get_dset_by_domain"], [40, 2, 1, "", "init_business"]], "domainlab.tasks.b_task_classif": [[40, 1, 1, "", "NodeTaskDictClassif"]], "domainlab.tasks.b_task_classif.NodeTaskDictClassif": [[40, 2, 1, "", "count_domain_class"], [40, 2, 1, "", "init_business"]], "domainlab.tasks.task_dset": [[40, 5, 1, "", "mk_task_dset"]], "domainlab.tasks.task_folder": [[40, 1, 1, "", "NodeTaskFolder"], [40, 1, 1, "", "NodeTaskFolderClassNaMismatch"]], "domainlab.tasks.task_folder.NodeTaskFolder": [[40, 4, 1, "", "dict_domain2imgroot"], [40, 4, 1, "", "extensions"], [40, 2, 1, "", "get_dset_by_domain"]], "domainlab.tasks.task_folder.NodeTaskFolderClassNaMismatch": [[40, 2, 1, "", "get_dset_by_domain"]], "domainlab.tasks.task_folder_mk": [[40, 5, 1, "", "mk_task_folder"]], "domainlab.tasks.task_mini_vlcs": [[40, 5, 1, "", "addtask2chain"]], "domainlab.tasks.task_mnist_color": [[40, 1, 1, "", "NodeTaskMNISTColor10"]], "domainlab.tasks.task_mnist_color.NodeTaskMNISTColor10": [[40, 2, 1, "", "get_dset_by_domain"], [40, 2, 1, "", "get_list_domains"], [40, 4, 1, "", "isize"], [40, 4, 1, "", "list_str_y"]], "domainlab.tasks.task_pathlist": [[40, 1, 1, "", "NodeTaskPathListDummy"], [40, 5, 1, "", "mk_node_task_path_list"]], "domainlab.tasks.task_pathlist.NodeTaskPathListDummy": [[40, 2, 1, "", "get_dset_by_domain"]], "domainlab.tasks.task_utils": [[40, 5, 1, "", "parse_domain_id"]], "domainlab.tasks.utils_task": [[40, 1, 1, "", "DsetClassVecDecorator"], [40, 1, 1, "", "DsetClassVecDecoratorImgPath"], [40, 1, 1, "", "DsetDomainVecDecorator"], [40, 1, 1, "", "DsetDomainVecDecoratorImgPath"], [40, 1, 1, "", "ImSize"], [40, 1, 1, "", "LoaderDomainLabel"], [40, 5, 1, "", "img_loader2dir"], [40, 5, 1, "", "mk_loader"], [40, 5, 1, "", "mk_onehot"], [40, 5, 1, "", "tensor1hot2ind"]], "domainlab.tasks.utils_task.DsetClassVecDecorator": [[40, 2, 1, "", "fun_class_local_na2vec_new"], [40, 4, 1, "", "targets"]], "domainlab.tasks.utils_task.DsetDomainVecDecorator": [[40, 4, 1, "", "targets"]], "domainlab.tasks.utils_task.ImSize": [[40, 4, 1, "", "c"], [40, 4, 1, "", "h"], [40, 4, 1, "", "w"]], "domainlab.tasks.utils_task_dset": [[40, 1, 1, "", "DsetIndDecorator4XYD"], [40, 1, 1, "", "DsetZip"]], "domainlab.tasks.zoo_tasks": [[40, 1, 1, "", "TaskChainNodeGetter"], [40, 5, 1, "", "get_task"]], "domainlab.utils": [[41, 0, 0, "-", "flows_gen_img_model"], [41, 0, 0, "-", "generate_benchmark_plots"], [41, 0, 0, "-", "get_git_tag"], [41, 0, 0, "-", "hyperparameter_gridsearch"], [41, 0, 0, "-", "hyperparameter_retrieval"], [41, 0, 0, "-", "hyperparameter_sampling"], [41, 0, 0, "-", "logger"], [41, 0, 0, "-", "override_interface"], [41, 0, 0, "-", "perf"], [41, 0, 0, "-", "perf_metrics"], [41, 0, 0, "-", "sanity_check"], [41, 0, 0, "-", "test_img"], [41, 0, 0, "-", "u_import"], [41, 0, 0, "-", "u_import_net_module"], [41, 0, 0, "-", "utils_class"], [41, 0, 0, "-", "utils_classif"], [41, 0, 0, "-", "utils_cuda"], [41, 0, 0, "-", "utils_img_sav"]], "domainlab.utils.flows_gen_img_model": [[41, 1, 1, "", "FlowGenImgs"], [41, 5, 1, "", "fun_gen"]], "domainlab.utils.flows_gen_img_model.FlowGenImgs": [[41, 2, 1, "", "gen_img_loader"], [41, 2, 1, "", "gen_img_xyd"]], "domainlab.utils.generate_benchmark_plots": [[41, 5, 1, "", "boxplot"], [41, 5, 1, "", "boxplot_stochastic"], [41, 5, 1, "", "boxplot_systematic"], [41, 5, 1, "", "gen_benchmark_plots"], [41, 5, 1, "", "gen_plots"], [41, 5, 1, "", "max_0_x"], [41, 5, 1, "", "radar_plot"], [41, 5, 1, "", "round_vals_in_dict"], [41, 5, 1, "", "scatterplot"], [41, 5, 1, "", "scatterplot_matrix"]], "domainlab.utils.get_git_tag": [[41, 5, 1, "", "get_git_tag"]], "domainlab.utils.hyperparameter_gridsearch": [[41, 5, 1, "", "add_next_param_from_list"], [41, 5, 1, "", "add_references_and_check_constraints"], [41, 5, 1, "", "add_shared_params_to_param_grids"], [41, 5, 1, "", "build_param_grid_of_shared_params"], [41, 5, 1, "", "grid_task"], [41, 5, 1, "", "lognormal_grid"], [41, 5, 1, "", "loguniform_grid"], [41, 5, 1, "", "normal_grid"], [41, 5, 1, "", "rais_error_if_num_not_specified"], [41, 5, 1, "", "round_to_discreate_grid_normal"], [41, 5, 1, "", "round_to_discreate_grid_uniform"], [41, 5, 1, "", "sample_grid"], [41, 5, 1, "", "sample_gridsearch"], [41, 5, 1, "", "uniform_grid"]], "domainlab.utils.hyperparameter_retrieval": [[41, 5, 1, "", "get_gamma_reg"]], "domainlab.utils.hyperparameter_sampling": [[41, 1, 1, "", "CategoricalHyperparameter"], [41, 1, 1, "", "Hyperparameter"], [41, 1, 1, "", "ReferenceHyperparameter"], [41, 1, 1, "", "SampledHyperparameter"], [41, 5, 1, "", "check_constraints"], [41, 5, 1, "", "create_samples_from_shared_samples"], [41, 5, 1, "", "get_hyperparameter"], [41, 5, 1, "", "get_shared_samples"], [41, 5, 1, "", "is_dict_with_key"], [41, 5, 1, "", "sample_hyperparameters"], [41, 5, 1, "", "sample_parameters"], [41, 5, 1, "", "sample_task"], [41, 5, 1, "", "sample_task_only_shared"]], "domainlab.utils.hyperparameter_sampling.CategoricalHyperparameter": [[41, 2, 1, "", "datatype"], [41, 2, 1, "", "sample"]], "domainlab.utils.hyperparameter_sampling.Hyperparameter": [[41, 2, 1, "", "datatype"], [41, 2, 1, "", "get_val"], [41, 2, 1, "", "sample"]], "domainlab.utils.hyperparameter_sampling.ReferenceHyperparameter": [[41, 2, 1, "", "datatype"], [41, 2, 1, "", "sample"]], "domainlab.utils.hyperparameter_sampling.SampledHyperparameter": [[41, 2, 1, "", "datatype"], [41, 2, 1, "", "sample"]], "domainlab.utils.logger": [[41, 1, 1, "", "Logger"]], "domainlab.utils.logger.Logger": [[41, 2, 1, "", "get_logger"], [41, 3, 1, "", "logger"]], "domainlab.utils.override_interface": [[41, 5, 1, "", "override_interface"]], "domainlab.utils.perf": [[41, 1, 1, "", "PerfClassif"]], "domainlab.utils.perf.PerfClassif": [[41, 2, 1, "", "cal_acc"], [41, 2, 1, "", "gen_fun_acc"], [41, 2, 1, "", "get_list_pred_target"]], "domainlab.utils.perf_metrics": [[41, 1, 1, "", "PerfMetricClassif"]], "domainlab.utils.perf_metrics.PerfMetricClassif": [[41, 2, 1, "", "cal_metrics"]], "domainlab.utils.sanity_check": [[41, 1, 1, "", "SanityCheck"]], "domainlab.utils.sanity_check.SanityCheck": [[41, 2, 1, "", "dataset_sanity_check"], [41, 2, 1, "", "save_san_check_for_domain"]], "domainlab.utils.test_img": [[41, 5, 1, "", "mk_img"], [41, 5, 1, "", "mk_rand_label_onehot"], [41, 5, 1, "", "mk_rand_xyd"]], "domainlab.utils.u_import": [[41, 5, 1, "", "import_path"]], "domainlab.utils.u_import_net_module": [[41, 5, 1, "", "build_external_obj_net_module_feat_extract"]], "domainlab.utils.utils_class": [[41, 5, 1, "", "store_args"]], "domainlab.utils.utils_classif": [[41, 5, 1, "", "get_label_na"], [41, 5, 1, "", "logit2preds_vpic"], [41, 5, 1, "", "mk_dummy_label_list_str"]], "domainlab.utils.utils_cuda": [[41, 5, 1, "", "get_device"]], "domainlab.utils.utils_img_sav": [[41, 5, 1, "", "mk_fun_sav_img"], [41, 5, 1, "", "sav_add_title"]]}, "objtypes": {"0": "py:module", "1": "py:class", "2": "py:method", "3": "py:attribute", "4": "py:property", "5": "py:function"}, "objnames": {"0": ["py", "module", "Python module"], "1": ["py", "class", "Python class"], "2": ["py", "method", "Python method"], "3": ["py", "attribute", "Python attribute"], "4": ["py", "property", "Python property"], "5": ["py", "function", "Python function"]}, "titleterms": {"trainer": [0, 1, 3, 6, 13, 19, 29, 30], "dial": [0, 24], "domain": [0, 1, 2, 6, 7, 12, 14, 15, 19, 22, 44], "invari": [0, 1, 3, 15, 18, 44], "adversari": [0, 14], "learn": [0, 19], "gener": [0, 1, 2, 6, 8, 11, 15, 19, 44], "network": [0, 2, 4, 6, 13, 14, 15, 22, 44], "structur": [0, 8, 16, 21], "loss": [0, 2], "function": [0, 2], "gradient": [0, 1], "revers": 0, "layer": 0, "exampl": [0, 1, 2, 3, 4, 6, 12, 13, 15, 18, 19, 22, 23, 24, 44], "train": [0, 1, 2, 6, 15, 22, 44], "diva": [0, 7, 15], "model": [0, 2, 4, 13, 14, 15, 20, 23, 24, 39], "set": [0, 8], "hyper": 0, "paramet": [0, 4, 15, 24], "well": 0, "fishr": 1, "varianc": 1, "out": [1, 15], "distribut": [1, 11, 18, 44], "quantifi": 1, "inconsist": 1, "between": [1, 8], "match": [1, 6, 18], "dure": 1, "implement": [1, 2, 4, 6, 13, 15, 20, 22, 23], "hduva": [2, 6], "hierarch": 2, "variat": [2, 15], "auto": 2, "encod": [2, 35], "FOR": 2, "unsupervis": 2, "overview": 2, "evid": 2, "lower": 2, "bound": 2, "overal": 2, "hyperparamet": [2, 8, 9, 11, 24], "warmup": 2, "us": [2, 6, 13, 22], "custom": [2, 6, 13, 15, 20, 22], "net": 2, "sandwich": 2, "topic": 2, "classif": [2, 7], "color": [2, 7, 15], "mnist": [2, 6, 7, 15], "2": [2, 20, 22], "so": 2, "work": 2, "also": 2, "singl": 2, "neural": [2, 4, 6, 13, 14, 15, 22, 44], "alex": 2, "irm": 3, "risk": [3, 7], "minim": [3, 7], "irmv1": 3, "jigen": [4, 24], "sanniti": 4, "check": [4, 8, 22], "tile": 4, "shuffl": 4, "simpl": 5, "move": 5, "averag": 5, "matchdg": [6, 13, 24], "causal": [6, 18], "motiv": 6, "theori": 6, "dataset": [7, 22], "generalis": 7, "erm": [7, 13], "emper": 7, "result": [7, 8], "detail": [7, 8], "prompt": 7, "explan": [7, 9], "benchmark": [8, 9, 11, 44], "domainlab": [8, 10, 17, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44], "depend": [8, 17], "instal": [8, 17, 44], "unix": 8, "window": 8, "up": 8, "run": 8, "standalon": 8, "machin": 8, "comput": 8, "node": 8, "without": 8, "gpu": 8, "hpc": 8, "cluster": 8, "slurm": 8, "error": 8, "map": 8, "job": 8, "id": 8, "sampl": [8, 9, 11], "index": 8, "obtain": 8, "output": [8, 21], "folder": [8, 22], "partial": 8, "plot": 8, "from": [8, 22], "csv": 8, "file": [8, 11, 22, 44], "further": [9, 24], "setup": [9, 11], "constraint": 9, "let": 10, "know": 10, "where": 10, "your": 10, "pac": 10, "data": [10, 22], "were": 10, "locat": 10, "yaml": 11, "descript": 11, "uniform": 11, "loguniform": 11, "normal": 11, "lognorm": 11, "cathegor": 11, "referenc": 11, "hperparamet": 11, "special": 11, "argument": [11, 20, 24, 44], "combin": 11, "share": 11, "task": [11, 22, 24, 40, 44], "specif": [11, 20, 22, 23, 24, 44], "deep": 12, "coral": 12, "correl": 12, "align": 12, "adapt": 12, "specifi": [13, 22], "command": [13, 24, 44], "line": [13, 44], "case": 13, "larger": 13, "imag": [13, 15], "dann": [13, 14], "autoencod": 15, "version": [15, 17], "leav": 15, "one": 15, "choos": 15, "test": [15, 22], "softwar": 16, "architectur": 16, "design": 16, "code": 16, "creat": [17, 20], "virtual": 17, "environ": 17, "strongli": 17, "recommend": 17, "develop": 17, "via": 17, "github": 17, "manag": 17, "releas": 17, "irl": 18, "mechan": 18, "through": 18, "mldg": 19, "meta": 19, "option": [20, 24], "1": [20, 22], "extend": [20, 23], "class": [20, 22], "add": 20, "alongsid": 20, "exist": 20, "step": [20, 23], "requir": 20, "abstract": 20, "method": [20, 44], "addit": 20, "3": [20, 22], "builder": 20, "possibl": 22, "directli": 22, "taskpathlist": 22, "taskfold": 22, "organ": 22, "api": [22, 44], "user": 22, "defin": 22, "extern": 22, "imagepath": 22, "algorithm": 22, "python": [22, 44], "make": 22, "saniti": 22, "8": 22, "instanc": 22, "each": 22, "onli": 22, "domain2": 22, "trainerbas": 23, "usag": [24, 44], "guid": 24, "essenti": 24, "advanc": 24, "configur": [24, 44], "vae": [24, 34, 35], "resourc": 24, "packag": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 44], "subpackag": [25, 26, 29, 31, 34], "submodul": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41], "arg_pars": 25, "modul": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41], "cli": 25, "mk_exp": 25, "content": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42], "algo": [26, 27, 28, 29, 30], "a_algo_build": 26, "builder_api_model": 26, "builder_custom": 26, "builder_dann": 26, "builder_diva": 26, "builder_erm": 26, "builder_hduva": 26, "builder_jigen1": 26, "util": [26, 41], "zoo_algo": 26, "msel": 27, "a_model_sel": 27, "c_msel_oracl": 27, "c_msel_tr_loss": 27, "c_msel_val": 27, "observ": 28, "a_observ": 28, "b_obvisitor": 28, "c_obvisitor_cleanup": 28, "c_obvisitor_gen": 28, "a_train": 29, "args_dial": 29, "hyper_schedul": 29, "mmd_base": 29, "train_bas": 29, "train_causirl": 29, "train_cor": 29, "train_dial": 29, "train_ema": 29, "train_fishr": 29, "train_hyper_schedul": 29, "train_irm": 29, "train_matchdg": 29, "train_mldg": 29, "zoo_train": 29, "compo": [30, 31, 32, 33, 34, 35], "matchdg_arg": 30, "matchdg_match": 30, "matchdg_util": 30, "a_nn_build": 31, "builder_nn_alex": 31, "builder_nn_conv_bn_relu_2": 31, "builder_nn_external_from_fil": 31, "utils_conv_get_flat_dim": 31, "zoo_nn": 31, "nn_zoo": 32, "net_adversari": 32, "net_classif": 32, "net_conv_conv_bn_pool_2": 32, "net_gat": 32, "nn": 32, "nn_alex": 32, "nn_torchvis": 32, "pcr": 33, "p_chain_handl": 33, "request": 33, "a_model_build": 34, "a_vae_build": 34, "c_vae_adaptor_model_recon": 34, "c_vae_builder_classif": 34, "c_vae_recon": 34, "utils_request_chain_build": 34, "zoo_vae_builders_classif": 34, "zoo_vae_builders_classif_top": 34, "decoder_concat_vec_reshape_conv": 35, "decoder_concat_vec_reshape_conv_gated_conv": 35, "decoder_cond_prior": 35, "decoder_loss": 35, "encoder_dirichlet": 35, "encoder_domain_top": 35, "encoder_domain_topic_img2top": 35, "encoder_domain_topic_img_topic2zd": 35, "encoder_xyd_parallel": 35, "encoder_xydt_elev": 35, "encoder_zi": 35, "dset": 36, "a_dset_mnist_color_rgb_solo": 36, "dset_img_path_list": 36, "dset_mnist_color_solo_default": 36, "dset_poly_domains_mnist_color_default": 36, "dset_subfold": 36, "utils_color_palett": 36, "utils_data": 36, "utils_wrapdset_patch": 36, "exp": 37, "exp_cuda_se": 37, "exp_main": 37, "exp_util": 37, "exp_protocol": 38, "aggregate_result": 38, "run_experi": 38, "a_model": 39, "a_model_classif": 39, "args_jigen": 39, "args_va": 39, "interface_vae_xyd": 39, "model_custom": 39, "model_dann": 39, "model_diva": 39, "model_erm": 39, "model_hduva": 39, "model_jigen": 39, "model_vae_xyd_classif": 39, "a_task": 40, "a_task_classif": 40, "b_task": 40, "b_task_classif": 40, "task_dset": 40, "task_fold": 40, "task_folder_mk": 40, "task_mini_vlc": 40, "task_mnist_color": 40, "task_pathlist": 40, "task_util": 40, "utils_task": 40, "utils_task_dset": 40, "zoo_task": 40, "flows_gen_img_model": 41, "generate_benchmark_plot": 41, "get_git_tag": 41, "hyperparameter_gridsearch": 41, "hyperparameter_retriev": 41, "hyperparameter_sampl": 41, "logger": 41, "override_interfac": 41, "perf": 41, "perf_metr": 41, "sanity_check": 41, "test_img": 41, "u_import": 41, "u_import_net_modul": 41, "utils_class": 41, "utils_classif": 41, "utils_cuda": 41, "utils_img_sav": 41, "welcom": 42, "s": 42, "document": 42, "indic": 42, "tabl": 42, "introduct": 44, "modular": 44, "shift": 44, "get": 44, "start": 44, "avail": 44, "commandlin": 44, "programm": 44, "against": 44, "differ": 44, "citat": 44, "recommonmark": 45, "0": 45, "5": 45, "dev0": 45}, "envversion": {"sphinx.domains.c": 2, "sphinx.domains.changeset": 1, "sphinx.domains.citation": 1, "sphinx.domains.cpp": 6, "sphinx.domains.index": 1, "sphinx.domains.javascript": 2, "sphinx.domains.math": 2, "sphinx.domains.python": 3, "sphinx.domains.rst": 2, "sphinx.domains.std": 2, "sphinx.ext.viewcode": 1, "sphinx.ext.intersphinx": 1, "sphinx.ext.todo": 2, "nbsphinx": 4, "sphinx": 56}}) \ No newline at end of file +Search.setIndex({docnames:["docDIAL","docFishr","docHDUVA","docIRM","docJiGen","docMA","docMatchDG","doc_MNIST_classification","doc_benchmark","doc_benchmark_further_explanation","doc_benchmark_pacs","doc_benchmark_yaml","doc_coral","doc_custom_nn","doc_dann","doc_diva","doc_extend_contribute","doc_install","doc_irl","doc_mldg","doc_model","doc_output","doc_tasks","doc_trainer","doc_usage_cmd","domainlab","domainlab.algos","domainlab.algos.msels","domainlab.algos.observers","domainlab.algos.trainers","domainlab.algos.trainers.compos","domainlab.compos","domainlab.compos.nn_zoo","domainlab.compos.pcr","domainlab.compos.vae","domainlab.compos.vae.compos","domainlab.dsets","domainlab.exp","domainlab.exp_protocol","domainlab.models","domainlab.tasks","domainlab.utils","index","modules","readme_link","requirements","tips"],envversion:{"sphinx.domains.c":2,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":4,"sphinx.domains.index":1,"sphinx.domains.javascript":2,"sphinx.domains.math":2,"sphinx.domains.python":3,"sphinx.domains.rst":2,"sphinx.domains.std":2,"sphinx.ext.intersphinx":1,"sphinx.ext.todo":2,"sphinx.ext.viewcode":1,nbsphinx:4,sphinx:56},filenames:["docDIAL.md","docFishr.md","docHDUVA.md","docIRM.md","docJiGen.md","docMA.md","docMatchDG.md","doc_MNIST_classification.md","doc_benchmark.md","doc_benchmark_further_explanation.md","doc_benchmark_pacs.md","doc_benchmark_yaml.md","doc_coral.md","doc_custom_nn.md","doc_dann.md","doc_diva.md","doc_extend_contribute.md","doc_install.md","doc_irl.md","doc_mldg.md","doc_model.md","doc_output.md","doc_tasks.md","doc_trainer.md","doc_usage_cmd.md","domainlab.rst","domainlab.algos.rst","domainlab.algos.msels.rst","domainlab.algos.observers.rst","domainlab.algos.trainers.rst","domainlab.algos.trainers.compos.rst","domainlab.compos.rst","domainlab.compos.nn_zoo.rst","domainlab.compos.pcr.rst","domainlab.compos.vae.rst","domainlab.compos.vae.compos.rst","domainlab.dsets.rst","domainlab.exp.rst","domainlab.exp_protocol.rst","domainlab.models.rst","domainlab.tasks.rst","domainlab.utils.rst","index.rst","modules.rst","readme_link.rst","requirements.txt","tips.txt"],objects:{"":[[25,0,0,"-","domainlab"]],"domainlab.algos":[[26,0,0,"-","a_algo_builder"],[26,0,0,"-","builder_api_model"],[26,0,0,"-","builder_custom"],[26,0,0,"-","builder_dann"],[26,0,0,"-","builder_diva"],[26,0,0,"-","builder_erm"],[26,0,0,"-","builder_hduva"],[26,0,0,"-","builder_jigen1"],[27,0,0,"-","msels"],[28,0,0,"-","observers"],[29,0,0,"-","trainers"],[26,0,0,"-","utils"],[26,0,0,"-","zoo_algos"]],"domainlab.algos.a_algo_builder":[[26,1,1,"","NodeAlgoBuilder"]],"domainlab.algos.a_algo_builder.NodeAlgoBuilder":[[26,2,1,"","extend"],[26,2,1,"","init_business"],[26,2,1,"","init_next_model"],[26,2,1,"","is_myjob"],[26,3,1,"","na_prefix"],[26,4,1,"","name"]],"domainlab.algos.builder_api_model":[[26,1,1,"","NodeAlgoBuilderAPIModel"]],"domainlab.algos.builder_api_model.NodeAlgoBuilderAPIModel":[[26,2,1,"","init_business"]],"domainlab.algos.builder_custom":[[26,5,1,"","make_basic_trainer"]],"domainlab.algos.builder_dann":[[26,1,1,"","NodeAlgoBuilderDANN"]],"domainlab.algos.builder_dann.NodeAlgoBuilderDANN":[[26,2,1,"","init_business"],[26,2,1,"","reset_aux_net"]],"domainlab.algos.builder_diva":[[26,1,1,"","NodeAlgoBuilderDIVA"]],"domainlab.algos.builder_diva.NodeAlgoBuilderDIVA":[[26,2,1,"","get_trainer"],[26,2,1,"","init_business"]],"domainlab.algos.builder_erm":[[26,1,1,"","NodeAlgoBuilderERM"]],"domainlab.algos.builder_erm.NodeAlgoBuilderERM":[[26,2,1,"","init_business"]],"domainlab.algos.builder_hduva":[[26,1,1,"","NodeAlgoBuilderHDUVA"]],"domainlab.algos.builder_hduva.NodeAlgoBuilderHDUVA":[[26,2,1,"","init_business"]],"domainlab.algos.builder_jigen1":[[26,1,1,"","NodeAlgoBuilderJiGen"]],"domainlab.algos.builder_jigen1.NodeAlgoBuilderJiGen":[[26,2,1,"","init_business"]],"domainlab.algos.msels":[[27,0,0,"-","a_model_sel"],[27,0,0,"-","c_msel_oracle"],[27,0,0,"-","c_msel_tr_loss"],[27,0,0,"-","c_msel_val"]],"domainlab.algos.msels.a_model_sel":[[27,1,1,"","AMSel"]],"domainlab.algos.msels.a_model_sel.AMSel":[[27,2,1,"","accept"],[27,2,1,"","base_update"],[27,4,1,"","best_te_metric"],[27,4,1,"","best_val_acc"],[27,2,1,"","early_stop"],[27,2,1,"","if_stop"],[27,4,1,"","max_es"],[27,4,1,"","model_selection_epoch"],[27,4,1,"","observer4msel"],[27,2,1,"","reset"],[27,4,1,"","sel_model_te_acc"],[27,2,1,"","update"],[27,4,1,"","val_threshold"]],"domainlab.algos.msels.c_msel_oracle":[[27,1,1,"","MSelOracleVisitor"]],"domainlab.algos.msels.c_msel_oracle.MSelOracleVisitor":[[27,2,1,"","accept"],[27,2,1,"","base_update"],[27,2,1,"","early_stop"],[27,4,1,"","oracle_last_setpoint_sel_te_acc"]],"domainlab.algos.msels.c_msel_tr_loss":[[27,1,1,"","MSelTrLoss"]],"domainlab.algos.msels.c_msel_tr_loss.MSelTrLoss":[[27,2,1,"","base_update"],[27,2,1,"","early_stop"],[27,4,1,"","max_es"],[27,2,1,"","reset"]],"domainlab.algos.msels.c_msel_val":[[27,1,1,"","MSelValPerf"]],"domainlab.algos.msels.c_msel_val.MSelValPerf":[[27,2,1,"","base_update"],[27,4,1,"","best_te_metric"],[27,4,1,"","best_val_acc"],[27,2,1,"","reset"],[27,4,1,"","sel_model_te_acc"]],"domainlab.algos.observers":[[28,0,0,"-","a_observer"],[28,0,0,"-","b_obvisitor"],[28,0,0,"-","c_obvisitor_cleanup"],[28,0,0,"-","c_obvisitor_gen"]],"domainlab.algos.observers.a_observer":[[28,1,1,"","AObVisitor"]],"domainlab.algos.observers.a_observer.AObVisitor":[[28,2,1,"","accept"],[28,2,1,"","after_all"],[28,2,1,"","clean_up"],[28,2,1,"","reset"],[28,2,1,"","set_task"],[28,2,1,"","update"]],"domainlab.algos.observers.b_obvisitor":[[28,1,1,"","ObVisitor"]],"domainlab.algos.observers.b_obvisitor.ObVisitor":[[28,2,1,"","accept"],[28,2,1,"","after_all"],[28,2,1,"","cal_oracle_perf"],[28,2,1,"","clean_up"],[28,2,1,"","dump_prediction"],[28,4,1,"","str_metric4msel"],[28,2,1,"","update"]],"domainlab.algos.observers.c_obvisitor_cleanup":[[28,1,1,"","ObVisitorCleanUp"]],"domainlab.algos.observers.c_obvisitor_cleanup.ObVisitorCleanUp":[[28,2,1,"","accept"],[28,2,1,"","after_all"],[28,2,1,"","clean_up"],[28,4,1,"","model_sel"],[28,2,1,"","update"]],"domainlab.algos.observers.c_obvisitor_gen":[[28,1,1,"","ObVisitorGen"]],"domainlab.algos.observers.c_obvisitor_gen.ObVisitorGen":[[28,2,1,"","after_all"]],"domainlab.algos.trainers":[[29,0,0,"-","a_trainer"],[29,0,0,"-","args_dial"],[30,0,0,"-","compos"],[29,0,0,"-","hyper_scheduler"],[29,0,0,"-","mmd_base"],[29,0,0,"-","train_basic"],[29,0,0,"-","train_causIRL"],[29,0,0,"-","train_coral"],[29,0,0,"-","train_dial"],[29,0,0,"-","train_ema"],[29,0,0,"-","train_fishr"],[29,0,0,"-","train_hyper_scheduler"],[29,0,0,"-","train_irm"],[29,0,0,"-","train_matchdg"],[29,0,0,"-","train_mldg"],[29,0,0,"-","zoo_trainer"]],"domainlab.algos.trainers.a_trainer":[[29,1,1,"","AbstractTrainer"],[29,5,1,"","mk_opt"]],"domainlab.algos.trainers.a_trainer.AbstractTrainer":[[29,2,1,"","after_batch"],[29,2,1,"","before_batch"],[29,2,1,"","before_tr"],[29,2,1,"","cal_reg_loss"],[29,2,1,"","cal_reg_loss_over_task_loss_ratio"],[29,4,1,"","decoratee"],[29,2,1,"","dset_decoration_args_algo"],[29,2,1,"","extend"],[29,2,1,"","get_model"],[29,2,1,"","init_business"],[29,2,1,"","is_myjob"],[29,4,1,"","list_tr_domain_size"],[29,4,1,"","model"],[29,4,1,"","name"],[29,4,1,"","p_na_prefix"],[29,2,1,"","post_tr"],[29,2,1,"","print_parameters"],[29,2,1,"","reset"],[29,4,1,"","str_metric4msel"],[29,2,1,"","tr_epoch"]],"domainlab.algos.trainers.args_dial":[[29,5,1,"","add_args2parser_dial"]],"domainlab.algos.trainers.compos":[[30,0,0,"-","matchdg_args"],[30,0,0,"-","matchdg_match"],[30,0,0,"-","matchdg_utils"]],"domainlab.algos.trainers.compos.matchdg_args":[[30,5,1,"","add_args2parser_matchdg"]],"domainlab.algos.trainers.compos.matchdg_match":[[30,1,1,"","MatchPair"]],"domainlab.algos.trainers.compos.matchdg_utils":[[30,1,1,"","MatchDictInit"],[30,1,1,"","MatchDictNumDomain2SizeDomain"],[30,1,1,"","MatchDictVirtualRefDset2EachDomain"],[30,5,1,"","dist_cosine_agg"],[30,5,1,"","dist_pairwise_cosine"],[30,5,1,"","fun_tensor_normalize"],[30,5,1,"","get_base_domain_size4match_dg"]],"domainlab.algos.trainers.compos.matchdg_utils.MatchDictInit":[[30,2,1,"","get_num_rows"]],"domainlab.algos.trainers.compos.matchdg_utils.MatchDictNumDomain2SizeDomain":[[30,2,1,"","get_num_rows"]],"domainlab.algos.trainers.compos.matchdg_utils.MatchDictVirtualRefDset2EachDomain":[[30,2,1,"","get_num_rows"]],"domainlab.algos.trainers.hyper_scheduler":[[29,1,1,"","HyperSchedulerWarmupExponential"],[29,1,1,"","HyperSchedulerWarmupLinear"]],"domainlab.algos.trainers.hyper_scheduler.HyperSchedulerWarmupExponential":[[29,2,1,"","warmup"]],"domainlab.algos.trainers.hyper_scheduler.HyperSchedulerWarmupLinear":[[29,2,1,"","set_steps"],[29,2,1,"","warmup"]],"domainlab.algos.trainers.mmd_base":[[29,1,1,"","TrainerMMDBase"]],"domainlab.algos.trainers.mmd_base.TrainerMMDBase":[[29,2,1,"","gaussian_kernel"],[29,2,1,"","mmd"],[29,2,1,"","my_cdist"]],"domainlab.algos.trainers.train_basic":[[29,1,1,"","TrainerBasic"],[29,5,1,"","list_divide"]],"domainlab.algos.trainers.train_basic.TrainerBasic":[[29,2,1,"","after_epoch"],[29,2,1,"","before_epoch"],[29,2,1,"","before_tr"],[29,2,1,"","cal_loss"],[29,2,1,"","log_loss"],[29,2,1,"","tr_batch"],[29,2,1,"","tr_epoch"]],"domainlab.algos.trainers.train_causIRL":[[29,1,1,"","TrainerCausalIRL"]],"domainlab.algos.trainers.train_causIRL.TrainerCausalIRL":[[29,2,1,"","gaussian_kernel"],[29,2,1,"","mmd"],[29,2,1,"","my_cdist"],[29,2,1,"","tr_batch"]],"domainlab.algos.trainers.train_coral":[[29,1,1,"","TrainerCoral"]],"domainlab.algos.trainers.train_coral.TrainerCoral":[[29,2,1,"","cross_domain_mmd"],[29,2,1,"","tr_epoch"]],"domainlab.algos.trainers.train_dial":[[29,1,1,"","TrainerDIAL"]],"domainlab.algos.trainers.train_dial.TrainerDIAL":[[29,2,1,"","gen_adversarial"]],"domainlab.algos.trainers.train_ema":[[29,1,1,"","TrainerMA"]],"domainlab.algos.trainers.train_ema.TrainerMA":[[29,2,1,"","after_epoch"],[29,2,1,"","move_average"]],"domainlab.algos.trainers.train_fishr":[[29,1,1,"","TrainerFishr"]],"domainlab.algos.trainers.train_fishr.TrainerFishr":[[29,2,1,"","cal_dict_variance_grads"],[29,2,1,"","cal_mean_across_dict"],[29,2,1,"","cal_power_single_dict"],[29,2,1,"","tr_epoch"],[29,2,1,"","var_grads_and_loss"],[29,2,1,"","variance_between_dict"]],"domainlab.algos.trainers.train_hyper_scheduler":[[29,1,1,"","TrainerHyperScheduler"]],"domainlab.algos.trainers.train_hyper_scheduler.TrainerHyperScheduler":[[29,2,1,"","before_batch"],[29,2,1,"","before_tr"],[29,2,1,"","set_scheduler"],[29,2,1,"","tr_epoch"]],"domainlab.algos.trainers.train_irm":[[29,1,1,"","TrainerIRM"]],"domainlab.algos.trainers.train_irm.TrainerIRM":[[29,2,1,"","tr_epoch"]],"domainlab.algos.trainers.train_matchdg":[[29,1,1,"","TrainerMatchDG"],[29,5,1,"","match_tensor_reshape"]],"domainlab.algos.trainers.train_matchdg.TrainerMatchDG":[[29,2,1,"","before_tr"],[29,2,1,"","dset_decoration_args_algo"],[29,2,1,"","init_business"],[29,2,1,"","mk_match_tensor"],[29,2,1,"","tr_batch"],[29,2,1,"","tr_epoch"]],"domainlab.algos.trainers.train_mldg":[[29,1,1,"","TrainerMLDG"]],"domainlab.algos.trainers.train_mldg.TrainerMLDG":[[29,2,1,"","before_tr"],[29,2,1,"","prepare_ziped_loader"],[29,2,1,"","tr_epoch"]],"domainlab.algos.trainers.zoo_trainer":[[29,1,1,"","TrainerChainNodeGetter"]],"domainlab.algos.utils":[[26,5,1,"","split_net_feat_last"]],"domainlab.algos.zoo_algos":[[26,1,1,"","AlgoBuilderChainNodeGetter"]],"domainlab.algos.zoo_algos.AlgoBuilderChainNodeGetter":[[26,2,1,"","register_external_node"]],"domainlab.arg_parser":[[25,1,1,"","ParseValuesOrKeyValuePairs"],[25,5,1,"","apply_dict_to_args"],[25,5,1,"","mk_parser_main"],[25,5,1,"","parse_cmd_args"]],"domainlab.compos":[[31,0,0,"-","a_nn_builder"],[31,0,0,"-","builder_nn_alex"],[31,0,0,"-","builder_nn_conv_bn_relu_2"],[31,0,0,"-","builder_nn_external_from_file"],[32,0,0,"-","nn_zoo"],[33,0,0,"-","pcr"],[31,0,0,"-","utils_conv_get_flat_dim"],[34,0,0,"-","vae"],[31,0,0,"-","zoo_nn"]],"domainlab.compos.a_nn_builder":[[31,1,1,"","AbstractFeatExtractNNBuilderChainNode"]],"domainlab.compos.a_nn_builder.AbstractFeatExtractNNBuilderChainNode":[[31,2,1,"","init_business"],[31,2,1,"","is_myjob"]],"domainlab.compos.builder_nn_alex":[[31,5,1,"","mkNodeFeatExtractNNBuilderNameAlex"]],"domainlab.compos.builder_nn_conv_bn_relu_2":[[31,5,1,"","mkNodeFeatExtractNNBuilderNameConvBnRelu2"]],"domainlab.compos.builder_nn_external_from_file":[[31,5,1,"","mkNodeFeatExtractNNBuilderExternFromFile"]],"domainlab.compos.nn_zoo":[[32,0,0,"-","net_adversarial"],[32,0,0,"-","net_classif"],[32,0,0,"-","net_conv_conv_bn_pool_2"],[32,0,0,"-","net_gated"],[32,0,0,"-","nn"],[32,0,0,"-","nn_alex"],[32,0,0,"-","nn_torchvision"]],"domainlab.compos.nn_zoo.net_adversarial":[[32,1,1,"","AutoGradFunMultiply"],[32,1,1,"","AutoGradFunReverseMultiply"],[32,1,1,"","Flatten"]],"domainlab.compos.nn_zoo.net_adversarial.AutoGradFunMultiply":[[32,2,1,"","backward"],[32,2,1,"","forward"]],"domainlab.compos.nn_zoo.net_adversarial.AutoGradFunReverseMultiply":[[32,2,1,"","backward"],[32,2,1,"","forward"]],"domainlab.compos.nn_zoo.net_adversarial.Flatten":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.net_classif":[[32,1,1,"","ClassifDropoutReluLinear"]],"domainlab.compos.nn_zoo.net_classif.ClassifDropoutReluLinear":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2":[[32,1,1,"","NetConvBnReluPool2L"],[32,1,1,"","NetConvDense"],[32,5,1,"","mk_conv_bn_relu_pool"]],"domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2.NetConvBnReluPool2L":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.net_conv_conv_bn_pool_2.NetConvDense":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.net_gated":[[32,1,1,"","Conv2d"],[32,1,1,"","GatedConv2d"],[32,1,1,"","GatedDense"]],"domainlab.compos.nn_zoo.net_gated.Conv2d":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.net_gated.GatedConv2d":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.net_gated.GatedDense":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.nn":[[32,1,1,"","DenseNet"],[32,1,1,"","LayerId"]],"domainlab.compos.nn_zoo.nn.DenseNet":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.nn.LayerId":[[32,2,1,"","forward"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.nn_alex":[[32,1,1,"","Alex4DeepAll"],[32,1,1,"","AlexNetBase"],[32,1,1,"","AlexNetNoLastLayer"]],"domainlab.compos.nn_zoo.nn_alex.Alex4DeepAll":[[32,3,1,"","training"]],"domainlab.compos.nn_zoo.nn_alex.AlexNetBase":[[32,2,1,"","fetch_net"],[32,3,1,"","training"]],"domainlab.compos.nn_zoo.nn_alex.AlexNetNoLastLayer":[[32,3,1,"","training"]],"domainlab.compos.nn_zoo.nn_torchvision":[[32,1,1,"","NetTorchVisionBase"]],"domainlab.compos.nn_zoo.nn_torchvision.NetTorchVisionBase":[[32,2,1,"","fetch_net"],[32,2,1,"","forward"],[32,2,1,"","show"],[32,3,1,"","training"]],"domainlab.compos.pcr":[[33,0,0,"-","p_chain_handler"],[33,0,0,"-","request"]],"domainlab.compos.pcr.p_chain_handler":[[33,1,1,"","AbstractChainNodeHandler"],[33,1,1,"","DummyBusiness"],[33,1,1,"","DummyChainNodeHandlerBeaver"],[33,1,1,"","DummyChainNodeHandlerLazy"],[33,1,1,"","Request4Chain"]],"domainlab.compos.pcr.p_chain_handler.AbstractChainNodeHandler":[[33,2,1,"","handle"],[33,2,1,"","init_business"],[33,2,1,"","is_myjob"],[33,2,1,"","print_options"],[33,2,1,"","set_parent"]],"domainlab.compos.pcr.p_chain_handler.DummyBusiness":[[33,3,1,"","message"]],"domainlab.compos.pcr.p_chain_handler.DummyChainNodeHandlerBeaver":[[33,2,1,"","init_business"],[33,2,1,"","is_myjob"]],"domainlab.compos.pcr.p_chain_handler.DummyChainNodeHandlerLazy":[[33,2,1,"","init_business"],[33,2,1,"","is_myjob"]],"domainlab.compos.pcr.p_chain_handler.Request4Chain":[[33,2,1,"","convert"]],"domainlab.compos.pcr.request":[[33,1,1,"","RequestArgs2ExpCmd"],[33,1,1,"","RequestTask"],[33,1,1,"","RequestVAEBuilderCHW"],[33,1,1,"","RequestVAEBuilderNN"]],"domainlab.compos.utils_conv_get_flat_dim":[[31,5,1,"","get_flat_dim"]],"domainlab.compos.vae":[[34,0,0,"-","a_model_builder"],[34,0,0,"-","a_vae_builder"],[34,0,0,"-","c_vae_adaptor_model_recon"],[34,0,0,"-","c_vae_builder_classif"],[34,0,0,"-","c_vae_recon"],[35,0,0,"-","compos"],[34,0,0,"-","utils_request_chain_builder"],[34,0,0,"-","zoo_vae_builders_classif"],[34,0,0,"-","zoo_vae_builders_classif_topic"]],"domainlab.compos.vae.a_model_builder":[[34,1,1,"","AbstractModelBuilderChainNode"]],"domainlab.compos.vae.a_model_builder.AbstractModelBuilderChainNode":[[34,2,1,"","config"]],"domainlab.compos.vae.a_vae_builder":[[34,1,1,"","AbstractVAEBuilderChainNode"]],"domainlab.compos.vae.a_vae_builder.AbstractVAEBuilderChainNode":[[34,2,1,"","build_decoder"],[34,2,1,"","build_encoder"],[34,2,1,"","init_business"]],"domainlab.compos.vae.c_vae_adaptor_model_recon":[[34,1,1,"","AdaptorReconVAEXYD"]],"domainlab.compos.vae.c_vae_adaptor_model_recon.AdaptorReconVAEXYD":[[34,2,1,"","cal_latent"],[34,2,1,"","cal_prior_zd"],[34,2,1,"","cal_prior_zy"],[34,2,1,"","recon_ydx"]],"domainlab.compos.vae.c_vae_builder_classif":[[34,1,1,"","ChainNodeVAEBuilderClassifCondPrior"]],"domainlab.compos.vae.c_vae_builder_classif.ChainNodeVAEBuilderClassifCondPrior":[[34,2,1,"","build_decoder"],[34,2,1,"","build_encoder"],[34,2,1,"","construct_classifier"],[34,2,1,"","construct_cond_prior"],[34,2,1,"","is_myjob"]],"domainlab.compos.vae.c_vae_recon":[[34,1,1,"","ReconVAEXYD"]],"domainlab.compos.vae.c_vae_recon.ReconVAEXYD":[[34,2,1,"","recon"],[34,2,1,"","recon_cf"]],"domainlab.compos.vae.compos":[[35,0,0,"-","decoder_concat_vec_reshape_conv"],[35,0,0,"-","decoder_concat_vec_reshape_conv_gated_conv"],[35,0,0,"-","decoder_cond_prior"],[35,0,0,"-","decoder_losses"],[35,0,0,"-","encoder"],[35,0,0,"-","encoder_dirichlet"],[35,0,0,"-","encoder_domain_topic"],[35,0,0,"-","encoder_domain_topic_img2topic"],[35,0,0,"-","encoder_domain_topic_img_topic2zd"],[35,0,0,"-","encoder_xyd_parallel"],[35,0,0,"-","encoder_xydt_elevator"],[35,0,0,"-","encoder_zy"]],"domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv":[[35,1,1,"","DecoderConcatLatentFcReshapeConv"]],"domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv.DecoderConcatLatentFcReshapeConv":[[35,2,1,"","cal_p_x_pars_loc_scale"],[35,2,1,"","concat_ydx"],[35,2,1,"","concat_ytdx"],[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv_gated_conv":[[35,1,1,"","DecoderConcatLatentFCReshapeConvGatedConv"]],"domainlab.compos.vae.compos.decoder_concat_vec_reshape_conv_gated_conv.DecoderConcatLatentFCReshapeConvGatedConv":[[35,3,1,"","training"]],"domainlab.compos.vae.compos.decoder_cond_prior":[[35,1,1,"","LSCondPriorLinearBnReluLinearSoftPlus"]],"domainlab.compos.vae.compos.decoder_cond_prior.LSCondPriorLinearBnReluLinearSoftPlus":[[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.decoder_losses":[[35,1,1,"","NLLPixelLogistic256"]],"domainlab.compos.vae.compos.encoder":[[35,1,1,"","LSEncoderConvBnReluPool"],[35,1,1,"","LSEncoderLinear"]],"domainlab.compos.vae.compos.encoder.LSEncoderConvBnReluPool":[[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder.LSEncoderLinear":[[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_dirichlet":[[35,1,1,"","EncoderH2Dirichlet"]],"domainlab.compos.vae.compos.encoder_dirichlet.EncoderH2Dirichlet":[[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_domain_topic":[[35,1,1,"","EncoderImg2TopicDirZd"]],"domainlab.compos.vae.compos.encoder_domain_topic.EncoderImg2TopicDirZd":[[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_domain_topic_img2topic":[[35,1,1,"","EncoderImg2TopicDistri"]],"domainlab.compos.vae.compos.encoder_domain_topic_img2topic.EncoderImg2TopicDistri":[[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_domain_topic_img_topic2zd":[[35,1,1,"","EncoderSandwichTopicImg2Zd"]],"domainlab.compos.vae.compos.encoder_domain_topic_img_topic2zd.EncoderSandwichTopicImg2Zd":[[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_xyd_parallel":[[35,1,1,"","XYDEncoderParallel"],[35,1,1,"","XYDEncoderParallelAlex"],[35,1,1,"","XYDEncoderParallelConvBnReluPool"],[35,1,1,"","XYDEncoderParallelExtern"],[35,1,1,"","XYDEncoderParallelUser"]],"domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallel":[[35,2,1,"","forward"],[35,2,1,"","infer_zy_loc"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallelAlex":[[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallelConvBnReluPool":[[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallelExtern":[[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_xyd_parallel.XYDEncoderParallelUser":[[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_xydt_elevator":[[35,1,1,"","XYDTEncoderArg"],[35,1,1,"","XYDTEncoderElevator"]],"domainlab.compos.vae.compos.encoder_xydt_elevator.XYDTEncoderArg":[[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_xydt_elevator.XYDTEncoderElevator":[[35,2,1,"","forward"],[35,2,1,"","infer_zy_loc"],[35,3,1,"","training"]],"domainlab.compos.vae.compos.encoder_zy":[[35,1,1,"","EncoderConnectLastFeatLayer2Z"]],"domainlab.compos.vae.compos.encoder_zy.EncoderConnectLastFeatLayer2Z":[[35,2,1,"","forward"],[35,3,1,"","training"]],"domainlab.compos.vae.utils_request_chain_builder":[[34,1,1,"","VAEChainNodeGetter"]],"domainlab.compos.vae.zoo_vae_builders_classif":[[34,1,1,"","ChainNodeVAEBuilderClassifCondPriorBase"],[34,1,1,"","NodeVAEBuilderArg"],[34,1,1,"","NodeVAEBuilderImgAlex"],[34,1,1,"","NodeVAEBuilderImgConvBnPool"],[34,1,1,"","NodeVAEBuilderUser"]],"domainlab.compos.vae.zoo_vae_builders_classif.ChainNodeVAEBuilderClassifCondPriorBase":[[34,2,1,"","build_decoder"],[34,2,1,"","build_encoder"],[34,2,1,"","config_img"],[34,2,1,"","is_myjob"]],"domainlab.compos.vae.zoo_vae_builders_classif.NodeVAEBuilderArg":[[34,2,1,"","build_encoder"],[34,2,1,"","is_myjob"]],"domainlab.compos.vae.zoo_vae_builders_classif.NodeVAEBuilderImgAlex":[[34,2,1,"","build_encoder"],[34,2,1,"","is_myjob"]],"domainlab.compos.vae.zoo_vae_builders_classif.NodeVAEBuilderImgConvBnPool":[[34,2,1,"","build_encoder"],[34,2,1,"","is_myjob"]],"domainlab.compos.vae.zoo_vae_builders_classif.NodeVAEBuilderUser":[[34,2,1,"","build_encoder"],[34,2,1,"","is_myjob"]],"domainlab.compos.vae.zoo_vae_builders_classif_topic":[[34,1,1,"","NodeVAEBuilderImgTopic"]],"domainlab.compos.vae.zoo_vae_builders_classif_topic.NodeVAEBuilderImgTopic":[[34,2,1,"","build_decoder"],[34,2,1,"","build_encoder"],[34,2,1,"","is_myjob"]],"domainlab.compos.zoo_nn":[[31,1,1,"","FeatExtractNNBuilderChainNodeGetter"]],"domainlab.dsets":[[36,0,0,"-","a_dset_mnist_color_rgb_solo"],[36,0,0,"-","dset_img_path_list"],[36,0,0,"-","dset_mnist_color_solo_default"],[36,0,0,"-","dset_poly_domains_mnist_color_default"],[36,0,0,"-","dset_subfolder"],[36,0,0,"-","utils_color_palette"],[36,0,0,"-","utils_data"],[36,0,0,"-","utils_wrapdset_patches"]],"domainlab.dsets.a_dset_mnist_color_rgb_solo":[[36,1,1,"","ADsetMNISTColorRGBSolo"]],"domainlab.dsets.a_dset_mnist_color_rgb_solo.ADsetMNISTColorRGBSolo":[[36,2,1,"","get_background_color"],[36,2,1,"","get_foreground_color"],[36,2,1,"","get_num_colors"]],"domainlab.dsets.dset_img_path_list":[[36,1,1,"","DsetImPathList"]],"domainlab.dsets.dset_img_path_list.DsetImPathList":[[36,2,1,"","get_list_tuple_img_label"]],"domainlab.dsets.dset_mnist_color_solo_default":[[36,1,1,"","DsetMNISTColorSoloDefault"]],"domainlab.dsets.dset_mnist_color_solo_default.DsetMNISTColorSoloDefault":[[36,2,1,"","get_background_color"],[36,2,1,"","get_foreground_color"],[36,2,1,"","get_num_colors"],[36,4,1,"","palette"]],"domainlab.dsets.dset_poly_domains_mnist_color_default":[[36,1,1,"","DsetMNISTColorMix"],[36,1,1,"","DsetMNISTColorMixNoDomainLabel"]],"domainlab.dsets.dset_subfolder":[[36,1,1,"","DsetSubFolder"],[36,5,1,"","fetch_img_paths"],[36,5,1,"","has_file_allowed_extension"]],"domainlab.dsets.utils_data":[[36,1,1,"","DsetInMemDecorator"],[36,5,1,"","fun_img_path_loader_default"],[36,5,1,"","mk_fun_label2onehot"],[36,5,1,"","plot_ds"],[36,5,1,"","plot_ds_list"]],"domainlab.dsets.utils_wrapdset_patches":[[36,1,1,"","WrapDsetPatches"]],"domainlab.dsets.utils_wrapdset_patches.WrapDsetPatches":[[36,2,1,"","get_tile"]],"domainlab.exp":[[37,0,0,"-","exp_cuda_seed"]],"domainlab.exp.exp_cuda_seed":[[37,5,1,"","set_seed"]],"domainlab.exp_protocol":[[38,0,0,"-","aggregate_results"]],"domainlab.exp_protocol.aggregate_results":[[38,5,1,"","agg_from_directory"],[38,5,1,"","agg_main"],[38,5,1,"","agg_results"]],"domainlab.models":[[39,0,0,"-","a_model"],[39,0,0,"-","a_model_classif"],[39,0,0,"-","args_jigen"],[39,0,0,"-","args_vae"],[39,0,0,"-","interface_vae_xyd"],[39,0,0,"-","model_custom"],[39,0,0,"-","model_dann"],[39,0,0,"-","model_diva"],[39,0,0,"-","model_erm"],[39,0,0,"-","model_hduva"],[39,0,0,"-","model_jigen"],[39,0,0,"-","model_vae_xyd_classif"]],"domainlab.models.a_model":[[39,1,1,"","AModel"]],"domainlab.models.a_model.AModel":[[39,2,1,"","cal_loss"],[39,2,1,"","cal_reg_loss"],[39,2,1,"","cal_task_loss"],[39,2,1,"","dset_decoration_args_algo"],[39,2,1,"","extend"],[39,2,1,"","extract_semantic_feat"],[39,2,1,"","forward"],[39,2,1,"","list_inner_product"],[39,2,1,"","load"],[39,4,1,"","metric4msel"],[39,4,1,"","multiplier4task_loss"],[39,4,1,"","name"],[39,4,1,"","net_invar_feat"],[39,4,1,"","p_na_prefix"],[39,2,1,"","print_parameters"],[39,2,1,"","reset_aux_net"],[39,2,1,"","reset_feature_extractor"],[39,2,1,"","save"],[39,2,1,"","set_saver"],[39,3,1,"","training"]],"domainlab.models.a_model_classif":[[39,1,1,"","AModelClassif"]],"domainlab.models.a_model_classif.AModelClassif":[[39,2,1,"","cal_logit_y"],[39,2,1,"","cal_loss_gen_adv"],[39,2,1,"","cal_perf_metric"],[39,2,1,"","cal_task_loss"],[39,2,1,"","create_perf_obj"],[39,4,1,"","dim_y"],[39,2,1,"","evaluate"],[39,2,1,"","extend"],[39,2,1,"","extract_semantic_feat"],[39,2,1,"","infer_y_vpicn"],[39,3,1,"","match_feat_fun_na"],[39,4,1,"","metric4msel"],[39,4,1,"","net_classifier"],[39,2,1,"","pred2file"],[39,2,1,"","read_prediction_file"],[39,3,1,"","training"]],"domainlab.models.args_jigen":[[39,5,1,"","add_args2parser_jigen"]],"domainlab.models.args_vae":[[39,5,1,"","add_args2parser_vae"]],"domainlab.models.interface_vae_xyd":[[39,1,1,"","InterfaceVAEXYD"]],"domainlab.models.interface_vae_xyd.InterfaceVAEXYD":[[39,2,1,"","init"],[39,2,1,"","init_p_zx4batch"]],"domainlab.models.model_custom":[[39,1,1,"","AModelCustom"]],"domainlab.models.model_custom.AModelCustom":[[39,2,1,"","dict_net_module_na2arg_na"],[39,3,1,"","training"]],"domainlab.models.model_dann":[[39,5,1,"","mk_dann"]],"domainlab.models.model_diva":[[39,5,1,"","mk_diva"]],"domainlab.models.model_erm":[[39,5,1,"","mk_erm"]],"domainlab.models.model_hduva":[[39,5,1,"","mk_hduva"]],"domainlab.models.model_jigen":[[39,5,1,"","mk_jigen"]],"domainlab.models.model_vae_xyd_classif":[[39,1,1,"","VAEXYDClassif"]],"domainlab.models.model_vae_xyd_classif.VAEXYDClassif":[[39,2,1,"","extract_semantic_feat"],[39,4,1,"","multiplier4task_loss"],[39,3,1,"","training"]],"domainlab.tasks":[[40,0,0,"-","a_task"],[40,0,0,"-","a_task_classif"],[40,0,0,"-","b_task"],[40,0,0,"-","b_task_classif"],[40,0,0,"-","task_dset"],[40,0,0,"-","task_folder"],[40,0,0,"-","task_folder_mk"],[40,0,0,"-","task_mini_vlcs"],[40,0,0,"-","task_mnist_color"],[40,0,0,"-","task_pathlist"],[40,0,0,"-","task_utils"],[40,0,0,"-","utils_task"],[40,0,0,"-","utils_task_dset"],[40,0,0,"-","zoo_tasks"]],"domainlab.tasks.a_task":[[40,1,1,"","NodeTaskDG"]],"domainlab.tasks.a_task.NodeTaskDG":[[40,2,1,"","conf_without_args"],[40,2,1,"","get_list_domains"],[40,2,1,"","get_list_domains_tr_te"],[40,2,1,"","get_na"],[40,2,1,"","init_business"],[40,2,1,"","is_myjob"],[40,4,1,"","isize"],[40,4,1,"","list_domain_tr"],[40,4,1,"","loader_te"],[40,4,1,"","loader_tr"],[40,4,1,"","loader_val"],[40,2,1,"","set_list_domains"],[40,4,1,"","task_name"]],"domainlab.tasks.a_task_classif":[[40,1,1,"","NodeTaskDGClassif"]],"domainlab.tasks.a_task_classif.NodeTaskDGClassif":[[40,4,1,"","dim_y"],[40,4,1,"","list_str_y"],[40,2,1,"","sample_sav"]],"domainlab.tasks.b_task":[[40,1,1,"","NodeTaskDict"]],"domainlab.tasks.b_task.NodeTaskDict":[[40,2,1,"","decorate_dset"],[40,2,1,"","get_dset_by_domain"],[40,2,1,"","init_business"]],"domainlab.tasks.b_task_classif":[[40,1,1,"","NodeTaskDictClassif"]],"domainlab.tasks.b_task_classif.NodeTaskDictClassif":[[40,2,1,"","count_domain_class"],[40,2,1,"","init_business"]],"domainlab.tasks.task_dset":[[40,5,1,"","mk_task_dset"]],"domainlab.tasks.task_folder":[[40,1,1,"","NodeTaskFolder"],[40,1,1,"","NodeTaskFolderClassNaMismatch"]],"domainlab.tasks.task_folder.NodeTaskFolder":[[40,4,1,"","dict_domain2imgroot"],[40,4,1,"","extensions"],[40,2,1,"","get_dset_by_domain"]],"domainlab.tasks.task_folder.NodeTaskFolderClassNaMismatch":[[40,2,1,"","get_dset_by_domain"]],"domainlab.tasks.task_folder_mk":[[40,5,1,"","mk_task_folder"]],"domainlab.tasks.task_mini_vlcs":[[40,5,1,"","addtask2chain"]],"domainlab.tasks.task_mnist_color":[[40,1,1,"","NodeTaskMNISTColor10"]],"domainlab.tasks.task_mnist_color.NodeTaskMNISTColor10":[[40,2,1,"","get_dset_by_domain"],[40,2,1,"","get_list_domains"],[40,4,1,"","isize"],[40,4,1,"","list_str_y"]],"domainlab.tasks.task_pathlist":[[40,1,1,"","NodeTaskPathListDummy"],[40,5,1,"","mk_node_task_path_list"]],"domainlab.tasks.task_pathlist.NodeTaskPathListDummy":[[40,2,1,"","get_dset_by_domain"]],"domainlab.tasks.task_utils":[[40,5,1,"","parse_domain_id"]],"domainlab.tasks.utils_task":[[40,1,1,"","DsetClassVecDecorator"],[40,1,1,"","DsetClassVecDecoratorImgPath"],[40,1,1,"","DsetDomainVecDecorator"],[40,1,1,"","DsetDomainVecDecoratorImgPath"],[40,1,1,"","ImSize"],[40,1,1,"","LoaderDomainLabel"],[40,5,1,"","img_loader2dir"],[40,5,1,"","mk_loader"],[40,5,1,"","mk_onehot"],[40,5,1,"","tensor1hot2ind"]],"domainlab.tasks.utils_task.DsetClassVecDecorator":[[40,2,1,"","fun_class_local_na2vec_new"],[40,4,1,"","targets"]],"domainlab.tasks.utils_task.DsetDomainVecDecorator":[[40,4,1,"","targets"]],"domainlab.tasks.utils_task.ImSize":[[40,4,1,"","c"],[40,4,1,"","h"],[40,4,1,"","w"]],"domainlab.tasks.utils_task_dset":[[40,1,1,"","DsetIndDecorator4XYD"],[40,1,1,"","DsetZip"]],"domainlab.tasks.zoo_tasks":[[40,1,1,"","TaskChainNodeGetter"],[40,5,1,"","get_task"]],"domainlab.utils":[[41,0,0,"-","flows_gen_img_model"],[41,0,0,"-","generate_benchmark_plots"],[41,0,0,"-","get_git_tag"],[41,0,0,"-","hyperparameter_gridsearch"],[41,0,0,"-","hyperparameter_retrieval"],[41,0,0,"-","hyperparameter_sampling"],[41,0,0,"-","logger"],[41,0,0,"-","override_interface"],[41,0,0,"-","perf"],[41,0,0,"-","perf_metrics"],[41,0,0,"-","sanity_check"],[41,0,0,"-","test_img"],[41,0,0,"-","u_import"],[41,0,0,"-","u_import_net_module"],[41,0,0,"-","utils_class"],[41,0,0,"-","utils_classif"],[41,0,0,"-","utils_cuda"],[41,0,0,"-","utils_img_sav"]],"domainlab.utils.flows_gen_img_model":[[41,1,1,"","FlowGenImgs"],[41,5,1,"","fun_gen"]],"domainlab.utils.flows_gen_img_model.FlowGenImgs":[[41,2,1,"","gen_img_loader"],[41,2,1,"","gen_img_xyd"]],"domainlab.utils.generate_benchmark_plots":[[41,5,1,"","boxplot"],[41,5,1,"","boxplot_stochastic"],[41,5,1,"","boxplot_systematic"],[41,5,1,"","gen_benchmark_plots"],[41,5,1,"","gen_plots"],[41,5,1,"","max_0_x"],[41,5,1,"","radar_plot"],[41,5,1,"","round_vals_in_dict"],[41,5,1,"","scatterplot"],[41,5,1,"","scatterplot_matrix"]],"domainlab.utils.get_git_tag":[[41,5,1,"","get_git_tag"]],"domainlab.utils.hyperparameter_gridsearch":[[41,5,1,"","add_next_param_from_list"],[41,5,1,"","add_references_and_check_constraints"],[41,5,1,"","add_shared_params_to_param_grids"],[41,5,1,"","build_param_grid_of_shared_params"],[41,5,1,"","grid_task"],[41,5,1,"","lognormal_grid"],[41,5,1,"","loguniform_grid"],[41,5,1,"","normal_grid"],[41,5,1,"","rais_error_if_num_not_specified"],[41,5,1,"","round_to_discreate_grid_normal"],[41,5,1,"","round_to_discreate_grid_uniform"],[41,5,1,"","sample_grid"],[41,5,1,"","sample_gridsearch"],[41,5,1,"","uniform_grid"]],"domainlab.utils.hyperparameter_retrieval":[[41,5,1,"","get_gamma_reg"]],"domainlab.utils.hyperparameter_sampling":[[41,1,1,"","CategoricalHyperparameter"],[41,1,1,"","Hyperparameter"],[41,1,1,"","ReferenceHyperparameter"],[41,1,1,"","SampledHyperparameter"],[41,5,1,"","check_constraints"],[41,5,1,"","create_samples_from_shared_samples"],[41,5,1,"","get_hyperparameter"],[41,5,1,"","get_shared_samples"],[41,5,1,"","is_dict_with_key"],[41,5,1,"","sample_hyperparameters"],[41,5,1,"","sample_parameters"],[41,5,1,"","sample_task"],[41,5,1,"","sample_task_only_shared"]],"domainlab.utils.hyperparameter_sampling.CategoricalHyperparameter":[[41,2,1,"","datatype"],[41,2,1,"","sample"]],"domainlab.utils.hyperparameter_sampling.Hyperparameter":[[41,2,1,"","datatype"],[41,2,1,"","get_val"],[41,2,1,"","sample"]],"domainlab.utils.hyperparameter_sampling.ReferenceHyperparameter":[[41,2,1,"","datatype"],[41,2,1,"","sample"]],"domainlab.utils.hyperparameter_sampling.SampledHyperparameter":[[41,2,1,"","datatype"],[41,2,1,"","sample"]],"domainlab.utils.logger":[[41,1,1,"","Logger"]],"domainlab.utils.logger.Logger":[[41,2,1,"","get_logger"],[41,3,1,"","logger"]],"domainlab.utils.override_interface":[[41,5,1,"","override_interface"]],"domainlab.utils.perf":[[41,1,1,"","PerfClassif"]],"domainlab.utils.perf.PerfClassif":[[41,2,1,"","cal_acc"],[41,2,1,"","gen_fun_acc"],[41,2,1,"","get_list_pred_target"]],"domainlab.utils.perf_metrics":[[41,1,1,"","PerfMetricClassif"]],"domainlab.utils.perf_metrics.PerfMetricClassif":[[41,2,1,"","cal_metrics"]],"domainlab.utils.sanity_check":[[41,1,1,"","SanityCheck"]],"domainlab.utils.sanity_check.SanityCheck":[[41,2,1,"","dataset_sanity_check"],[41,2,1,"","save_san_check_for_domain"]],"domainlab.utils.test_img":[[41,5,1,"","mk_img"],[41,5,1,"","mk_rand_label_onehot"],[41,5,1,"","mk_rand_xyd"]],"domainlab.utils.u_import":[[41,5,1,"","import_path"]],"domainlab.utils.u_import_net_module":[[41,5,1,"","build_external_obj_net_module_feat_extract"]],"domainlab.utils.utils_class":[[41,5,1,"","store_args"]],"domainlab.utils.utils_classif":[[41,5,1,"","get_label_na"],[41,5,1,"","logit2preds_vpic"],[41,5,1,"","mk_dummy_label_list_str"]],"domainlab.utils.utils_cuda":[[41,5,1,"","get_device"]],"domainlab.utils.utils_img_sav":[[41,5,1,"","mk_fun_sav_img"],[41,5,1,"","sav_add_title"]],domainlab:[[26,0,0,"-","algos"],[25,0,0,"-","arg_parser"],[31,0,0,"-","compos"],[36,0,0,"-","dsets"],[37,0,0,"-","exp"],[38,0,0,"-","exp_protocol"],[25,5,1,"","g_list_model_penalized_reg_agg"],[39,0,0,"-","models"],[40,0,0,"-","tasks"],[41,0,0,"-","utils"]]},objnames:{"0":["py","module","Python module"],"1":["py","class","Python class"],"2":["py","method","Python method"],"3":["py","attribute","Python attribute"],"4":["py","property","Python property"],"5":["py","function","Python function"]},objtypes:{"0":"py:module","1":"py:class","2":"py:method","3":"py:attribute","4":"py:property","5":"py:function"},terms:{"0":[0,1,2,3,5,6,7,8,9,11,12,15,18,22,24,30,32,35,39,40,41,44],"000":9,"001":11,"00322":0,"00390625":35,"01":11,"01719":12,"05":[6,24,30],"07500":6,"09436":[2,39],"1":[0,1,2,3,4,5,6,7,8,11,12,14,15,19,24,26,29,31,32,33,34,35,36,39,40,41,44],"10":[7,9,32,40,41],"100":[4,11,36],"100000":44,"1024":32,"10427":15,"10985":1,"10_0000":41,"10e5":15,"11":[29,32],"11646":18,"12":32,"13848":37,"1412":0,"14144163":8,"14356":44,"16":[7,11],"1607":12,"17":[14,39],"1905":15,"1912":1,"192":32,"1e":7,"1e4":11,"1e5":[0,7,11,15,22,24],"1e6":11,"1st":22,"1st_domain":22,"2":[0,1,3,5,6,7,8,11,13,15,24,29,32,33,34,35,36,39,40,41,42,44],"2006":6,"2016":[14,39],"2018":19,"2019":[4,39],"2020":[15,39],"2021":[2,39],"2022":[1,29],"2024":44,"2030":[14,39],"2096":[14,39],"2101":[2,39],"2104":0,"2206":18,"224":[22,40],"225":22,"229":22,"2403":44,"256":[32,35],"28":31,"2_2":1,"2e4":11,"2nd":22,"2nd_domain":22,"3":[0,2,3,4,5,6,7,8,11,12,13,15,17,18,26,29,31,32,33,35,36,39,40,41],"30":11,"31":[11,39],"32":[7,8,11,12,18,19,22,29,36],"384":32,"3rd":22,"3rd_domain":22,"4":[2,7,8,11,22,32,33,36,40],"406":22,"4096":32,"41":8,"456":22,"485":22,"5":[5,7,8,11,22,29,31,32,40,41],"50":11,"500":7,"6":[6,7,8,13,32,37],"64":[7,11,29,32],"6572":0,"7":[0,6,7,8,12,15,18,22,32],"700000":44,"797":7,"798":7,"7e5":[0,2,6,15,22,24],"8":[4,7,11,13,19,29,32,36,41],"80":11,"800":7,"832":7,"84":11,"858":7,"88428":36,"9":[7,17,32],"9216":32,"958":7,"959":7,"96":11,"961":7,"978":7,"995":7,"999":7,"abstract":[26,27,28,29,33,34,36,39,40],"boolean":[27,32],"break":[29,33],"case":[6,7,8,9,22,40,41],"class":[2,6,15,23,24,25,26,27,28,29,30,31,32,33,34,35,36,39,40,41,44],"const":25,"default":[8,9,11,21,24,25,39,40],"do":[6,8,11,20,22,23,25,27,31,33,34,40,41],"final":[1,2,6,27,29,41],"float":[9,11,39,41],"function":[1,3,6,8,20,22,24,29,31,32,33,34,36,38,39,40,41,42,44],"import":[22,40,41],"int":[9,11,35,39,40,41],"long":25,"new":[2,15,20,23,26,29,39,40,44],"public":14,"ram\u00e9":29,"return":[20,22,26,27,28,29,31,32,33,34,35,36,39,40,41],"static":[32,39,40,41,44],"super":[23,39],"switch":11,"true":[6,9,28,29,32,33,34,36,40,41],"try":[6,8],"var":[1,29,35],"while":[0,6,7,23,32,34],A:[1,7,22,33,39,41,44],At:[0,11],By:[6,15,21],FOR:42,For:[1,2,3,4,5,6,7,8,9,11,20,22,23,24,28,29,32,33,34,39,40,44],If:[8,9,11,20,32,41],In:[0,2,3,6,8,9,11,15,20,22,24,31,32,44],It:[0,8,17,20,22,32],No:[11,19],One:[3,8,11,20,39,44],The:[0,1,2,3,4,6,7,8,9,11,13,14,15,20,21,22,26,29,32,34,39,40,41,44],There:[11,20],These:15,To:[4,8,9,10,13,20,22,24,33,39,44],_0:0,_:[1,3],__doc__:22,__get__item:40,__init__:[16,33],_cal_reg_loss:20,_d:3,_e:1,_list_domain:40,_model:29,_modul:36,a_algo_build:[16,25,43],a_dset_mnist_color_rgb_solo:[16,25,43],a_model:[16,25,43],a_model_build:[25,31],a_model_classif:[16,25,43],a_model_sel:[25,26],a_nn_build:[16,25,43],a_observ:[25,26],a_task:[16,25,43],a_task_classif:[16,25,43],a_train:[25,26],a_vae_build:[25,31],aaai:19,ab:[0,1],abil:8,abort:9,about:[7,11,24,27,41],abov:[4,8,10,11,20,27,44],absolut:[22,40],abstractchainnodehandl:[26,29,31,33,34,40],abstractfeatextractnnbuilderchainnod:31,abstractmodelbuilderchainnod:34,abstracttrain:29,abstractvaebuilderchainnod:34,acc:[7,8,27,39,41],acc_val:27,accept:[26,27,28,32,33,41],access:8,accomod:29,accord:[22,34,41],accuaraci:27,accuraci:15,aconf:29,across:[1,3,6,22,27,29,40],action:[23,25,39],activ:[8,12,17,32],actual:[0,15,41],ad:[2,26,29,31,39,40],adam:11,adapt:[6,29,33,42],adaptiveavgpool2d:32,adaptor:34,adaptorreconvaexyd:34,add:[11,22,31,40,41],add_args2parser_di:29,add_args2parser_jigen:39,add_args2parser_matchdg:30,add_args2parser_va:39,add_next_param_from_list:41,add_references_and_check_constraint:41,add_shared_params_to_param_grid:41,addit:[2,15,23,24,39],addition:[6,15,27,39,40],address:[9,15],addtask2chain:40,adsetmnistcolorrgbsolo:36,adv:0,advantag:[4,15],adversari:[24,26,29,39,42],ae:34,affect:[1,6,27,33],after:[6,8,20,21,23,28,29,39,41],after_al:28,after_batch:[23,29],after_epoch:[23,29],afterward:[32,41],again:1,against:[8,41],agg_from_directori:38,agg_main:38,agg_partial_bm:[8,24],agg_precision_recall_f1:41,agg_result:[38,41],aggreg:[8,21,24,25,38,39,41],aggregate_result:[16,25,43],aggregr:38,aggrst:21,aggtag:24,aim:44,al:[3,4,14,19,29,39],alex4deepal:32,alex:[29,42],alexandr:[1,29],alexej:[29,44],alexnet:[1,2,4,6,13,15,19,22,24,32],alexnetbas:32,alexnetnolastlay:32,algo:[8,16,25,41,43],algobuilderchainnodegett:26,algorithm:[0,6,7,11,20,21,24,26,29,31,39,44],algoritm:46,alia:32,align:[1,29,42,44],all:[1,3,6,7,8,9,11,15,20,22,24,29,32,33,36,38,39,40,41,44],allow:[15,22,23,36,40],along:[2,25,39,41,44],alpha:[0,2,32,39],alpha_d:15,alpha_i:15,alreadi:[7,11,22,36],also:[4,7,8,9,11,15,20,22,24,26,32,33,36,39,41,42,44],alter:21,altern:[2,8],although:32,alwai:[10,15,29,31,33,34,40,41],amodel:[20,39],amodelclassif:[20,39],amodelcustom:[20,39],among:[6,11],amongst:9,amsel:27,an:[0,1,2,3,4,6,7,8,9,15,20,21,22,24,26,29,32,33,36,39,40,41,44],ani:[6,29,32,44],anoth:[4,6,9,11,22,29,32,35,39,40],anymor:41,anyth:[8,23,41],aobvisitor:28,apart:9,apath:[22,24,26],api:[26,29],appear:[7,11],append:[21,29,40,44],appendix:[3,29],appli:[1,9,11,20,23,24,25,32,41],apply_dict_to_arg:25,approach:[2,6,20,39],approx:1,approxim:1,ar:[0,1,2,4,6,7,8,9,11,15,20,21,22,24,25,29,32,39,40,41,44],arbitrari:[32,39],architectur:41,archiv:0,arg:[3,6,25,26,28,29,30,31,32,33,35,36,39,40,41],arg_nam:35,arg_name4net:31,arg_name_net_path:31,arg_name_of_net:31,arg_pars:[16,20,42,43],arg_path_nam:35,arg_path_of_net:31,arg_val:31,argpars:[25,29],args_dial:[25,26],args_jigen:[16,25,43],args_nnam:31,args_va:[16,25,43],argument:[2,8,9,13,21,22,25,26,31,32,33,34,39,40,41],arjovski:[3,29],armin:44,around:[1,29,40],arpit:[5,29],arrang:35,arrow:[2,6],art:44,artifici:19,arxiv:[0,1,2,6,12,15,18,39,44],aspect:8,associ:[6,44],assum:[8,22,30,36,40],assumend:40,assumpt:[1,6],attack:0,attempt:41,attribut:[26,32,34,41],au:29,augment:44,auroc:[7,8,41],author:[2,6,44],auto:[22,39,42],autoencod:[2,39,42],autograd:32,autogradfunmultipli:32,autogradfunreversemultipli:32,automat:32,auxiliari:[2,44],auxilliari:26,avail:[0,6,7,9,24,33,40],averag:[1,29,42],avgpool:32,avoid:[10,31,33,34,40],b:[1,6,44],b_obvisitor:[25,26],b_task:[16,25,43],b_task_classif:[16,25,43],babu:44,back:41,backbon:26,background:[7,36],backpack:[1,29],backprop:1,backward:[23,32],bar:[3,44,46],base:[2,6,8,20,23,25,26,27,28,29,30,31,32,33,34,35,36,39,40,41],base_upd:27,baseclass:41,bash:[8,44],basi:[7,24],basic:[29,40],batch:[3,6,7,9,11,23,24,25,29,32,35,39,40,41,44],batch_idx:29,batch_siz:[36,39,40,41],batch_tensor_ref_domain2each:29,batchnorm2d:7,batchsiz:[22,29,31,36],bb5af1d77658133af8be8c9b1a13139722315c3a:36,becaus:20,becom:[1,8],been:8,beer:44,befor:[6,7,8,11,23,24,29],before_batch:[23,29],before_epoch:[23,29],before_tr:[23,29],begin:[6,23],behavior:[23,33],behaviour:[9,41],being:[1,4,9,20,22,39],below:[1,3,8,9,13,15,21,22,27,44],benchmark:[13,16,24,38,41],benefit:39,best:27,best_te_metr:27,best_val_acc:27,beta:2,beta_d:[2,24,39],beta_i:[2,24,39],beta_t:[2,24,39],beta_x:[2,24,39],better:[15,40],between:[0,3,6,9,21,22,24,37,39,41,42],bi:3,bia:[7,32],big:0,biggl:1,bin:35,bin_siz:35,bioconda:8,bird:22,blob:[13,35,36,39],block:29,blue:7,bm_dir:38,booktitl:2,bool:[28,32,35,36,38,39,41],boost:[5,29,44],both:[1,4,6,7,8,20,33,34,36,39],bound:[9,42],boxplot:41,boxplot_stochast:41,boxplot_systemat:41,bridg:[34,35],brief:23,brokenpip:8,bs:[1,2,4,5,6,7,9,11,12,13,15,18,19,22,24,40,44],bs_match:30,bsize:40,buettner:[2,39],build:[2,23,26,34,39,40,41],build_decod:34,build_encod:34,build_external_obj_net_module_feat_extract:41,build_feat_extract_net:13,build_param_grid_of_shared_param:41,builder:[26,31,34],builder_api_model:[16,25,43],builder_custom:[16,25,43],builder_dann:[16,25,43],builder_diva:[16,25,43],builder_erm:[16,25,43],builder_hduva:[16,25,43],builder_jigen1:[16,25,43],builder_nn_alex:[16,25,43],builder_nn_conv_bn_relu_2:[16,25,43],builder_nn_external_from_fil:[16,25,43],built:20,bulk:23,busi:[31,33,34],c:[6,8,35,39,40,44],c_msel_oracl:[25,26],c_msel_tr_loss:[25,26],c_msel_val:[25,26],c_obvisitor_cleanup:[25,26],c_obvisitor_gen:[25,26],c_vae_adaptor_model_recon:[25,31],c_vae_builder_classif:[25,31],c_vae_recon:[25,31],caim:[5,29],cal:39,cal_acc:41,cal_dict_variance_grad:29,cal_lat:34,cal_logit_i:39,cal_loss:[29,39],cal_loss_gen_adv:39,cal_mean_across_dict:29,cal_metr:41,cal_oracle_perf:28,cal_p_x_pars_loc_scal:35,cal_perf_metr:39,cal_power_single_dict:29,cal_prior_zd:34,cal_prior_zi:34,cal_reg_loss:[29,39],cal_reg_loss_over_task_loss_ratio:29,cal_task_loss:[20,39],calcul:[20,28,29,35,39],call:[20,22,23,28,32,33,35,39,41],caltech:[2,4,6,13,15,19,22,24,44],can:[0,1,2,3,4,7,8,9,11,13,15,20,21,22,23,24,26,29,31,32,33,34,39,40,41,44],cannot:[0,11],car:22,care:32,carla:44,carlucci:[4,39],carri:[22,40],cartesian:[30,41],cat:8,categor:[9,11,41],categoricalhyperparamet:41,causal:[29,42],causalirl:18,cd:8,cdot:3,ce_:0,ceil_mod:32,certain:1,chain:[20,22,26,29,31,33,34,40],chain_node_build:39,chainnod:33,chainnodevaebuilderclassifcondprior:34,chainnodevaebuilderclassifcondpriorbas:34,chair:22,chang:[8,10,17,29,32,33,34,39,41,46],channel:[22,29,32,40],characterist:0,chart:8,check:[7,23,24,27,29,33,36,39,41,42],check_constraint:41,chen:44,child:[29,34,39,40,41],choic:[3,8,24,25],choos:[3,7,11,13,24,41,42],chosen:[6,11,36],circ:[3,6],cite:2,claasif:39,class1:40,class2:40,class_name_model:26,class_to_idx:36,classif:[0,3,4,6,15,20,22,24,39,40,41,42,44],classifdropoutrelulinear:32,classifi:[0,3,4,6,7,15,22,24,32,34,39],classificaiton:[3,39],classmethod:41,classsifi:2,clean:[8,28],clean_up:28,clear:[15,29],clear_count:27,cli:[16,42,43],clone:[8,17,20,44],close:15,cls_fun_nll_p_x:35,cmd:26,code:[1,8,10,13,20,22,26,31,33,44],coeff_reg:39,cohort:44,collect:[8,29],color:[36,40,42],color_schem:36,colorjitt:22,column:[38,41],com:[13,35,36,39,45],comb:26,combin:[0,5,7,26,29,36,38,41,44],come:[7,24,29,30,33],command:[7,8,20,21,22,25,31,33,37],commandlin:[34,39],comment:[21,22],commit:[8,21],common:[6,9,22,29,34,39,40],commonmark:45,compar:[7,21,24],comparison:[13,32],compat:[32,39],complement:40,complet:[8,21,23],complex:[2,22,39],compli:41,compo:[16,25,26,29,39,40,43],compon:[24,26,29,32,39,44],component_nam:41,compos:22,composit:[22,40],comprehens:24,comput:[1,4,20,29,32,35,39],concat_ydx:35,concat_ytdx:35,concaten:35,concatent:2,concatn:34,concentr:2,concept:[4,7,44],concret:[9,11,33,34,39],conda:[8,17],condit:[2,6,15,34],conf:[44,46],conf_samp:41,conf_without_arg:40,confer:[1,4,19,29,39],confid:39,config:[8,24,34,41],config_img:34,configur:[8,17,21,23,34,40,41],configuraiton:44,confus:24,connect:[31,32,35],consid:[1,8,9,29,36,41],consider:41,consist:[0,40],constraind:41,constraint:[3,11,29,41],construct:[2,6,11,31,33,34,39,40],construct_classifi:34,construct_cond_prior:34,constructor:[33,34,35,39],contain:[7,8,10,15,21,22,29,36,40,41,44],content:43,context:32,continu:[2,11],contradictori:9,contrast:[6,29],contribut:[6,24],control:[3,29],contructor:35,conv2d:[7,32],conv_bn_pool_2:[0,2,3,5,6,7,11,12,15,18,22,24],conv_strid:[31,32,35],conveni:21,converg:7,convert:[33,34,39,40],convex:[5,29],convolut:[4,29,31,32,35],coral:[29,42],cord:[1,29],core:[8,41],corentin:[1,29],correct:[4,7,20,22,40,41],correl:[4,6,29,42],correspond:[6,7,8,21,22,29,32,40,44],cosin:[6,24],cosinesimilar:30,could:[2,3,8,10,11,13,20,22,39,44],count:[24,40],count_domain_class:40,countefactu:34,counter:23,counterfactu:[21,34],coupl:[28,33,34],cover:22,cpu:[8,39],crash:11,creat:[0,4,8,9,11,22,29,30,33,39,40,41,44],create_perf_obj:39,create_samples_from_shared_sampl:41,creator:30,criterion:7,crope:36,cross:[0,3,4,20,29,39,44],cross_domain_mmd:29,csv:[21,38,41],csv_file:8,ctr:24,ctx:32,cuda:[8,24,37],current:[8,9,11,21,26,29,32,35,39,40,41],custom:[4,9,23,24,26,32,39,42],cv:8,cvf:[4,39],d:[2,3,6,15,29,34,35,39,40,44],d_1:44,d_2:44,d_dataset:41,d_dim:41,d_e:29,d_i:29,da:19,dan:39,dancett:[1,29],dangel:1,dann:[24,39,42,44],dann_diva:44,dash:[2,6],data:[2,4,6,7,15,23,24,25,29,32,34,36,39,40,41,44],datafram:41,dataframe_in:41,dataload:24,dataset:[0,10,11,24,29,30,36,39,40,41,44],dataset_sanity_check:41,datasetfold:36,datatyp:[9,11,41],datri:8,ddset:[29,39],deafult:40,debug:[4,6,13,15,18,19,22,24],decid:[11,28],decis:[11,44],declar:9,decod:[15,34,35,39],decoder_concat_vec_reshape_conv:[31,34],decoder_concat_vec_reshape_conv_gated_conv:[31,34],decoder_cond_prior:[31,34],decoder_loss:[31,34],decoderconcatlatentfcreshapeconv:35,decoderconcatlatentfcreshapeconvgatedconv:35,decompos:[2,3,39],decor:[26,28,29,36,39,40,41,44],decorate:[27,29,39],decorate_dset:40,decoupl:[27,33,34,44],decreas:15,deep:[15,29,39,42,44],deepal:39,def:[13,22,23,41],defin:[6,8,9,10,11,20,23,24,32,33,34,40,41],definit:[1,6,25,31],deleg:[29,32,39],delet:[21,24,32],delta_:29,demo_benchmark:8,demo_config_single_run_diva:24,demo_custom_model:22,demo_hypeparameter_sampl:9,demo_hyperparameter_sampl:9,demo_task_path_list_smal:[4,22],demonstr:0,denot:[1,2,3,4,6,39],dense_lay:32,densenet:[32,35],densiti:15,depend:[6,9,11,24,44],depth:32,descend:0,describ:0,descript:[7,41],design:[33,44],desir:6,despit:4,dest:[25,41],destin:41,detail:[1,3,9,11,14,23,24,29,39,44],determin:[39,41],devansh:[5,29],develop:[23,44],deviat:[9,11],devic:[20,24,28,29,34,35,39,41],df_column_in:41,dg:40,diag:1,diagon:[1,29],dial:[9,11,42,44],dial_epsilon:[0,24],dial_lr:[0,24],dial_noise_scal:[0,24],dial_steps_perturb:[0,24],dict:[25,30,41],dict_class_label_ind2nam:40,dict_d2filepath_list_img_t:40,dict_d2filepath_list_img_tr:40,dict_d2filepath_list_img_v:40,dict_data:29,dict_domain2imgroot:[22,40],dict_domain_folder_name2class:[22,40],dict_domain_img_tran:[22,40],dict_folder_name2class_glob:40,dict_net_module_na2arg_na:[20,39],dict_param_grid:41,dictionari:[20,22,24,26,29,30,34,39,40,41],differ:[0,1,2,6,7,8,9,11,15,21,22,24,26,30,31,37,38,39,40,41],differenti:32,digit:[7,41],dilat:32,dim:[3,29,36,39,40,41],dim_cf:34,dim_d:40,dim_i:[13,30,32,39,40,41],dim_input:35,dim_out:31,dim_out_h:32,dim_target:41,dim_top:35,dimens:[3,24,29,30,31,32,34,39,40,41],dimension:2,dimes:11,diminish:25,direct:[8,29,32],directli:[6,11,32,35],director:29,directori:[8,9,10,11,17,20,21,22,24,36,38,40],diretori:10,dirichlet:[2,35],disabl:24,discourag:10,discrep:29,discrimin:0,discuss:[36,37],disentangl:[2,15,39],disk:[21,28,39,40],dispatch:[8,28,40],dist:6,dist_cosine_agg:30,dist_pairwise_cosin:30,distanc:[6,29,30],distinct:2,distinguish:41,distinguish_hyperparam:41,distinguish_param_setup:41,distribut:[2,3,9,24,29,35,39,40,41,42],diva:[11,22,24,26,31,39,42,44],diva_fbopt_ful:8,diverg:2,divid:[3,35],dmem:[9,24],doc:[32,37],doc_benchmark_yaml:8,doc_task:7,document:[8,9,22,44],doe:[0,4,6,8,11,15,22,27,30,33,40,41],dog:22,domain1:[22,40],domain3:22,domain:[3,4,5,9,11,20,24,26,27,28,29,30,31,34,36,39,40,41,42],domain_classification_loss:39,domainlab:[0,1,2,3,7,11,13,16,20],domainlab_arg:[9,11],domainlab_py39:17,don:[22,27],done:[1,6,7,8,11,20,28,29],dot:44,doubl:40,download:[24,44],dpath:24,drive:22,drop_last:40,dropout:32,ds_list:36,dset1:40,dset2:40,dset:[16,25,40,43],dset_decoration_args_algo:[29,39],dset_img_path_list:[16,25,43],dset_mnist_color_solo_default:[16,25,43],dset_poly_domains_mnist_color_default:[16,25,43],dset_subfold:[16,25,43],dsetclassvecdecor:40,dsetclassvecdecoratorimgpath:40,dsetdomainvecdecor:40,dsetdomainvecdecoratorimgpath:40,dsetimpathlist:36,dsetinddecorator4xyd:40,dsetinmemdecor:36,dsetmnistcolormix:36,dsetmnistcolormixnodomainlabel:36,dsetmnistcolorsolodefault:36,dsetsubfold:36,dsetzip:40,due:[8,29],dummi:33,dummybusi:33,dummychainnodehandlerbeav:33,dummychainnodehandlerlazi:33,dump:[39,41],dump_predict:28,dure:[0,2,8,15,29,32,39,42],e:[1,3,4,8,9,11,24,29,31,32,33,39,41,44],e_:2,each:[1,3,5,6,8,9,11,15,21,23,28,29,30,31,32,34,35,36,39,40,41,44],earli:[7,9,11,24,27],early_stop:27,easier:29,easili:20,effici:1,eiter:24,either:[11,24,29,32,34,40,41],elbo:[2,24,39],element:[29,39],ell:[3,29,44],els:[28,41],embed:[39,41],emper:[22,39],empir:[1,3,39,44],enabl:[24,40],encod:[15,31,32,34,39,42],encoder_dirichlet:[31,34],encoder_domain_top:[31,34],encoder_domain_topic_img2top:[31,34],encoder_domain_topic_img_topic2zd:[31,34],encoder_xyd_parallel:[31,34],encoder_xydt_elev:[31,34],encoder_zi:[31,34],encoderconnectlastfeatlayer2z:35,encoderh2dirichlet:35,encoderimg2topicdirzd:35,encoderimg2topicdistri:35,encodersandwichtopicimg2zd:35,encourag:[2,3,15],end:[6,7,24,35,36],endse:[9,11],enforc:[6,9,32],enhanc:44,ensembl:[5,29],ensur:[0,7,23,31,33,34,37],enter:39,entir:8,entri:[22,29,39],entropi:[0,3,4,20,39,44],enumer:23,environ:44,epo:[5,6,7,8,9,11,12,13,24,41,44],epo_reg_loss_tr:29,epo_t:24,epoch:[2,5,6,7,9,11,23,24,27,28,29,44],epochs_ctr:[6,13,24],epos_min:24,epos_per_match_upd:[6,24],epsilon:[0,1,29],eq:15,equal:9,equival:32,erm:[0,1,3,5,6,12,18,19,22,24,26,29,32,39],err:8,error:[0,9,41],es:[9,11,24],establish:23,estim:[1,3,29,41],et:[3,4,14,19,29,39],evalu:[11,28,39,40,44],even:[2,11,31,33,34,39],everi:[6,24,32,39],evid:42,exact:[7,41],exactli:[11,30],exampl:[5,7,8,9,10,11,20,29,33,39,42],except:[39,40],execut:[8,11,40,41],exist:[2,15,21,22,31,39],exp:[6,16,20,25,26,43],exp_cuda_se:[16,25,43],exp_main:[16,25,43],exp_protocol:[16,25,43],exp_util:[16,25,43],expect:[22,41],experi:[6,8,9,11,24,26,33,44],explain:[0,4,8,14,22],explicitli:41,exponenti:29,express:11,exptag:24,extend:[4,24,25,26,29,32,39],extens:[22,36,40],extern:[20,26,41],extra:[8,21,29,32,39,44],extract:[0,3,4,7,9,24,31,32,35,39,41,44],extract_semantic_feat:[26,39],extractor:[0,3,26,31,35,39,44],f1:[7,8,41],f:[6,11,23,35,44],f_name:36,fabio:[4,39],facilit:[2,39],fact:39,factor:[24,34,44],factori:[33,41],fail:[8,33,44],fair:[13,32],fals:[7,9,13,25,27,28,29,31,32,33,34,36,38,40,41],famili:[31,39],far:7,fassion:41,featextractnnbuilderchainnodegett:31,featur:[0,3,4,6,9,15,22,24,26,31,32,35,39,40,41,44],fed:[4,15,39],feed:[22,35],feistner:44,felix:1,fetch:[26,32,35,36],fetch_img_path:36,fetch_net:32,few:7,fiel:39,field:33,figur:[0,1,2,6,7],file:[2,13,17,20,21,24,26,31,33,34,36,38,39,40,41],filenam:[36,39,41],fill:[11,41],find:[0,1,2,6,8,22,24,29],fine:41,finish:41,first:[0,1,4,6,8,17,29,32,39,40,41],fisher:1,fishr:[29,42],fit:2,fix:[11,40,41],flag:34,flag_accept:29,flag_do_not_weave_til:36,flag_erm:29,flag_pretrain:[31,32,35],flag_rand_color:36,flag_update_batch:29,flag_update_epoch:29,flat:31,flatten:[32,39],flip:[36,44],florian:[2,39],flow:[29,40,44],flowgenimg:41,flows_gen_img_model:[16,25,43],fly:22,fmcarlucci:36,fname:36,folder:[13,21,36,40,41],folder_a:22,folder_b:22,folder_c:22,folder_na:41,folder_nam:41,foldernam:41,follow:[0,1,2,4,6,7,8,9,11,13,14,15,20,22,32,44],foral:3,forc:1,foreground:36,forg:8,form:[3,6,8,36,41,44],format:[8,25,36,41],former:32,formula:32,formular:11,forward:[23,32,35,39,40],found:[7,8,9,13,22,41],fourth:8,frac:[1,2,6],frame:41,framework:[2,20,23,39],frederik:1,frequenc:24,from:[0,2,4,5,6,7,9,11,13,15,20,21,23,24,25,26,27,29,30,32,33,34,35,37,38,39,40,41,44],fstring:36,fulfil:[0,41],full:[7,41],fulli:[8,15,31,32],fun:41,fun_class_local_na2vec_new:40,fun_gen:41,fun_img_path_loader_default:36,fun_label2onehot:36,fun_tensor_norm:30,further:[1,44],furthermor:[4,9,15],g:[1,3,8,9,24,31,32,33,39,41,44],g_list_model_penalized_reg_agg:25,g_method_na:41,g_model_na:41,gamma:[24,41],gamma_:[0,6],gamma_d:[0,7,11,15,22,24,39,44],gamma_i:[0,2,6,7,11,15,22,24,39,44],gamma_reg:[0,4,6,11,24],ganin:[14,39],gatedconv2d:32,gateddens:32,gaussian:[0,29],gaussian_kernel:29,gen:[15,21,24,41],gen_adversari:29,gen_benchmark_plot:41,gen_bencmark_plot:41,gen_fun_acc:41,gen_img_load:41,gen_img_xyd:41,gen_plot:[8,24,41],gener:[4,5,7,20,21,22,24,28,29,34,36,38,39,40,41,42],generate_benchmark_plot:[16,25,43],georg:44,geq:9,get:[3,8,22,25,26,27,29,31,35,39,40,41],get_background_color:36,get_base_domain_size4match_dg:30,get_devic:41,get_dset_by_domain:40,get_flat_dim:31,get_foreground_color:36,get_gamma_reg:41,get_git_tag:[16,25,43],get_hyperparamet:41,get_label_na:41,get_list_domain:40,get_list_domains_tr_t:40,get_list_pred_target:41,get_list_tuple_img_label:36,get_logg:41,get_model:29,get_na:40,get_node_na:20,get_num_color:36,get_num_row:30,get_shared_sampl:41,get_task:[22,40],get_til:36,get_train:26,get_val:41,getitem:40,getter:40,git:45,github:[13,35,36,39,44,45],give:[7,22,24],given:[4,6,7,8,24,28,32,35,36,40,41],global:25,go:[0,41],goal:[0,1,4,6,29],goe:29,good:8,gossmann:44,gpu:[39,41],grad:[3,29,32],grad_output:32,gradient:[3,23,29,32,42],gradual:29,graphic:[2,6,8],grep:8,grid:[4,9,11,24,39,41],grid_df:41,grid_df_prior:41,grid_img:41,grid_len:[4,24,36],grid_task:41,gridpoint:41,gridsearch:[11,41],gruber:44,guid:[20,44],h:[1,6,40],h_a:1,h_b:1,h_d:2,h_e:1,h_t:2,ha:[8,13,22,29,30,32,33,34,36,39,41,44],half:29,hand:1,handl:[23,33,36,44],handler:33,happen:[11,29],har:0,hard:[3,7,21,22],hardcod:[26,31,34,40],has_file_allowed_extens:36,hashtag1_seed_1:21,hashtag1_seed_1_instance_wise_predict:21,hat:44,have:[0,6,7,8,9,11,17,22,23,30,32,39,40,41,44],hduva:[24,26,39,42],head:8,header:38,heavi:[31,33,34],height:[22,32,40],help:[8,24,25,44],henc:[6,11],hennig:1,here:[7,8,10,11,20,22,23,24,34,36,39,44],hessian:1,hidden:[2,32,35],hidden_dim:35,hierarch:[39,41,42],high:[6,7],highest:6,highli:1,hoc:29,hook:32,hot:[2,39,40,41],how:[0,7,11,24,27,33,44],howev:[8,15],html:[32,36,37],http:[0,1,2,6,12,13,15,18,32,35,36,37,39,44,45],huan:[5,29],human:22,hund:22,huski:22,hyper:[2,24,29,35,39,41,42],hyper_prior:35,hyper_prior_dim:35,hyper_schedul:[25,26],hyperparam:41,hyperparamet:[0,3,4,6,7,15,41,42,44],hyperparameter_gridsearch:[16,25,43],hyperparameter_retriev:[25,43],hyperparameter_sampl:[16,25,43],hyperparamt:[11,41],hyperschedul:29,hyperschedulerwarmupexponenti:29,hyperschedulerwarmuplinear:29,i:[1,3,4,6,8,11,29,35,36,39],i_c:[30,31,33,35,40],i_ch:41,i_channel:[31,32,35],i_h:[30,31,33,35,40,41],i_w:[30,31,33,35,40],iaf:35,iclr:2,id:40,idea:[3,8],ident:[21,32,38,40],idx:40,ieee:[4,39],if_stop:27,ignor:[13,25,32,36],ils:39,im:41,imag:[0,1,2,4,6,8,21,22,24,29,31,32,35,36,39,40,41,42],img:[29,35,36,41],img_class_loss:39,img_h:29,img_h_dim:35,img_loader2dir:40,img_natur:29,img_trans_t:[22,40],img_trans_tr:40,img_w:29,implement:[7,26,33,34,35,36,40,41,42,44],implicitli:1,import_path:41,importantli:[1,34],improv:[5,15,29],imsiz:[22,40],in_channel:7,in_featur:32,includ:[6,7,9,11,23,24,36],incomplet:39,inconsist:[29,42],incorrect:9,ind:[36,40],ind_batch:[23,29],ind_color:36,ind_til:36,independ:[15,20,35,39],index:[3,24,29,36,39,40,41,42,44,46],indic:[11,22],individu:[24,41],infer:[2,15,22,32,40],infer_y_vpicn:39,infer_zy_loc:35,influenc:6,info:41,inform:[1,2,6,7,8,11,15,22,27,29,39,40,41],inherit:[20,23,29,33,39,41],init:[33,39],init_busi:[20,26,29,31,33,34,40],init_next_model:26,init_p_zx4batch:39,init_param:41,initi:[23,26,29,31,33,34,39,41],initialis:[6,11],inject:0,inlin:22,inner:[3,29,39],innermost:27,inplac:32,inproceed:2,input:[0,2,15,29,30,32,33,34,39,40,41,44],input_channel:32,input_dict:41,input_dim:34,input_dir:38,input_fil:38,input_flat_s:32,input_s:32,insid:[8,9,21,29,44],instanc:[4,21,24,28,29,32,39,41],instanti:[39,41],instead:[2,6,28,32,33,40,41],integ:[11,41],integr:[31,34],intellig:19,intend:32,intens:[34,35],inter:44,interest:29,interfac:[28,33,35,39,40,41],interface_class:41,interface_vae_xyd:[16,25,43],interfacevaexyd:39,intern:[0,1,29],interpret:[3,8],intersect:3,interv:[11,24],interven:27,intrepret:41,introduc:[0,2,6,23,39],intuit:1,invari:[24,29,39,42],investig:8,invit:28,invok:33,io:44,irl:42,irm:42,irmv1:[29,42],is_dict_with_kei:41,is_myjob:[26,29,31,33,34,40],is_valid_fil:36,isinst:40,isiz:[22,32,35,40],isol:[33,41],item:[29,36,39],iter:[11,40],its:[1,2,6,7,8,11,24,29,33,39],itself:41,j:[3,6,29,35],jigen:[26,29,36,39,42,44],jigen_ppath:24,jigendg:36,jigsaw:[4,39],jigsawload:36,job:[31,33,34,38],join:[22,38],jointli:3,journal:[14,39,44],jpeg:40,jpg:[22,40],json:8,just:[29,32],jvp:32,k:[6,40],karg:[33,34],kde:41,keep:[8,39],keep_model:[0,15,21,24],kei:[6,9,20,22,23,25,29,30,39,40,41],kernel:29,kernel_s:[7,32],kind:[39,41,44],kl:2,know:40,knowledg:[22,40],kth:30,kunstner:1,kwarg:[29,32,33,34,39],l29:35,l2:3,l93:36,l:[1,6,8],l_2:6,label:[0,2,4,6,15,22,29,34,35,39,40,41],label_transform:36,lambda:[1,3],landscap:[1,29],larg:15,last:[8,26,27,32,35,41],last_feat_dim:32,latent:[2,15,24,35,39],latter:32,launch:[8,44],layei:5,layer1:29,layer2:29,layer:[2,3,12,26,29,31,32,35,42],layerid:32,lead:[6,11,15,44],learn:[1,2,6,7,9,11,14,15,23,24,29,39,42,44],learnabl:15,least:2,leav:42,left:[0,6,46],len:36,length:[4,24,39],leq:1,let:[1,3,29],level:[1,2,3,6,11,24,27,29,35,39,41],li:19,light:[31,33,34],like:[7,8,11,23,26,36,39,41,44],likelihood:[2,35],limit:44,limits_:6,line:[8,20,21,22,24,25,31,33,37,40,41],linear:32,linearli:32,link:10,linkifi:45,linux:8,lisa:44,list:[9,11,22,25,29,34,36,38,39,40,41,46],list_b_reg_loss:29,list_class_dir:36,list_class_na:40,list_d_tr:39,list_dict:29,list_dict_var_paragrad:29,list_divid:29,list_domain:40,list_domain_id:40,list_domain_na:40,list_domain_tr:40,list_inner_product:39,list_loss:39,list_multipli:39,list_penalized_reg:25,list_str_d:39,list_str_i:[22,39,40],list_str_na:41,list_tr_domain_s:[29,30],list_transform:36,list_tuple_path_cls_ind:36,list_val:29,load:[28,29,39,40,41],loader:[24,28,29,36,39,40,41],loader_t:[39,40,41],loader_tr:[23,40],loader_v:40,loaderdomainlabel:40,local:[1,8,29,36,40],locat:[22,35],log:[2,6,8,23,24,29,41],log_loss:29,logger:[16,24,25,43],logger_18603:41,logger_nam:41,logic:23,login:8,logit2preds_vp:41,logit:[39,41],loglevel:[24,41],lognorm:[9,41],lognormal_grid:41,loguniform:[9,41],loguniform_grid:41,look:[6,11],loss:[1,3,4,6,20,24,27,29,39,42,44],loss_task:29,lot:7,low:0,lower:[9,42],lowercas:36,lr:[0,7,9,11,24],lscondpriorlinearbnrelulinearsoftplu:35,lsencoderconvbnrelupool:35,lsencoderlinear:35,ly:9,m2r2:45,m:[4,6,39,44],ma:[5,29],ma_trainer2_trainer3:29,machin:[0,1,14,29,39],macro:41,magnifi:24,magnitud:11,mai:[6,8,11],main:[0,29],main_out:[0,1,2,3,4,5,6,7,8,12,13,15,18,19,22,24,44],make:[0,8,10,23,29,33,40,41,44],make_basic_train:26,mamba:8,mandatori:[8,9],mani:[0,11,32],manual:[8,37],map:[2,22,39,40,41,44],margin:15,markdown:8,marrlab:[13,39,44],master:[13,35,36,39],match:[24,29,30,40,42],match_feat_fun_na:39,match_tensor_reshap:29,matchdg:[29,30,42],matchdg_arg:[26,29],matchdg_match:[26,29],matchdg_util:[26,29],matchdictinit:30,matchdictnumdomain2sizedomain:30,matchdictvirtualrefdset2eachdomain:30,matchpair:30,mathbb:3,mathcal:[0,1,3,6],matplotlib:41,matrix:[1,6,24],matter:[22,40],matthieu:[1,29],max:[1,9,11,41],max_0_x:41,max_:27,max_pool_strid:32,maxim:[0,7,11,39,44],maximilian:39,maximum:[24,27,29,39,41],maxpool2d:[7,32],maxpool_2d:32,md:[7,8],mdict:29,mean:[2,9,11,21,22,29,32,35,41],measur:[1,24],mechan:42,medic:[15,39],member:[29,33],memori:[24,36],mensch:22,mention:[11,20],merg:36,messag:[8,33],meta:[29,42],metavar:25,method2overrid:41,method:[1,4,7,8,11,23,26,29,33,34,39,40,41],metric4msel:39,metric:[27,28,29,39,41],metric_t:[28,39],might:[8,11,21,22,31,33],min:[3,6,9,11,41],min_:3,mini:[3,29,39],mini_vlc:[1,2,4,6,13,15,19,22],minibatch:[29,39],minim:[1,6,8,22,24,29,39,42],minimum:24,misc:44,miss:[4,15],mix:[24,36,40],mixed_codec:16,mk_conv_bn_relu_pool:32,mk_dann:39,mk_diva:39,mk_dummy_label_list_str:41,mk_erm:39,mk_exp:[16,42,43],mk_fun_label2onehot:36,mk_fun_sav_img:41,mk_hduva:39,mk_img:41,mk_jigen:39,mk_loader:40,mk_match_tensor:29,mk_node_task_path_list:40,mk_onehot:40,mk_opt:29,mk_parser_main:25,mk_rand_label_onehot:41,mk_rand_xyd:41,mk_task_dset:40,mk_task_fold:[22,40],mknodefeatextractnnbuilderexternfromfil:31,mknodefeatextractnnbuildernamealex:31,mknodefeatextractnnbuildernameconvbnrelu2:31,mldg:[9,11,42,44],mldg_dial:44,mmd:29,mmd_base:[25,26],mnist:[36,40,42],mnist_benchmark:11,mnistcolor10:[0,2,3,5,6,7,11,12,15,18,22,24],modal:2,mode:[9,11,22,23,24,29,32],model:[1,3,5,6,7,9,11,12,16,18,19,21,22,25,26,27,28,29,31,32,34,35,36,40,41,42,43,44],model_custom:[16,25,43],model_dann:[16,25,43],model_diva:[16,25,43],model_erm:[16,25,43],model_hduva:[16,25,43],model_jigen:[16,25,43],model_ld:28,model_loc:41,model_sel:28,model_selection_epoch:27,model_vae_xyd_classif:[16,25,43],modeldan:39,modeldiva:39,modelerm:39,modelhduva:39,modeljigen:39,modifi:39,modul:[20,42,43],more:[0,1,2,3,7,8,9,11,22,24,29,31,39],moreov:8,most:[1,22,29,34],mostli:40,motiv:42,move:[29,39,42],move_averag:29,mpath:41,msel:[16,25,26],mseloraclevisitor:27,mseltrloss:27,mselvalperf:27,mu:44,mu_a:44,mu_b:44,mu_c:44,mu_d:44,muhammad:44,multipl:[9,11,24,33,44],multipli:[3,15,24,39,44],multiplier4task_loss:39,multiprocess:8,must:[8,9,11,20,32,39,40,41],my_cdist:29,my_custom_arg_nam:22,mycustomtrain:23,myst:45,n:[0,4,6,8,29,39],n_d:6,n_domain:36,na:22,na_adaptor:34,na_cf:34,na_domain:40,na_prefix:26,na_t:40,na_tr:40,nabla_:[3,29],naiv:29,name:[8,9,11,17,20,22,24,26,29,31,33,34,36,39,40,41,44],narg:25,nat:0,natur:0,navig:46,nbsphinx:45,nbsphinx_link:45,nearest:[6,41],necesarilli:30,necessari:[20,23,24,39],necessarili:29,need:[2,6,7,11,15,20,22,23,29,32,34,39,40,41],needs_input_grad:32,neg:35,neglect:39,neighbor:6,neq:6,net:[6,11,13,15,22,26,39,42,44],net_adversari:[25,31],net_class_d:[33,35],net_class_i:[33,35],net_classif:[25,31],net_classifi:39,net_classifier_class:39,net_classifier_permut:39,net_conv:35,net_conv_conv_bn_pool_2:[25,31],net_discrimin:39,net_encod:[26,39],net_fc_z2flat_img:35,net_gat:[25,31],net_infer_zd:35,net_infer_zd_top:35,net_infer_zi:35,net_infer_zx:35,net_invar_feat:39,net_p_x_log_var:35,net_p_x_mean:35,net_torchvis:32,net_x:[33,35],netconvbnrelupool2l:32,netconvdens:32,nettorchvisionbas:32,network:[1,7,9,11,12,24,26,29,31,35,39,41,42],neural:[9,12,24,26,29,31,35,39,41,42],neuron:[6,7],next:[9,11,26],next_model:26,nicknam:22,nine:7,nllpixellogistic256:35,nn:[7,25,30,31,35,39],nn_alex:[25,31],nn_torchvis:[25,31],nn_zoo:[16,25,31],nname:[0,1,2,3,4,5,6,7,9,11,12,13,15,18,19,22,24,31],nname_argna2v:[22,24],nname_dom:[0,7,11,15,22,24],nname_encoder_sandwich_x2h4zd:[2,6,11,24],nname_encoder_x2topic_h:[2,6,11,24],no_dump:24,nocu:[1,24],node:[6,26,29,31,33,34,40,41],nodealgobuild:[20,26],nodealgobuilderapimodel:26,nodealgobuilderdann:26,nodealgobuilderdiva:26,nodealgobuildererm:26,nodealgobuilderhduva:26,nodealgobuilderjigen:26,nodetaskdg:40,nodetaskdgclassif:40,nodetaskdict:40,nodetaskdictclassif:40,nodetaskfold:40,nodetaskfolderclassnamismatch:40,nodetaskmnistcolor10:40,nodetaskpathlistdummi:40,nodevaebuilderarg:34,nodevaebuilderimgalex:34,nodevaebuilderimgconvbnpool:34,nodevaebuilderimgtop:34,nodevaebuilderus:34,nohup:8,nois:0,nomin:[30,36],non:32,none:[20,22,25,26,27,29,30,31,32,33,34,35,36,39,40,41],nonlinear:[12,29],norm:[3,29,32],normal:[9,22,35,36,41],normal_grid:41,notabl:15,note:[2,7,8,9,26,32,36,37,39,41,44],novel:44,npath:[2,6,9,11,13,15,22,24,31,44],npath_:11,npath_argna2v:[22,24],npath_dom:[15,22,24,44],npath_encoder_sandwich_x2h4zd:[2,24],npath_encoder_x2topic_h:[2,24],nperm:[4,11,24,39],nrow:41,nsum_:29,nsum_i:29,num:[9,11,41],num_batch:29,num_class:41,num_domains_tr:30,num_featur:7,num_of_input_channel:7,num_param_sampl:[9,11],num_perms2classifi:36,num_sampl:41,num_shared_param_sampl:[9,11],num_top:35,num_work:40,number:[2,4,6,7,8,9,11,24,29,32,39,40,41,44],numer:[40,41],nutan:44,o:6,ob:6,obj1:41,obj2:41,obj:[33,41],object:[1,6,26,27,28,29,30,31,33,34,35,39,40,41,44],observ:[2,6,16,20,25,26,27,29,39,41],observer4msel:27,obtain:15,obvisitor:28,obvisitorcleanup:28,obvisitorgen:28,off:3,offer:[8,9,22,44],old:34,omega:6,onc:33,one:[0,1,2,3,4,6,8,10,11,20,22,29,30,32,33,34,35,36,39,40,41,42,44],onehot:36,ones:[7,41],onli:[1,4,6,7,8,9,11,13,15,20,21,24,29,33,34,35,36,39,40,41],openai:35,oper:[32,33,39,40],optim:[0,1,3,29,32,33],option:[8,9,11,23,39,40,41],option_str:25,optmiz:3,oracl:[27,28],oracle_last_setpoint_sel_te_acc:27,order:[1,15,20,22,34,35,40],org:[0,1,2,6,12,15,18,32,36,37,44],organ:40,origin:[4,29,35,39],orthogon:3,os:22,other:[1,2,4,6,7,8,9,11,20,23,24,29,32,33,36,39,40,41,44],othervis:11,otherwis:[6,29,40,41],our:[2,7,8,22,41,44],out:[8,22,24,27,29,32,40,42],out_channel:7,out_featur:32,out_hidden_s:32,outer:29,outlin:20,outp_dir:[8,24],output:[0,9,11,24,29,31,32,38,39,44],output_channel:32,output_dim:34,output_dir:[8,9,11,41],output_fil:38,output_folder_na:41,output_s:32,outsid:9,over:[6,11,41],overal:[15,39,42],overrid:[23,29,31,33,34,41],overridden:32,override_interfac:[16,25,43],overview:[23,42],own:[11,22],p1:[8,41],p2:[8,41],p4:9,p:[32,34,35,39],p_:2,p_chain_handl:[25,26,29,31,34,40],p_dropout:32,p_na_prefix:[29,39],pac:[11,44],pack:1,packag:[0,1,7,8,21,22,42,43],pacs_mini_10:16,pacs_split:16,pad:32,page:[42,44],pair:[6,7,40],pairwis:[29,30],palett:[36,40],panda:[0,41],paper:[0,2,5,6,15,29,36],par:29,par_setpoint:29,parallel:[4,35],param1:8,param:[8,9,11,26,29,31,32,33,34,35,36,40,41],param_config:41,param_grid:41,param_idx:[24,41],param_index:[8,41],param_nam:41,paramet:[1,2,6,8,9,11,20,22,26,29,31,32,34,35,36,38,39,40,41,42],parameter:15,parent:[26,29,33,35,39,40],parent_class:39,parmaet:29,pars:[20,25],parse_cmd_arg:25,parse_domain_id:40,parser:[29,30,39,45],parsevaluesorkeyvaluepair:25,part:[0,2,29,41],partial:[24,38],particular:[7,29],pass:[11,29,32,41],patch:[4,39],patches_permutation4jigsaw:16,path2filelist:36,path:[2,8,9,11,13,22,24,31,32,35,36,40,41,44],path_dir:36,patrick:44,patter:34,pattern:[4,26,27,28,29,31,34,35,39,44],pcr:[16,25,26,29,31,34,40],pd:41,pdf:[0,2,6,12,15,18,44],penalti:[3,44],per:[11,29],perf:[16,25,43],perf_metr:[16,25,43],perfclassif:41,perfmetricclassif:41,perform:[5,6,7,8,15,23,24,27,28,29,32,39,41,44],perm_task_loss:39,permuat:36,permut:[4,24,29,36,39],persist:[21,28,39,41],perturb:[0,1],phase:6,phi:[0,3,6,29,44],phi_:[2,6],phi_d:2,phi_i:2,phi_x:2,philipp:1,pick:11,pil:8,pip:[8,17,44],pixel:[0,35],plai:34,plain:22,platform:44,pleas:[2,8,9],plot:[7,24,36,38,41,44],plot_d:36,plot_ds_list:36,plu:4,pmlr:[1,15,29,39],png:[8,40],poetri:17,point:[6,7,9,11,41],pool:[35,39],posit:[1,6],possibl:[1,11,20,31,41,44],post_tr:29,posterior:2,potenti:[2,21],power:[11,29,44],ppath:36,pperm:[4,11,24],practic:3,pre:[33,44],preced:11,precis:[7,8,41],pred2fil:39,predefin:[11,24],predfin:24,predict:[0,4,6,15,21,28,35,39,40,41],prefix:[24,29,39,41],prepar:[8,15,22,23],prepare_ziped_load:29,preprint:39,preprocessinga:44,preserv:24,prevent:9,previou:11,primari:20,print:[22,23,29,32,39,40],print_diff:41,print_opt:33,print_paramet:[29,39],prior:[2,34,35],prob_no_perm:36,prob_permut:39,probabilist:2,probabl:[0,4,39,41],probe:29,problem:[4,15],problemat:41,procedur:[0,6,11],proceed:[4,19,39],process:[0,6,8,11,23],product:[3,29,30,39,41],program:7,project:44,propag:23,properti:[26,27,28,29,36,39,40,41],proport:24,protocol:44,provid:[6,7,8,22,25,36,40,41,44],pulp:8,pure:0,puzzl:[4,39],py:[0,1,2,3,4,5,6,7,8,11,12,13,15,16,17,18,19,20,22,23,24,35,36,39,44,45,46],pypi:44,pyplot:41,pyproject:17,python3:24,python:[0,1,2,3,4,5,6,7,8,11,12,13,15,17,18,19,20,24,26,31,40,41],pytorch:[8,20,21,22,26,32,35,36,37,39,40],q:2,q_:2,q_zd:35,q_zx:35,q_zy:35,quad:[3,6],quantifi:42,question:29,quieri:39,r:[1,3,8,17,29,32,44],r_a:44,r_b:44,r_c:44,r_d:44,radar:8,radar_dist:8,radar_plot:41,rahul:44,rais:41,rais_error_if_num_not_specifi:41,rame:1,ramp:29,rand:11,random:[0,8,9,11,24,29,36,37,40,44],randomgrayscal:22,randomhorizontalflip:22,randomli:[6,8,9,11,29],randomresizedcrop:22,rang:[11,36],rao:44,rate:[0,7,9,11,23,24],ratio:[4,40],raw_split:36,re:[35,36,40],read:8,read_prediction_fil:39,reader:44,real:[9,29,31,33,34],recal:[7,8,41],recip:32,recognit:[4,39],recom:8,recommend:44,recon:34,recon_cf:34,recon_ydx:34,reconstruct:[15,34,39],reconvaexyd:34,recours:41,recov:39,rectangl:44,recurs:[29,41],red:7,reduc:29,reduce_dim:35,redund:[33,34],ref_domain:29,refactor:34,refer:[1,4,8,9,11,15,41],referenc:[9,41],referenced_param:41,referencehyperparamet:41,reg:[0,6,24,41],regardless:3,regist:[23,31,32],register_external_nod:26,regress:41,regular:[1,6,20,24,29,39,41],reimplement:35,reject:9,rel:[4,6,10,21,24,40],relat:[2,22],reli:8,relu:[7,32,35],remov:41,remove_last_lay:[13,31,41],replac:[32,41],repositori:[8,10,13,17,20,22,24,44],repres:[1,2,9,28,29,32,39,41,44],represent:[0,2,3,6,35,39,40],reproduc:[8,9,24,37],request4chain:33,request:[25,26,29,31,34,39,40,41],requestargs2expcmd:33,requesttask:33,requestvaebuilderchw:33,requestvaebuildernn:33,requir:[2,4,10,17,25,32,40],research:[5,14,29,39],reset:[23,26,27,28,29,39],reset_aux_net:[26,39],reset_feature_extractor:39,reshap:39,residu:[2,15,39],resiz:[22,36],resnet50:11,resnet50domainb:11,resnet:[2,6,13,15,22,44],respec:2,respect:[6,8,9,10,15,22,23,39,41],respons:[26,28,29,31,33,34,40],restrain:24,restult:41,result:[3,4,9,11,21,24,33,38,41,44],resurs:41,retrain:34,retriev:[20,32,41],reus:[31,34,44],revers:[7,42],rgba:36,right:[0,1,6,23,39],rightarrow:6,rinohtyp:45,risk:[1,22,29,39,42,44],rm:8,robust:0,robustml:2,rockenschaub:44,role:34,root:[20,22,36,40],root_img:36,round:[11,41],round_to_discreate_grid_norm:41,round_to_discreate_grid_uniform:41,round_vals_in_dict:41,routin:23,row:[9,36,41],rst:46,rtfd:45,rule_result:8,run:[11,20,21,24,32,41,44],run_benchmark_slurm:[8,44],run_benchmark_standalon:8,run_experi:[8,16,25,43],s:[2,22,23,24,35,36,39,40,44],safest:33,safeti:33,salesforc:[5,29],same:[2,6,7,8,11,20,21,22,30,31,32,39,40,44],sampl:[6,15,40,41],sample_df:41,sample_grid:41,sample_gridsearch:41,sample_hyperparamet:41,sample_num:41,sample_p_zd:34,sample_p_zi:34,sample_paramet:41,sample_sav:40,sample_task:41,sample_task_only_shar:41,sampledhyperparamet:41,sampling_se:[9,11,41],san_check:[4,7,9,22,24],san_num:[7,22,24],sandwich:[35,42],saniti:[24,39,41],sanity_check:[16,25,43],sanitycheck:41,sanniti:42,sapien:22,satisfi:41,sav_add_titl:41,save:[7,8,11,24,27,32,39,40,41],save_for_backward:32,save_for_forward:32,save_san_check_for_domain:41,saved_model:21,saver:41,scalar:[29,41],scalar_zx2fil:34,scale:[22,32,35,41],scatterpl:8,scatterplot:41,scatterplot_matrix:41,scenario:[34,44],schedul:[23,29],scheme:26,schwarz:44,scm:6,score:1,scratch:[6,20],script:[8,16,24],search:[9,11,42],second:[1,6,7,8,29,40],section:[3,9,11,29],see:[0,1,3,8,9,10,13,20,22,24,29,36,39,44],seed:[7,8,11,24,37,41,44],seen:7,segment:39,sel_model_te_acc:27,select:[5,6,8,11,24,26,27,28,29,31,34,39,40],selector:[27,28],self:[20,23,29,32,33,40,41],semant:[7,39],semi:15,sensit:8,separ:[22,34,35],sepcif:8,seq2seq:39,sequenti:[7,32],serv:[2,3,33,39],servic:[31,33,34],session:23,set:[1,2,4,6,7,9,11,15,23,24,25,26,29,30,37,39,40,41,42,44],set_list_domain:40,set_par:33,set_sav:39,set_schedul:29,set_se:37,set_step:29,set_task:28,setpoint:27,setter:40,setup:[15,17,41],sever:[21,23,29,34,36,44],sh:[8,44],sh_list_error:8,shade:6,shall:[6,11,22,41],shape:[30,39],share:[9,30,34,39,41,44],shared_conf_samp:41,shared_config:41,shared_config_ful:41,shared_df:41,shared_sampl:41,shared_samples_ful:41,should:[7,9,22,27,28,29,31,32,33,35,37,39,41,44],show:[7,20,32,33],shown:[2,7,20,22],shrestha:44,shuffl:[24,36,39,40,42],shuffling_off:24,side:1,sigma:0,signatur:[13,22],signific:41,silent:[25,32],sim:[0,3,6],similar:[6,8,21,22,24,39],simpl:[23,29,42],simpli:[3,44],simplic:7,sinc:[6,13,15,32,33,39,40,41],singl:[15,24,29,36,38,42,44],sit:22,site:44,size:[0,2,6,7,9,11,15,24,29,35,39,40,44],size_domain_k:30,sketch:[4,22],skip:[11,41],skip_gen:41,skip_plot:38,slot:40,slurm_log:8,small:[0,1,6,29],smk:16,snakemak:[8,38],so:[3,6,8,20,21,22,29,31,32,33,34,40,42,44],sofa:22,softmax:39,softplu:35,softwar:[41,44],solo:36,solut:33,solv:[3,4,39],some:[8,11,15,22,24,29,31,33,34],someth:[8,21,22],sort:11,sould:11,sourc:[0,1,2,6,18,20,22,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,44],sp_matrix:8,sp_matrix_dist:8,sp_matrix_dist_reg:8,sp_matrix_reg:8,space:[1,2,6,15,24,39,41,44],span:11,spatial:4,special:[0,23],specif:[2,7,8,9,15,39,40,41],specifi:[2,4,7,8,9,11,15,20,21,24,26,31,39,40,41,44],specificli:8,spectrum:36,specul:40,speed:41,sphinx_materi:45,split:[4,21,24,26,29,40],split_net_feat_last:26,spliter:39,squar:[3,29,36],ssystemt:41,stabl:[32,36,37],standalon:11,standard:[0,9,11,36,39],start:[0,6,8,11,22,23,24,29,31,34,41],startse:[9,11],state:[23,29,44],statement:40,std:[9,11,41],steadi:29,stem:39,step:[0,9,11,24,29,41],still:[0,8,33],stochast:[8,41],stochastic_vari:8,stop:[7,9,11,24,27,28],store:[15,22,24,29,32,41],store_arg:41,str:[36,38,40,41],str_metric4msel:[28,29],str_trainer:29,straight:40,straightforward:22,strategi:29,stride:[7,32],string:[26,28,29,31,33,34,36,39,40,41],structur:[2,6,7,22,24,25,29,36,39,40,41,42,44],stuhl:22,sub:[22,33,40],subclass:[31,32,33,34,39],subdirectori:8,subdomain:36,subfold:[22,36],subfolder_na:[40,41],subgrid:11,sublimin:6,submiss:8,submit:8,submodul:[42,43],subpackag:[42,43],subsampl:[24,41],subset:3,subset_step:36,subspac:39,substack:6,succ:40,success_nod:[26,33,34],successfulli:33,successor_nod:[29,31,34],suffic:1,suffici:[7,44],suffix:39,sum:[6,27,29],sum_:[1,3,6],sum_i:29,summar:7,sun2021hierarch:2,sun2024domainlab:44,sun:[2,39,44],supervis:[2,15,39],support:[8,9,15],suppos:[8,17,22,34],suppress:24,sure:[8,23,33],symbol:10,system:8,t0:7,t:[1,6,22,27,29,32,34,36,37],tab:46,tabl:7,tackl:0,tag:24,tailor:23,take:[11,32,34,35],taken:11,target:[0,29,39,40],target_dim:[32,41],target_transform:36,task1:9,task1_algo1_git:21,task1_test_domain1:21,task1_test_domain1_tagnam:21,task2_test_domain3_tagnam:21,task:[0,1,2,3,4,5,6,7,9,12,13,15,16,18,19,20,21,25,26,28,29,30,33,36,39,41,43],task_config:41,task_custom:40,task_dann:11,task_demo:22,task_diva_di:11,task_dset:[16,25,43],task_dset_custom:22,task_fold:[16,25,43],task_folder_mk:[16,25,43],task_jigen:11,task_loss:39,task_mini_vlc:[16,25,43],task_mnist_color:[16,25,43],task_nam:[11,40,41],task_pacs_path_list:11,task_pathlist:[16,25,43],task_sampl:40,task_util:[16,25,43],task_vlc:[22,24,44],taskchainnodegett:40,taskclassif:40,taskna:[22,40],taskpathlist:40,tau:[0,6,11,24,30],te_d:[0,1,2,3,4,5,6,7,8,9,12,13,15,18,19,22,24,41,44],te_id:40,techniqu:[1,9],tell:44,tensor1hot2ind:40,tensor:[6,24,25,29,30,32,35,39,41],tensor_batch_x:30,tensor_d:[20,23,29,39],tensor_i:[20,23,29,39],tensor_ind:41,tensor_label:40,tensor_x:[20,23,29,32,39],term:[1,2,4,6,39],termin:7,test:[7,8,9,11,21,24,27,28,39,40,41,42,44],test_domain:11,test_exp:33,test_img:[16,25,43],test_mk_exp:34,test_mk_exp_dann:39,test_mk_exp_diva:39,test_mk_exp_erm:39,test_mk_exp_jigen:39,text:[0,1,6,21,22],textnorm:1,textur:35,tf_util:35,than:[39,40],thei:[11,23,32,41,44],theire:11,them:[0,32,34,41],theori:42,thep:11,therefor:[0,6,11,15],theta:[1,2,29],theta_:2,theta_d:2,theta_i:2,theta_x:2,thi:[0,1,2,3,4,6,7,8,9,11,13,15,17,20,21,22,23,24,26,29,32,33,34,35,39,40,41,44],third:8,though:[11,32],three:[0,2,6,15,39],threshold:[0,27],thress:22,through:[2,11,24,35,41,42],thu:[3,7,9,39,41,44],tild:0,tile:[24,36,39,42],time:[4,6,11,22,34,40,41],titl:[2,41,44],tmux:8,togeth:[6,7,39,40],toi:24,toml:17,tool:[8,11],top:11,topic:[24,32,35,39,42],topic_dim:[2,24,34,35,39],torch:[30,32,35,36,37,39,40],torchvis:[22,31,32,36,40,41],total:[4,11,24],total_loss:39,total_step:29,totensor:22,tpath:[4,9,11,22,24,44],tr_batch:[23,29],tr_d:[0,2,5,6,7,9,11,12,15,18,22,24],tr_epoch:[23,29],tr_id:40,tr_observ:27,track:24,trade:3,train:[4,7,9,11,14,21,23,24,26,27,28,29,32,34,35,36,39,40,41,42],train_bas:[25,26],train_causirl:[25,26],train_cor:[25,26],train_dial:[25,26],train_ema:[25,26],train_fishr:[25,26],train_hyper_schedul:[25,26],train_irm:[25,26],train_matchdg:[25,26],train_mldg:[25,26],trainer:[5,9,11,12,16,18,20,23,24,25,26,27,28,39,40,42,44],trainerbas:29,trainercausalirl:29,trainerchainnodegett:29,trainercor:29,trainerdi:29,trainerfishr:29,trainerhyperschedul:29,trainerirm:29,trainerma:29,trainermatchdg:29,trainermldg:29,trainermmdbas:29,trans_img:36,trans_t:22,trans_target:36,trans_tr:22,transform:[7,12,22,36,40,44],translat:8,transpar:35,tree:8,treshold:27,tri:[25,41],trim:29,trust:41,tupel:29,tupl:[32,36,40,41],tuple_data_domains_batch:29,turn:15,tutori:44,two:[1,2,3,4,6,8,9,11,15,20,29,39],txt:[8,17,21],type:[25,32,39,40],typedef:40,typic:23,u_import:[16,25,43],u_import_net_modul:[16,25,43],umer:44,unbias:[3,29],undefin:9,under:[1,3,20,41],underli:[6,7],underset:6,understand:[4,24],uniform:[9,41],uniform_grid:41,uniqu:[11,22,40,44],unit:24,unknown:[6,25],unobserv:[0,2,24],unseen:44,unstabl:8,unsupervis:[15,39,42],until:32,up:[2,11,23,24,28,29,41],updat:[6,8,23,24,26,27,28,29],upgrd:8,upon:[21,23,29,35,36],upper:9,us:[0,1,3,4,7,8,9,11,15,17,20,23,24,25,27,28,29,32,33,34,35,36,38,39,40,41,42,44],usa:[5,29],usag:[39,41],use_param_index:41,useful:11,user:[2,4,8,9,11,13,15,21,24,26,29,31,36,39,40,41,44],usual:44,util:[16,23,25,36,40,43,44],utils_class:[16,25,43],utils_classif:[16,25,43],utils_color_palett:[16,25,43],utils_conv_get_flat_dim:[16,25,43],utils_cuda:[16,25,43],utils_data:[16,25,43],utils_img_sav:[16,25,43],utils_request_chain_build:[25,31,39],utils_task:[16,25,43],utils_task_dset:[16,25,43],utils_wrapdset_patch:[16,25,43],v:[1,29,39],v_e:1,v_i:29,vabuild:34,vae:[2,16,25,31,32,39],vaechainnodegett:[34,39],vaexydclassif:39,val:[27,30,41],val_threshold:27,valid:[9,22,23,24,27,28,36,40],valu:[7,8,9,11,20,22,24,25,29,32,35,39,40,41],var_grads_and_loss:29,variabl:[2,6,15,29,35,39,44],varianc:[0,2,15,29,35,39,42],variance_between_dict:29,variat:[8,39,41,42],variational_plot:8,variou:[8,22,24,44],varphi:44,vec_d:[34,41],vec_domain:40,vec_i:[29,34,39,41],vec_top:35,vec_z:35,vector:[29,35,39,40,41],vehicl:22,vendor:44,verbos:24,veri:[11,44],verion:7,version:[7,8,34,42,44],via:[2,3,4,8,13,22,27,28,29,39,40,44],virtual:[29,44],virtual_ref_dset_s:30,vision:[4,36,39],visitor:[27,28,39],visual:8,vital:34,vjp:32,vlc:24,vlcs_diva_mldg_di:44,vlcs_mini:[16,24],vogel:22,vol:19,vpicn:39,w:[3,24,29,32,40],wa:[7,11,27,41],wai:[11,22,44],wang:[5,29],want:[7,30,34,41],warm:[2,24],warmup:[24,29,42],wdot:29,we:[1,2,3,6,7,8,9,10,11,20,22,27,28,29,30,34,40,41,44],weav:36,websit:46,weight:[0,1,2,4,5,6,24,29,31,33,34,39],well:[2,7,8,22,24,42],went:41,were:[11,32],what:[22,23,44],when:[11,23,27,29,33,40,41],where:[2,3,6,7,8,13,15,22,23,24,25,29,30,35,39,40,41,44],wherebi:[1,4],wheter:29,whether:[11,28,32,44],which:[0,1,2,3,4,6,7,8,9,10,11,15,20,21,22,24,27,29,32,33,34,35,39,40,41,44],whole:[25,33],width:[22,32,40],wih:40,wise:[0,21,23,29,35],within:[1,2,8,9,32,39,40],without:[2,11,24,32,35,39,40,44],won:34,work:[8,11,21,31,33,34,42,44],workshop:2,wrap:40,wrapdsetpatch:36,write:3,written:39,x1:[29,30],x2:[29,30],x:[0,2,3,6,29,32,34,35,39,40,44],x_1:40,x_2:40,x_:35,x_a:6,x_adv:39,x_arg:41,x_b:6,x_c:6,x_e:29,x_i:6,x_j:6,x_k:6,x_natur:39,xiong:[5,29],xudong:[2,29,39,44],xx:40,xxx:29,xyd:39,xydencoderparallel:35,xydencoderparallelalex:35,xydencoderparallelconvbnrelupool:35,xydencoderparallelextern:35,xydencoderparallelus:35,xydtencoderarg:35,xydtencoderelev:35,y:[2,3,6,15,29,34,39,40,44],y_1:40,y_2:40,y_:6,y_dim:41,y_e:29,y_i:6,y_j:6,yaml:[8,24,41,44],yaroslav:[14,39],year:[2,44],yet:[8,41],yield:[0,6],yingbo:[5,29],yml:9,you:[0,2,8,11,17,22,23,24,32,33,44],your:[8,11,23],your_benchmark_configur:44,yy:40,z:35,z_d:[2,34],z_dim:[32,35],z_vec:32,z_x:2,z_y:[2,34],zd:[24,34,35],zd_dim:[2,11,15,24,34,35,39],zd_mean:2,zd_q:35,zd_scale:2,zdata:16,zero:[7,40],zhou:[5,29],zip:40,zoo_algo:[16,25,43],zoo_nn:[16,25,43],zoo_task:[16,25,43],zoo_train:[23,25,26],zoo_vae_builders_classif:[25,31],zoo_vae_builders_classif_top:[25,31],zoutput:[8,11,21,36],ztest_fil:16,zx2fill:34,zx:[34,35],zx_dim:[2,11,15,24,34,35,39],zx_q:35,zy:[34,35],zy_dim:[2,11,15,24,34,35,39],zy_q:35,zz:40},titles:["Trainer DIAL","Trainer Fishr","Model HDUVA","Trainer IRM (Invariant Risk Minimization)","Model JiGen","Simple Moving Average","Trainer MatchDG","colored MNIST classification","Benchmarking with DomainLab","Further explanations to Benchmark Setup","Let DomainLab know where your PACS data were located","Benchmark yaml files","Deep CORAL","Specify neural network in command line","Model DANN","Model DIVA","Software Architecture and Design","Installation of DomainLab","Causal IRL","Trainer MLDG","Model Specification","Output structure","Task Specification","Model Specification","DomainLab Usage Guide","domainlab package","domainlab.algos package","domainlab.algos.msels package","domainlab.algos.observers package","domainlab.algos.trainers package","domainlab.algos.trainers.compos package","domainlab.compos package","domainlab.compos.nn_zoo package","domainlab.compos.pcr package","domainlab.compos.vae package","domainlab.compos.vae.compos package","domainlab.dsets package","domainlab.exp package","domainlab.exp_protocol package","domainlab.models package","domainlab.tasks package","domainlab.utils package","Welcome to domainlab\u2019s documentation!","domainlab","Introduction","recommonmark==0.5.0.dev0","<no title>"],titleterms:{"0":45,"1":[20,22],"2":[2,20,22],"3":[20,22],"5":45,"8":22,"abstract":20,"case":13,"class":[20,22],"function":[0,2],FOR:2,a_algo_build:26,a_dset_mnist_color_rgb_solo:36,a_model:39,a_model_build:34,a_model_classif:39,a_model_sel:27,a_nn_build:31,a_observ:28,a_task:40,a_task_classif:40,a_train:29,a_vae_build:34,adapt:12,add:20,addit:20,advanc:24,adversari:[0,14],against:44,aggregate_result:38,alex:2,algo:[26,27,28,29,30],algorithm:22,align:12,alongsid:20,also:2,api:[22,44],architectur:16,arg_pars:25,args_dial:29,args_jigen:39,args_va:39,argument:[11,20,24,44],auto:2,autoencod:15,avail:44,averag:5,b_obvisitor:28,b_task:40,b_task_classif:40,benchmark:[8,9,11,44],between:[1,8],bound:2,builder:20,builder_api_model:26,builder_custom:26,builder_dann:26,builder_diva:26,builder_erm:26,builder_hduva:26,builder_jigen1:26,builder_nn_alex:31,builder_nn_conv_bn_relu_2:31,builder_nn_external_from_fil:31,c_msel_oracl:27,c_msel_tr_loss:27,c_msel_val:27,c_obvisitor_cleanup:28,c_obvisitor_gen:28,c_vae_adaptor_model_recon:34,c_vae_builder_classif:34,c_vae_recon:34,cathegor:11,causal:[6,18],check:[4,8,22],choos:15,citat:44,classif:[2,7],cli:25,cluster:8,code:16,color:[2,7,15],combin:11,command:[13,24,44],commandlin:44,compo:[30,31,32,33,34,35],comput:8,configur:[24,44],constraint:9,content:[25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],coral:12,correl:12,creat:[17,20],csv:8,custom:[2,6,13,15,20,22],dann:[13,14],data:[10,22],dataset:[7,22],decoder_concat_vec_reshape_conv:35,decoder_concat_vec_reshape_conv_gated_conv:35,decoder_cond_prior:35,decoder_loss:35,deep:12,defin:22,depend:[8,17],descript:11,design:16,detail:[7,8],dev0:45,develop:17,dial:[0,24],differ:44,directli:22,distribut:[1,11,18,44],diva:[0,7,15],document:42,domain2:22,domain:[0,1,2,6,7,12,14,15,19,22,44],domainlab:[8,10,17,22,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44],dset:36,dset_img_path_list:36,dset_mnist_color_solo_default:36,dset_poly_domains_mnist_color_default:36,dset_subfold:36,dure:1,each:22,emper:7,encod:[2,35],encoder_dirichlet:35,encoder_domain_top:35,encoder_domain_topic_img2top:35,encoder_domain_topic_img_topic2zd:35,encoder_xyd_parallel:35,encoder_xydt_elev:35,encoder_zi:35,environ:17,erm:[7,13],error:8,essenti:24,evid:2,exampl:[0,1,2,3,4,6,12,13,15,18,19,22,23,24,44],exist:20,exp:37,exp_cuda_se:37,exp_main:37,exp_protocol:38,exp_util:37,explan:[7,9],extend:[20,23],extern:22,file:[8,11,22,44],fishr:1,flows_gen_img_model:41,folder:[8,22],from:[8,22],further:[9,24],gener:[0,1,2,6,8,11,15,19,44],generalis:7,generate_benchmark_plot:41,get:44,get_git_tag:41,github:17,gpu:8,gradient:[0,1],guid:24,hduva:[2,6],hierarch:2,hpc:8,hperparamet:11,hyper:0,hyper_schedul:29,hyperparamet:[2,8,9,11,24],hyperparameter_gridsearch:41,hyperparameter_retriev:41,hyperparameter_sampl:41,id:8,imag:[13,15],imagepath:22,implement:[1,2,4,6,13,15,20,22,23],inconsist:1,index:8,indic:42,instal:[8,17,44],instanc:22,interface_vae_xyd:39,introduct:44,invari:[0,1,3,15,18,44],irl:18,irm:3,irmv1:3,jigen:[4,24],job:8,know:10,larger:13,layer:0,learn:[0,19],leav:15,let:10,line:[13,44],locat:10,logger:41,lognorm:11,loguniform:11,loss:[0,2],lower:2,machin:8,make:22,manag:17,map:8,match:[1,6,18],matchdg:[6,13,24],matchdg_arg:30,matchdg_match:30,matchdg_util:30,mechan:18,meta:19,method:[20,44],minim:[3,7],mk_exp:25,mldg:19,mmd_base:29,mnist:[2,6,7,15],model:[0,2,4,13,14,15,20,23,24,39],model_custom:39,model_dann:39,model_diva:39,model_erm:39,model_hduva:39,model_jigen:39,model_vae_xyd_classif:39,modul:[25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41],modular:44,motiv:6,move:5,msel:27,net:2,net_adversari:32,net_classif:32,net_conv_conv_bn_pool_2:32,net_gat:32,network:[0,2,4,6,13,14,15,22,44],neural:[2,4,6,13,14,15,22,44],nn:32,nn_alex:32,nn_torchvis:32,nn_zoo:32,node:8,normal:11,observ:28,obtain:8,one:15,onli:22,option:[20,24],organ:22,out:[1,15],output:[8,21],overal:2,override_interfac:41,overview:2,p_chain_handl:33,pac:10,packag:[25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,44],paramet:[0,4,15,24],partial:8,pcr:33,perf:41,perf_metr:41,plot:8,possibl:22,programm:44,prompt:7,python:[22,44],quantifi:1,recommend:17,recommonmark:45,referenc:11,releas:17,request:33,requir:20,resourc:24,result:[7,8],revers:0,risk:[3,7],run:8,run_experi:38,s:42,sampl:[8,9,11],sandwich:2,saniti:22,sanity_check:41,sanniti:4,set:[0,8],setup:[9,11],share:11,shift:44,shuffl:4,simpl:5,singl:2,slurm:8,so:2,softwar:16,special:11,specif:[11,20,22,23,24,44],specifi:[13,22],standalon:8,start:44,step:[20,23],strongli:17,structur:[0,8,16,21],submodul:[25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41],subpackag:[25,26,29,31,34],tabl:42,task:[11,22,24,40,44],task_dset:40,task_fold:40,task_folder_mk:40,task_mini_vlc:40,task_mnist_color:40,task_pathlist:40,task_util:40,taskfold:22,taskpathlist:22,test:[15,22],test_img:41,theori:6,through:18,tile:4,topic:2,train:[0,1,2,6,15,22,44],train_bas:29,train_causirl:29,train_cor:29,train_dial:29,train_ema:29,train_fishr:29,train_hyper_schedul:29,train_irm:29,train_matchdg:29,train_mldg:29,trainer:[0,1,3,6,13,19,29,30],trainerbas:23,u_import:41,u_import_net_modul:41,uniform:11,unix:8,unsupervis:2,up:8,us:[2,6,13,22],usag:[24,44],user:22,util:[26,41],utils_class:41,utils_classif:41,utils_color_palett:36,utils_conv_get_flat_dim:31,utils_cuda:41,utils_data:36,utils_img_sav:41,utils_request_chain_build:34,utils_task:40,utils_task_dset:40,utils_wrapdset_patch:36,vae:[24,34,35],varianc:1,variat:[2,15],version:[15,17],via:17,virtual:17,warmup:2,welcom:42,well:0,were:10,where:10,window:8,without:8,work:2,yaml:11,your:10,zoo_algo:26,zoo_nn:31,zoo_task:40,zoo_train:29,zoo_vae_builders_classif:34,zoo_vae_builders_classif_top:34}}) \ No newline at end of file diff --git a/docs/build/html/tips.html b/docs/build/html/tips.html index ccd600f31..eb3d2be92 100644 --- a/docs/build/html/tips.html +++ b/docs/build/html/tips.html @@ -1,10 +1,10 @@ - + - + @@ -55,7 +55,6 @@ - @@ -308,13 +307,6 @@ Causal IRL - -
  • - - - Invariant Causal Mechanisms through Distribution Matching - -
  • @@ -366,7 +358,7 @@

    Created using - Sphinx 5.0.2. + Sphinx 4.4.0. and Material for Sphinx