diff --git a/_downloads/5fdddbed2260616231dbf7b0d94bb665/train.txt b/_downloads/5fdddbed2260616231dbf7b0d94bb665/train.txt index 502220837..c7de0dfca 100644 --- a/_downloads/5fdddbed2260616231dbf7b0d94bb665/train.txt +++ b/_downloads/5fdddbed2260616231dbf7b0d94bb665/train.txt @@ -1,16 +1,16 @@ -2024-07-09 14:11:39 (INFO): Running in non-distributed local mode -2024-07-09 14:11:39 (INFO): Project root: /home/runner/work/fairchem/fairchem/src/fairchem -2024-07-09 14:11:40 (INFO): amp: true +2024-07-10 16:28:36 (INFO): Running in non-distributed local mode +2024-07-10 16:28:36 (INFO): Project root: /home/runner/work/fairchem/fairchem/src/fairchem +2024-07-10 16:28:37 (INFO): amp: true cmd: - checkpoint_dir: fine-tuning/checkpoints/2024-07-09-14-11-12-ft-oxides - commit: bf2b26e + checkpoint_dir: fine-tuning/checkpoints/2024-07-10-16-27-44-ft-oxides + commit: 36a17ad identifier: ft-oxides - logs_dir: fine-tuning/logs/tensorboard/2024-07-09-14-11-12-ft-oxides + logs_dir: fine-tuning/logs/tensorboard/2024-07-10-16-27-44-ft-oxides print_every: 10 - results_dir: fine-tuning/results/2024-07-09-14-11-12-ft-oxides + results_dir: fine-tuning/results/2024-07-10-16-27-44-ft-oxides seed: 0 - timestamp_id: 2024-07-09-14-11-12-ft-oxides - version: 0.1.dev1+gbf2b26e + timestamp_id: 2024-07-10-16-27-44-ft-oxides + version: 0.1.dev1+g36a17ad dataset: a2g_args: r_energy: true @@ -141,83 +141,83 @@ val_dataset: format: ase_db src: val.db -2024-07-09 14:11:40 (INFO): Loading dataset: ase_db -2024-07-09 14:11:40 (INFO): rank: 0: Sampler created... -2024-07-09 14:11:40 (INFO): Batch balancing is disabled for single GPU training. -2024-07-09 14:11:40 (INFO): rank: 0: Sampler created... -2024-07-09 14:11:40 (INFO): Batch balancing is disabled for single GPU training. -2024-07-09 14:11:40 (INFO): rank: 0: Sampler created... -2024-07-09 14:11:40 (INFO): Batch balancing is disabled for single GPU training. -2024-07-09 14:11:40 (INFO): Loading model: gemnet_oc -2024-07-09 14:11:40 (WARNING): Unrecognized arguments: ['symmetric_edge_symmetrization'] -2024-07-09 14:11:43 (INFO): Loaded GemNetOC with 38864438 parameters. -2024-07-09 14:11:43 (WARNING): log_summary for Tensorboard not supported -2024-07-09 14:11:43 (WARNING): Using `weight_decay` from `optim` instead of `optim.optimizer_params`.Please update your config to use `optim.optimizer_params.weight_decay`.`optim.weight_decay` will soon be deprecated. -2024-07-09 14:11:43 (INFO): Loading checkpoint from: /tmp/ocp_checkpoints/gnoc_oc22_oc20_all_s2ef.pt -2024-07-09 14:11:43 (INFO): Overwriting scaling factors with those loaded from checkpoint. If you're generating predictions with a pretrained checkpoint, this is the correct behavior. To disable this, delete `scale_dict` from the checkpoint. +2024-07-10 16:28:37 (INFO): Loading dataset: ase_db +2024-07-10 16:28:37 (INFO): rank: 0: Sampler created... +2024-07-10 16:28:37 (INFO): Batch balancing is disabled for single GPU training. +2024-07-10 16:28:37 (INFO): rank: 0: Sampler created... +2024-07-10 16:28:37 (INFO): Batch balancing is disabled for single GPU training. +2024-07-10 16:28:37 (INFO): rank: 0: Sampler created... +2024-07-10 16:28:37 (INFO): Batch balancing is disabled for single GPU training. +2024-07-10 16:28:37 (INFO): Loading model: gemnet_oc +2024-07-10 16:28:37 (WARNING): Unrecognized arguments: ['symmetric_edge_symmetrization'] +2024-07-10 16:28:39 (INFO): Loaded GemNetOC with 38864438 parameters. +2024-07-10 16:28:39 (WARNING): log_summary for Tensorboard not supported +2024-07-10 16:28:39 (WARNING): Using `weight_decay` from `optim` instead of `optim.optimizer_params`.Please update your config to use `optim.optimizer_params.weight_decay`.`optim.weight_decay` will soon be deprecated. +2024-07-10 16:28:40 (INFO): Loading checkpoint from: /tmp/ocp_checkpoints/gnoc_oc22_oc20_all_s2ef.pt +2024-07-10 16:28:40 (INFO): Overwriting scaling factors with those loaded from checkpoint. If you're generating predictions with a pretrained checkpoint, this is the correct behavior. To disable this, delete `scale_dict` from the checkpoint. /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) -2024-07-09 14:12:06 (INFO): energy_mae: 6.77e+00, forcesx_mae: 4.11e-02, forcesy_mae: 3.63e-02, forcesz_mae: 5.26e-02, forces_mae: 4.33e-02, forces_cosine_similarity: 8.24e-02, forces_magnitude_error: 7.42e-02, energy_forces_within_threshold: 0.00e+00, loss: 6.86e+00, lr: 5.00e-04, epoch: 1.69e-01, step: 1.00e+01 -2024-07-09 14:12:07 (INFO): Evaluating on val. +2024-07-10 16:29:02 (INFO): energy_mae: 6.77e+00, forcesx_mae: 4.11e-02, forcesy_mae: 3.63e-02, forcesz_mae: 5.26e-02, forces_mae: 4.33e-02, forces_cosine_similarity: 8.24e-02, forces_magnitude_error: 7.42e-02, energy_forces_within_threshold: 0.00e+00, loss: 6.86e+00, lr: 5.00e-04, epoch: 1.69e-01, step: 1.00e+01 +2024-07-10 16:29:03 (INFO): Evaluating on val. device 0: 0%| | 0/2 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 50%|█████ | 1/2 [00:04<00:04, 4.77s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.65s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.87s/it] -2024-07-09 14:12:15 (INFO): energy_mae: 8.7333, forcesx_mae: 0.0231, forcesy_mae: 0.0188, forcesz_mae: 0.0198, forces_mae: 0.0206, forces_cosine_similarity: -0.0317, forces_magnitude_error: 0.0290, energy_forces_within_threshold: 0.0000, loss: 8.6591, epoch: 0.1695 -2024-07-09 14:12:15 (INFO): Predicting on test. + device 0: 50%|█████ | 1/2 [00:04<00:04, 4.75s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.64s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.85s/it] +2024-07-10 16:29:11 (INFO): energy_mae: 8.7333, forcesx_mae: 0.0231, forcesy_mae: 0.0188, forcesz_mae: 0.0198, forces_mae: 0.0206, forces_cosine_similarity: -0.0317, forces_magnitude_error: 0.0290, energy_forces_within_threshold: 0.0000, loss: 8.6590, epoch: 0.1695 +2024-07-10 16:29:11 (INFO): Predicting on test. device 0: 0%| | 0/2 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 50%|█████ | 1/2 [00:02<00:02, 2.95s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.76s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.87s/it] -2024-07-09 14:12:21 (INFO): Writing results to fine-tuning/results/2024-07-09-14-11-12-ft-oxides/ocp_predictions.npz -2024-07-09 14:12:46 (INFO): energy_mae: 1.07e+01, forcesx_mae: 2.63e-02, forcesy_mae: 1.76e-02, forcesz_mae: 1.90e-02, forces_mae: 2.10e-02, forces_cosine_similarity: -3.68e-02, forces_magnitude_error: 2.26e-02, energy_forces_within_threshold: 0.00e+00, loss: 1.07e+01, lr: 5.00e-04, epoch: 3.39e-01, step: 2.00e+01 -2024-07-09 14:12:48 (INFO): Evaluating on val. + device 0: 50%|█████ | 1/2 [00:03<00:03, 3.03s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.82s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.92s/it] +2024-07-10 16:29:17 (INFO): Writing results to fine-tuning/results/2024-07-10-16-27-44-ft-oxides/ocp_predictions.npz +2024-07-10 16:29:43 (INFO): energy_mae: 1.07e+01, forcesx_mae: 2.63e-02, forcesy_mae: 1.76e-02, forcesz_mae: 1.90e-02, forces_mae: 2.10e-02, forces_cosine_similarity: -3.68e-02, forces_magnitude_error: 2.26e-02, energy_forces_within_threshold: 0.00e+00, loss: 1.07e+01, lr: 5.00e-04, epoch: 3.39e-01, step: 2.00e+01 +2024-07-10 16:29:44 (INFO): Evaluating on val. device 0: 0%| | 0/2 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 50%|█████ | 1/2 [00:04<00:04, 4.80s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.79s/it] device 0: 100%|██████████| 2/2 [00:08<00:00, 4.02s/it] -2024-07-09 14:12:56 (INFO): energy_mae: 2.6849, forcesx_mae: 0.0235, forcesy_mae: 0.0180, forcesz_mae: 0.0129, forces_mae: 0.0181, forces_cosine_similarity: 0.0675, forces_magnitude_error: 0.0215, energy_forces_within_threshold: 0.0000, loss: 2.7128, epoch: 0.3390 -2024-07-09 14:12:57 (INFO): Predicting on test. + device 0: 50%|█████ | 1/2 [00:04<00:04, 4.82s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.73s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.97s/it] +2024-07-10 16:29:53 (INFO): energy_mae: 2.6840, forcesx_mae: 0.0235, forcesy_mae: 0.0180, forcesz_mae: 0.0129, forces_mae: 0.0181, forces_cosine_similarity: 0.0674, forces_magnitude_error: 0.0215, energy_forces_within_threshold: 0.0000, loss: 2.7120, epoch: 0.3390 +2024-07-10 16:29:53 (INFO): Predicting on test. device 0: 0%| | 0/2 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 50%|█████ | 1/2 [00:03<00:03, 3.01s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.87s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.97s/it] -2024-07-09 14:13:03 (INFO): Writing results to fine-tuning/results/2024-07-09-14-11-12-ft-oxides/ocp_predictions.npz -2024-07-09 14:13:22 (INFO): energy_mae: 4.87e+00, forcesx_mae: 1.81e-02, forcesy_mae: 1.51e-02, forcesz_mae: 2.08e-02, forces_mae: 1.80e-02, forces_cosine_similarity: -9.49e-03, forces_magnitude_error: 2.32e-02, energy_forces_within_threshold: 0.00e+00, loss: 4.91e+00, lr: 5.00e-04, epoch: 5.08e-01, step: 3.00e+01 -2024-07-09 14:13:24 (INFO): Evaluating on val. + device 0: 50%|█████ | 1/2 [00:03<00:03, 3.04s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.87s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.98s/it] +2024-07-10 16:29:59 (INFO): Writing results to fine-tuning/results/2024-07-10-16-27-44-ft-oxides/ocp_predictions.npz +2024-07-10 16:30:19 (INFO): energy_mae: 4.87e+00, forcesx_mae: 1.82e-02, forcesy_mae: 1.51e-02, forcesz_mae: 2.07e-02, forces_mae: 1.80e-02, forces_cosine_similarity: -9.75e-03, forces_magnitude_error: 2.32e-02, energy_forces_within_threshold: 0.00e+00, loss: 4.91e+00, lr: 5.00e-04, epoch: 5.08e-01, step: 3.00e+01 +2024-07-10 16:30:20 (INFO): Evaluating on val. device 0: 0%| | 0/2 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 50%|█████ | 1/2 [00:04<00:04, 4.62s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.50s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.77s/it] -2024-07-09 14:13:32 (INFO): energy_mae: 4.4434, forcesx_mae: 0.0268, forcesy_mae: 0.0290, forcesz_mae: 0.0330, forces_mae: 0.0296, forces_cosine_similarity: -0.0265, forces_magnitude_error: 0.0418, energy_forces_within_threshold: 0.0000, loss: 4.5329, epoch: 0.5085 -2024-07-09 14:13:57 (INFO): energy_mae: 3.76e+00, forcesx_mae: 3.12e-02, forcesy_mae: 2.67e-02, forcesz_mae: 3.30e-02, forces_mae: 3.03e-02, forces_cosine_similarity: -1.19e-01, forces_magnitude_error: 3.96e-02, energy_forces_within_threshold: 0.00e+00, loss: 3.82e+00, lr: 5.00e-04, epoch: 6.78e-01, step: 4.00e+01 -2024-07-09 14:13:59 (INFO): Evaluating on val. + device 0: 50%|█████ | 1/2 [00:04<00:04, 4.70s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.55s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.82s/it] +2024-07-10 16:30:28 (INFO): energy_mae: 4.4268, forcesx_mae: 0.0263, forcesy_mae: 0.0285, forcesz_mae: 0.0323, forces_mae: 0.0290, forces_cosine_similarity: -0.0244, forces_magnitude_error: 0.0407, energy_forces_within_threshold: 0.0000, loss: 4.5148, epoch: 0.5085 +2024-07-10 16:30:53 (INFO): energy_mae: 4.11e+00, forcesx_mae: 2.74e-02, forcesy_mae: 2.14e-02, forcesz_mae: 2.78e-02, forces_mae: 2.55e-02, forces_cosine_similarity: -1.74e-02, forces_magnitude_error: 3.37e-02, energy_forces_within_threshold: 0.00e+00, loss: 4.16e+00, lr: 5.00e-04, epoch: 6.78e-01, step: 4.00e+01 +2024-07-10 16:30:55 (INFO): Evaluating on val. device 0: 0%| | 0/2 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 50%|█████ | 1/2 [00:04<00:04, 4.65s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.57s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.82s/it] -2024-07-09 14:14:06 (INFO): energy_mae: 1.2410, forcesx_mae: 0.0160, forcesy_mae: 0.0177, forcesz_mae: 0.0147, forces_mae: 0.0161, forces_cosine_similarity: 0.1347, forces_magnitude_error: 0.0191, energy_forces_within_threshold: 0.0000, loss: 1.2659, epoch: 0.6780 -2024-07-09 14:14:07 (INFO): Predicting on test. + device 0: 50%|█████ | 1/2 [00:04<00:04, 4.42s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.56s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.79s/it] +2024-07-10 16:31:03 (INFO): energy_mae: 2.6334, forcesx_mae: 0.0111, forcesy_mae: 0.0120, forcesz_mae: 0.0123, forces_mae: 0.0118, forces_cosine_similarity: 0.0739, forces_magnitude_error: 0.0141, energy_forces_within_threshold: 0.0000, loss: 2.6224, epoch: 0.6780 +2024-07-10 16:31:03 (INFO): Predicting on test. device 0: 0%| | 0/2 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 50%|█████ | 1/2 [00:03<00:03, 3.20s/it] device 0: 100%|██████████| 2/2 [00:05<00:00, 2.95s/it] device 0: 100%|██████████| 2/2 [00:06<00:00, 3.08s/it] -2024-07-09 14:14:13 (INFO): Writing results to fine-tuning/results/2024-07-09-14-11-12-ft-oxides/ocp_predictions.npz -2024-07-09 14:14:34 (INFO): energy_mae: 2.93e+00, forcesx_mae: 1.71e-02, forcesy_mae: 1.54e-02, forcesz_mae: 1.70e-02, forces_mae: 1.65e-02, forces_cosine_similarity: 2.49e-02, forces_magnitude_error: 2.03e-02, energy_forces_within_threshold: 0.00e+00, loss: 2.96e+00, lr: 5.00e-04, epoch: 8.47e-01, step: 5.00e+01 -2024-07-09 14:14:36 (INFO): Evaluating on val. + device 0: 50%|█████ | 1/2 [00:03<00:03, 3.17s/it] device 0: 100%|██████████| 2/2 [00:06<00:00, 3.04s/it] device 0: 100%|██████████| 2/2 [00:06<00:00, 3.16s/it] +2024-07-10 16:31:09 (INFO): Writing results to fine-tuning/results/2024-07-10-16-27-44-ft-oxides/ocp_predictions.npz +2024-07-10 16:31:30 (INFO): energy_mae: 2.56e+00, forcesx_mae: 2.10e-02, forcesy_mae: 1.88e-02, forcesz_mae: 1.99e-02, forces_mae: 1.99e-02, forces_cosine_similarity: -8.99e-02, forces_magnitude_error: 2.54e-02, energy_forces_within_threshold: 0.00e+00, loss: 2.60e+00, lr: 5.00e-04, epoch: 8.47e-01, step: 5.00e+01 +2024-07-10 16:31:32 (INFO): Evaluating on val. device 0: 0%| | 0/2 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 50%|█████ | 1/2 [00:04<00:04, 4.37s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.48s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.71s/it] -2024-07-09 14:14:43 (INFO): energy_mae: 4.0707, forcesx_mae: 0.0248, forcesy_mae: 0.0212, forcesz_mae: 0.0229, forces_mae: 0.0230, forces_cosine_similarity: 0.0864, forces_magnitude_error: 0.0363, energy_forces_within_threshold: 0.0000, loss: 4.0942, epoch: 0.8475 -2024-07-09 14:15:02 (INFO): Total time taken: 198.98523592948914 + device 0: 50%|█████ | 1/2 [00:04<00:04, 4.73s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.68s/it] device 0: 100%|██████████| 2/2 [00:07<00:00, 3.95s/it] +2024-07-10 16:31:40 (INFO): energy_mae: 2.1851, forcesx_mae: 0.0230, forcesy_mae: 0.0225, forcesz_mae: 0.0232, forces_mae: 0.0229, forces_cosine_similarity: 0.0011, forces_magnitude_error: 0.0311, energy_forces_within_threshold: 0.0000, loss: 2.2285, epoch: 0.8475 +2024-07-10 16:31:59 (INFO): Total time taken: 199.46588730812073 diff --git a/_downloads/819e10305ddd6839cd7da05935b17060/mass-inference.txt b/_downloads/819e10305ddd6839cd7da05935b17060/mass-inference.txt index 8395c31fe..58f8f1bde 100644 --- a/_downloads/819e10305ddd6839cd7da05935b17060/mass-inference.txt +++ b/_downloads/819e10305ddd6839cd7da05935b17060/mass-inference.txt @@ -1,16 +1,16 @@ -2024-07-09 14:17:18 (INFO): Running in non-distributed local mode -2024-07-09 14:17:19 (INFO): Project root: /home/runner/work/fairchem/fairchem/src/fairchem -2024-07-09 14:17:20 (INFO): amp: true +2024-07-10 16:34:01 (INFO): Running in non-distributed local mode +2024-07-10 16:34:01 (INFO): Project root: /home/runner/work/fairchem/fairchem/src/fairchem +2024-07-10 16:34:02 (INFO): amp: true cmd: - checkpoint_dir: ./checkpoints/2024-07-09-14-17-36 - commit: bf2b26e + checkpoint_dir: ./checkpoints/2024-07-10-16-34-08 + commit: 36a17ad identifier: '' - logs_dir: ./logs/tensorboard/2024-07-09-14-17-36 + logs_dir: ./logs/tensorboard/2024-07-10-16-34-08 print_every: 10 - results_dir: ./results/2024-07-09-14-17-36 + results_dir: ./results/2024-07-10-16-34-08 seed: 0 - timestamp_id: 2024-07-09-14-17-36 - version: 0.1.dev1+gbf2b26e + timestamp_id: 2024-07-10-16-34-08 + version: 0.1.dev1+g36a17ad dataset: null evaluation_metrics: metrics: @@ -112,20 +112,20 @@ test_dataset: trainer: ocp val_dataset: null -2024-07-09 14:17:20 (INFO): rank: 0: Sampler created... -2024-07-09 14:17:20 (INFO): Batch balancing is disabled for single GPU training. -2024-07-09 14:17:20 (INFO): Loading model: gemnet_t -2024-07-09 14:17:22 (INFO): Loaded GemNetT with 31671825 parameters. -2024-07-09 14:17:22 (WARNING): log_summary for Tensorboard not supported -2024-07-09 14:17:22 (INFO): Loading checkpoint from: /tmp/ocp_checkpoints/gndt_oc22_all_s2ef.pt -2024-07-09 14:17:22 (INFO): Overwriting scaling factors with those loaded from checkpoint. If you're generating predictions with a pretrained checkpoint, this is the correct behavior. To disable this, delete `scale_dict` from the checkpoint. -2024-07-09 14:17:22 (WARNING): Scale factor comment not found in model -2024-07-09 14:17:22 (INFO): Predicting on test. +2024-07-10 16:34:02 (INFO): rank: 0: Sampler created... +2024-07-10 16:34:02 (INFO): Batch balancing is disabled for single GPU training. +2024-07-10 16:34:02 (INFO): Loading model: gemnet_t +2024-07-10 16:34:04 (INFO): Loaded GemNetT with 31671825 parameters. +2024-07-10 16:34:04 (WARNING): log_summary for Tensorboard not supported +2024-07-10 16:34:04 (INFO): Loading checkpoint from: /tmp/ocp_checkpoints/gndt_oc22_all_s2ef.pt +2024-07-10 16:34:04 (INFO): Overwriting scaling factors with those loaded from checkpoint. If you're generating predictions with a pretrained checkpoint, this is the correct behavior. To disable this, delete `scale_dict` from the checkpoint. +2024-07-10 16:34:04 (WARNING): Scale factor comment not found in model +2024-07-10 16:34:04 (INFO): Predicting on test. device 0: 0%| | 0/3 [00:00, ?it/s]/opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch_geometric/data/collate.py:145: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() storage = elem.storage()._new_shared(numel) - device 0: 33%|███████████▋ | 1/3 [00:02<00:04, 2.03s/it] device 0: 67%|███████████████████████▎ | 2/3 [00:04<00:02, 2.51s/it] device 0: 100%|███████████████████████████████████| 3/3 [00:06<00:00, 2.11s/it] device 0: 100%|███████████████████████████████████| 3/3 [00:06<00:00, 2.18s/it] -2024-07-09 14:17:28 (INFO): Writing results to ./results/2024-07-09-14-17-36/ocp_predictions.npz -2024-07-09 14:17:28 (INFO): Total time taken: 6.688116550445557 -Elapsed time = 13.0 seconds + device 0: 33%|███████████▋ | 1/3 [00:03<00:07, 3.82s/it] device 0: 67%|███████████████████████▎ | 2/3 [00:07<00:03, 3.47s/it] device 0: 100%|███████████████████████████████████| 3/3 [00:08<00:00, 2.43s/it] device 0: 100%|███████████████████████████████████| 3/3 [00:08<00:00, 2.76s/it] +2024-07-10 16:34:13 (INFO): Writing results to ./results/2024-07-10-16-34-08/ocp_predictions.npz +2024-07-10 16:34:13 (INFO): Total time taken: 8.430647373199463 +Elapsed time = 14.9 seconds diff --git a/_images/1bd3f6f137572394f45b2bea46d8a5d05c74e922aca09434f2a962c660c36a98.png b/_images/1bd3f6f137572394f45b2bea46d8a5d05c74e922aca09434f2a962c660c36a98.png deleted file mode 100644 index 011bcbc5a..000000000 Binary files a/_images/1bd3f6f137572394f45b2bea46d8a5d05c74e922aca09434f2a962c660c36a98.png and /dev/null differ diff --git a/_images/20119ebabd48f33476c0d005b06b2a0dd7a215cc88c5f66182ccd469bcff0e8a.png b/_images/20119ebabd48f33476c0d005b06b2a0dd7a215cc88c5f66182ccd469bcff0e8a.png deleted file mode 100644 index b0c9615ff..000000000 Binary files a/_images/20119ebabd48f33476c0d005b06b2a0dd7a215cc88c5f66182ccd469bcff0e8a.png and /dev/null differ diff --git a/_images/336759a1dc6295bc0edbf0a0fb8b76b26c6259d18205582f40cff7f41c093c24.png b/_images/336759a1dc6295bc0edbf0a0fb8b76b26c6259d18205582f40cff7f41c093c24.png new file mode 100644 index 000000000..f2e7dc4a4 Binary files /dev/null and b/_images/336759a1dc6295bc0edbf0a0fb8b76b26c6259d18205582f40cff7f41c093c24.png differ diff --git a/_images/33e748c54c65bf2c62d880db6f712a27cf22f1a234bc72238947a7f83abc602c.png b/_images/33e748c54c65bf2c62d880db6f712a27cf22f1a234bc72238947a7f83abc602c.png deleted file mode 100644 index 87d98ecf5..000000000 Binary files a/_images/33e748c54c65bf2c62d880db6f712a27cf22f1a234bc72238947a7f83abc602c.png and /dev/null differ diff --git a/_images/4c0a3eadc20e85c10fa416a87bd8a8e66ed2141f9d738463e7fc40ceaccf7418.png b/_images/4c0a3eadc20e85c10fa416a87bd8a8e66ed2141f9d738463e7fc40ceaccf7418.png deleted file mode 100644 index c5e5b8c3b..000000000 Binary files a/_images/4c0a3eadc20e85c10fa416a87bd8a8e66ed2141f9d738463e7fc40ceaccf7418.png and /dev/null differ diff --git a/_images/5619bffc5cbf78888b670444701903eb7a06f1d9b467bceef578151bc4f956c6.png b/_images/5619bffc5cbf78888b670444701903eb7a06f1d9b467bceef578151bc4f956c6.png new file mode 100644 index 000000000..e4bdf3930 Binary files /dev/null and b/_images/5619bffc5cbf78888b670444701903eb7a06f1d9b467bceef578151bc4f956c6.png differ diff --git a/_images/6a1d6b0df93617c90cb3eff7991ca38c17638566984c00e87abd508f9ee2ce4a.png b/_images/6a1d6b0df93617c90cb3eff7991ca38c17638566984c00e87abd508f9ee2ce4a.png new file mode 100644 index 000000000..0dcb0c11f Binary files /dev/null and b/_images/6a1d6b0df93617c90cb3eff7991ca38c17638566984c00e87abd508f9ee2ce4a.png differ diff --git a/_images/737b5c27afd393eda611180faa0e6913cf2913d35de9ee976ce8393895b088f2.png b/_images/737b5c27afd393eda611180faa0e6913cf2913d35de9ee976ce8393895b088f2.png new file mode 100644 index 000000000..cae090e99 Binary files /dev/null and b/_images/737b5c27afd393eda611180faa0e6913cf2913d35de9ee976ce8393895b088f2.png differ diff --git a/_images/86c6c30bcba70e4dc7045ed905b57d667d70fd06a61b46a3a2001c30c1b632ec.png b/_images/86c6c30bcba70e4dc7045ed905b57d667d70fd06a61b46a3a2001c30c1b632ec.png new file mode 100644 index 000000000..a116e8a4b Binary files /dev/null and b/_images/86c6c30bcba70e4dc7045ed905b57d667d70fd06a61b46a3a2001c30c1b632ec.png differ diff --git a/_images/aa0f94393b055271b577afa74863460b56e85787ad0daaaab7aaed93b90e85a5.png b/_images/aa0f94393b055271b577afa74863460b56e85787ad0daaaab7aaed93b90e85a5.png new file mode 100644 index 000000000..2067a91a4 Binary files /dev/null and b/_images/aa0f94393b055271b577afa74863460b56e85787ad0daaaab7aaed93b90e85a5.png differ diff --git a/_images/b82ea1738fba5c90b56c624f7c437d6122aebdef0d1d9f361a1986abbc441734.png b/_images/b82ea1738fba5c90b56c624f7c437d6122aebdef0d1d9f361a1986abbc441734.png new file mode 100644 index 000000000..23670819d Binary files /dev/null and b/_images/b82ea1738fba5c90b56c624f7c437d6122aebdef0d1d9f361a1986abbc441734.png differ diff --git a/_images/c138d78d11c7e14374028e10ca0934ab2696fbf6aa9d2b45d44bcffabf6080e2.png b/_images/c138d78d11c7e14374028e10ca0934ab2696fbf6aa9d2b45d44bcffabf6080e2.png deleted file mode 100644 index 376e9183c..000000000 Binary files a/_images/c138d78d11c7e14374028e10ca0934ab2696fbf6aa9d2b45d44bcffabf6080e2.png and /dev/null differ diff --git a/_images/c2d46ce94f5bd95474bf0f35cece29b3531bc7d573c9c716bd1a2006c183352d.png b/_images/c2d46ce94f5bd95474bf0f35cece29b3531bc7d573c9c716bd1a2006c183352d.png new file mode 100644 index 000000000..fa7926815 Binary files /dev/null and b/_images/c2d46ce94f5bd95474bf0f35cece29b3531bc7d573c9c716bd1a2006c183352d.png differ diff --git a/_images/cfcdc7e302775579c03049aef09d50bcdd1691a83bfa4ec1a20e1ac8536e0ce7.png b/_images/cfcdc7e302775579c03049aef09d50bcdd1691a83bfa4ec1a20e1ac8536e0ce7.png deleted file mode 100644 index 584583fc5..000000000 Binary files a/_images/cfcdc7e302775579c03049aef09d50bcdd1691a83bfa4ec1a20e1ac8536e0ce7.png and /dev/null differ diff --git a/_images/d8f000c5866310197f993a7572f8fecc25292a36dab7b71d7508e7eec2cc2434.png b/_images/d8f000c5866310197f993a7572f8fecc25292a36dab7b71d7508e7eec2cc2434.png deleted file mode 100644 index b37c0af57..000000000 Binary files a/_images/d8f000c5866310197f993a7572f8fecc25292a36dab7b71d7508e7eec2cc2434.png and /dev/null differ diff --git a/_images/deabea6d50267eb0199707fd11c4215082556e20ab5e2e17344d1dde87586a2e.png b/_images/deabea6d50267eb0199707fd11c4215082556e20ab5e2e17344d1dde87586a2e.png deleted file mode 100644 index 6603825bc..000000000 Binary files a/_images/deabea6d50267eb0199707fd11c4215082556e20ab5e2e17344d1dde87586a2e.png and /dev/null differ diff --git a/_images/ec3d78c828a48f27837aed2db2ea620c401fa3cc7fec032fead69ca09d84fb29.png b/_images/ec3d78c828a48f27837aed2db2ea620c401fa3cc7fec032fead69ca09d84fb29.png deleted file mode 100644 index 5922b3ab4..000000000 Binary files a/_images/ec3d78c828a48f27837aed2db2ea620c401fa3cc7fec032fead69ca09d84fb29.png and /dev/null differ diff --git a/_images/f9b0bc9c18d7454baec238a7cfbd8239d5be193923cbfba2bd94d16060c45393.png b/_images/f9b0bc9c18d7454baec238a7cfbd8239d5be193923cbfba2bd94d16060c45393.png new file mode 100644 index 000000000..4893e73ab Binary files /dev/null and b/_images/f9b0bc9c18d7454baec238a7cfbd8239d5be193923cbfba2bd94d16060c45393.png differ diff --git a/core/fine-tuning/fine-tuning-oxides.html b/core/fine-tuning/fine-tuning-oxides.html index baab1fc73..816f2b8ac 100644 --- a/core/fine-tuning/fine-tuning-oxides.html +++ b/core/fine-tuning/fine-tuning-oxides.html @@ -827,7 +827,7 @@
Elapsed time 67.7 seconds.
+Elapsed time 68.0 seconds.
@@ -906,7 +906,7 @@ Fine tuning the checkpoint
-/tmp/ipykernel_2423/1448814737.py:12: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2290/1448814737.py:12: DeprecationWarning: Please use atoms.calc = calc
atoms.set_calculator(calc)
@@ -1199,7 +1199,7 @@ Running the training jobHide code cell output
-Elapsed time = 206.6 seconds
+Elapsed time = 207.0 seconds
@@ -1215,7 +1215,7 @@ Running the training job
-'fine-tuning/checkpoints/2024-07-09-14-11-12-ft-oxides'
+'fine-tuning/checkpoints/2024-07-10-16-27-44-ft-oxides'
@@ -1274,7 +1274,7 @@ Running the training job
-
+
@@ -1285,7 +1285,7 @@ Running the training job
-New MAE = 0.324 eV/atom
+New MAE = 0.235 eV/atom
@@ -1304,7 +1304,7 @@ Running the training job
-
+
It is possible to continue refining the fit. The simple things to do are to use more epochs of training. Eventually the MAE will stabilize, and then it may be necessary to adjust other optimization parameters like the learning rate (usually you decrease it).
diff --git a/core/gotchas.html b/core/gotchas.html
index 6f0e44060..98d820224 100644
--- a/core/gotchas.html
+++ b/core/gotchas.html
@@ -731,12 +731,12 @@ I want the energy of a gas phase atom 544 '{} not present in this ' 'calculation'.format(name)
545 )
-File ~/work/fairchem/fairchem/src/fairchem/core/common/relaxation/ase_utils.py:237, in OCPCalculator.calculate(self, atoms, properties, system_changes)
- 234 data_object = self.a2g.convert(atoms)
- 235 batch = data_list_collater([data_object], otf_graph=True)
---> 237 predictions = self.trainer.predict(batch, per_image=False, disable_tqdm=True)
- 239 for key in predictions:
- 240 _pred = predictions[key]
+File ~/work/fairchem/fairchem/src/fairchem/core/common/relaxation/ase_utils.py:235, in OCPCalculator.calculate(self, atoms, properties, system_changes)
+ 232 data_object = self.a2g.convert(atoms)
+ 233 batch = data_list_collater([data_object], otf_graph=True)
+--> 235 predictions = self.trainer.predict(batch, per_image=False, disable_tqdm=True)
+ 237 for key in predictions:
+ 238 _pred = predictions[key]
File /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch/utils/_contextlib.py:115, in context_decorator.<locals>.decorate_context(*args, **kwargs)
112 @functools.wraps(func)
@@ -922,7 +922,7 @@ I get wildly different energies from the different modelsWARNING:root:No seed has been set in modelcheckpoint or OCPCalculator! Results may not be reproducible on re-run
-/tmp/ipykernel_2697/1677958285.py:11: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2561/1677958285.py:11: DeprecationWarning: Please use atoms.calc = calc
slab.set_calculator(calc)
@@ -959,7 +959,7 @@ I get wildly different energies from the different modelsWARNING:root:No seed has been set in modelcheckpoint or OCPCalculator! Results may not be reproducible on re-run
-/tmp/ipykernel_2697/2284884232.py:9: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2561/2284884232.py:9: DeprecationWarning: Please use atoms.calc = calc
slab.set_calculator(calc)
@@ -988,11 +988,11 @@ I get wildly different energies from the different modelsWARNING:root:No seed has been set in modelcheckpoint or OCPCalculator! Results may not be reproducible on re-run
-/tmp/ipykernel_2697/898007276.py:7: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2561/898007276.py:7: DeprecationWarning: Please use atoms.calc = calc
slab.set_calculator(calc)
-1.6709036827087402
+1.6888058185577393
@@ -1099,12 +1099,12 @@ You need at least four atoms for molecules with some models 544 '{} not present in this ' 'calculation'.format(name)
545 )
-File ~/work/fairchem/fairchem/src/fairchem/core/common/relaxation/ase_utils.py:237, in OCPCalculator.calculate(self, atoms, properties, system_changes)
- 234 data_object = self.a2g.convert(atoms)
- 235 batch = data_list_collater([data_object], otf_graph=True)
---> 237 predictions = self.trainer.predict(batch, per_image=False, disable_tqdm=True)
- 239 for key in predictions:
- 240 _pred = predictions[key]
+File ~/work/fairchem/fairchem/src/fairchem/core/common/relaxation/ase_utils.py:235, in OCPCalculator.calculate(self, atoms, properties, system_changes)
+ 232 data_object = self.a2g.convert(atoms)
+ 233 batch = data_list_collater([data_object], otf_graph=True)
+--> 235 predictions = self.trainer.predict(batch, per_image=False, disable_tqdm=True)
+ 237 for key in predictions:
+ 238 _pred = predictions[key]
File /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch/utils/_contextlib.py:115, in context_decorator.<locals>.decorate_context(*args, **kwargs)
112 @functools.wraps(func)
@@ -1309,12 +1309,12 @@ To tag or not? 544 '{} not present in this ' 'calculation'.format(name)
545 )
-File ~/work/fairchem/fairchem/src/fairchem/core/common/relaxation/ase_utils.py:237, in OCPCalculator.calculate(self, atoms, properties, system_changes)
- 234 data_object = self.a2g.convert(atoms)
- 235 batch = data_list_collater([data_object], otf_graph=True)
---> 237 predictions = self.trainer.predict(batch, per_image=False, disable_tqdm=True)
- 239 for key in predictions:
- 240 _pred = predictions[key]
+File ~/work/fairchem/fairchem/src/fairchem/core/common/relaxation/ase_utils.py:235, in OCPCalculator.calculate(self, atoms, properties, system_changes)
+ 232 data_object = self.a2g.convert(atoms)
+ 233 batch = data_list_collater([data_object], otf_graph=True)
+--> 235 predictions = self.trainer.predict(batch, per_image=False, disable_tqdm=True)
+ 237 for key in predictions:
+ 238 _pred = predictions[key]
File /opt/hostedtoolcache/Python/3.11.9/x64/lib/python3.11/site-packages/torch/utils/_contextlib.py:115, in context_decorator.<locals>.decorate_context(*args, **kwargs)
112 @functools.wraps(func)
@@ -1457,7 +1457,7 @@ To tag or not?
-/tmp/ipykernel_2697/3906293788.py:3: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2561/3906293788.py:3: DeprecationWarning: Please use atoms.calc = calc
atoms.set_calculator(calc)
@@ -1501,11 +1501,11 @@ To tag or not?
-/tmp/ipykernel_2697/4094489779.py:3: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2561/4094489779.py:3: DeprecationWarning: Please use atoms.calc = calc
atoms.set_calculator(calc)
--0.4297373294830322
+-0.4297369122505188
@@ -1552,21 +1552,21 @@ Stochastic simulation resultsWARNING:root:No seed has been set in modelcheckpoint or OCPCalculator! Results may not be reproducible on re-run
-/tmp/ipykernel_2697/3095522216.py:11: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2561/3095522216.py:11: DeprecationWarning: Please use atoms.calc = calc
slab.set_calculator(calc)
-1.2139860391616821 1.6319050749442676e-06
-1.2139861583709717
-1.2139849662780762
-1.2139837741851807
-1.213989019393921
-1.213986873626709
-1.2139880657196045
+1.2139862775802612 1.9376552365568983e-06
+1.2139883041381836
+1.2139859199523926
+1.2139892578125
1.2139837741851807
+1.2139856815338135
+1.2139835357666016
+1.2139892578125
+1.2139854431152344
1.2139861583709717
-1.2139866352081299
-1.2139849662780762
+1.2139854431152344
@@ -1604,11 +1604,11 @@ The forces don’t sum to zeroWARNING:root:No seed has been set in modelcheckpoint or OCPCalculator! Results may not be reproducible on re-run
-/tmp/ipykernel_2697/2981184886.py:11: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2561/2981184886.py:11: DeprecationWarning: Please use atoms.calc = calc
slab.set_calculator(calc)
-array([ 0.00848194, 0.01409447, -0.05882657], dtype=float32)
+array([ 0.00848049, 0.0140971 , -0.05883038], dtype=float32)
@@ -1621,7 +1621,7 @@ The forces don’t sum to zero
-array([ 1.3416866e-07, 7.9278834e-08, -1.1920929e-07], dtype=float32)
+array([-1.5000114e-07, 7.9977326e-08, 0.0000000e+00], dtype=float32)
diff --git a/core/inference.html b/core/inference.html
index 7fccd881a..a9108b8bb 100644
--- a/core/inference.html
+++ b/core/inference.html
@@ -646,12 +646,12 @@ Fast batched inference
---2024-07-09 14:17:02-- https://figshare.com/ndownloader/files/11948267
+--2024-07-10 16:33:49-- https://figshare.com/ndownloader/files/11948267
Resolving figshare.com (figshare.com)...
-52.212.236.71, 34.251.213.12, 2a05:d018:1f4:d000:4856:ab6:5d9e:fe0e, ...
-Connecting to figshare.com (figshare.com)|52.212.236.71|:443...
+52.215.106.69, 52.212.236.71, 2a05:d018:1f4:d003:3dee:9eef:2e6:4a2b, ...
+Connecting to figshare.com (figshare.com)|52.215.106.69|:443...
302 Found
-Location: https://s3-eu-west-1.amazonaws.com/pstorage-cmu-348901238291901/11948267/data.db?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAI266R7V6O36O5JUA/20240709/eu-west-1/s3/aws4_request&X-Amz-Date=20240709T141702Z&X-Amz-Expires=10&X-Amz-SignedHeaders=host&X-Amz-Signature=9994bf27457b7fdedb9b8a0dd3e1fad30d4b9d38fb5855ee72678db49831b765 [following]
---2024-07-09 14:17:02-- https://s3-eu-west-1.amazonaws.com/pstorage-cmu-348901238291901/11948267/data.db?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAI266R7V6O36O5JUA/20240709/eu-west-1/s3/aws4_request&X-Amz-Date=20240709T141702Z&X-Amz-Expires=10&X-Amz-SignedHeaders=host&X-Amz-Signature=9994bf27457b7fdedb9b8a0dd3e1fad30d4b9d38fb5855ee72678db49831b765
-Resolving s3-eu-west-1.amazonaws.com (s3-eu-west-1.amazonaws.com)...
-
-
-52.218.46.88, 52.92.1.24, 52.218.60.227, ...
-Connecting to s3-eu-west-1.amazonaws.com (s3-eu-west-1.amazonaws.com)|52.218.46.88|:443...
+Location: https://s3-eu-west-1.amazonaws.com/pstorage-cmu-348901238291901/11948267/data.db?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAI266R7V6O36O5JUA/20240710/eu-west-1/s3/aws4_request&X-Amz-Date=20240710T163350Z&X-Amz-Expires=10&X-Amz-SignedHeaders=host&X-Amz-Signature=68a7ba32898abaeaf1242dae7185303908cff55cedeb4576e44d30011f62dd34 [following]
+--2024-07-10 16:33:50-- https://s3-eu-west-1.amazonaws.com/pstorage-cmu-348901238291901/11948267/data.db?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAI266R7V6O36O5JUA/20240710/eu-west-1/s3/aws4_request&X-Amz-Date=20240710T163350Z&X-Amz-Expires=10&X-Amz-SignedHeaders=host&X-Amz-Signature=68a7ba32898abaeaf1242dae7185303908cff55cedeb4576e44d30011f62dd34
+Resolving s3-eu-west-1.amazonaws.com (s3-eu-west-1.amazonaws.com)... 52.218.98.147, 52.218.28.19, 52.92.32.104, ...
+Connecting to s3-eu-west-1.amazonaws.com (s3-eu-west-1.amazonaws.com)|52.218.98.147|:443...
-data.db 0%[ ] 58.38K 247KB/s
-
-
-data.db 0%[ ] 169.38K 358KB/s
-
-
-data.db 1%[ ] 636.38K 897KB/s
+data.db 0%[ ] 160.38K 759KB/s
-data.db 6%[> ] 2.48M 2.61MB/s
+data.db 3%[ ] 1.37M 3.23MB/s
-data.db 18%[==> ] 7.57M 6.39MB/s
+data.db 20%[===> ] 8.48M 13.4MB/s
-data.db 32%[=====> ] 13.32M 9.37MB/s
+data.db 41%[=======> ] 17.21M 20.4MB/s
-data.db 46%[========> ] 19.05M 11.5MB/s
+data.db 63%[===========> ] 25.95M 24.6MB/s
-data.db 60%[===========> ] 24.90M 13.1MB/s
+data.db 84%[===============> ] 34.60M 27.2MB/s
-data.db 74%[=============> ] 30.63M 14.3MB/s
-
-
-data.db 88%[================> ] 36.33M 15.3MB/s
-
-
-data.db 100%[===================>] 41.13M 16.4MB/s in 2.5s
+data.db 100%[===================>] 41.13M 29.0MB/s in 1.4s
-2024-07-09 14:17:05 (16.4 MB/s) - ‘data.db’ saved [43125760/43125760]
+2024-07-10 16:33:52 (29.0 MB/s) - ‘data.db’ saved [43125760/43125760]
@@ -758,29 +743,29 @@ Fast batched inference
-id|age|user |formula|calculator| energy|natoms| fmax|pbc| volume|charge| mass
- 1| 0s|runner|Au37 |unknown | -96.129| 37|0.159|TTT|5771.976| 0.000|7287.763
- 2| 0s|runner|Au28 |unknown | -65.693| 28|3.139|TTT|5068.935| 0.000|5515.064
- 3| 0s|runner|Au29 |unknown | -71.706| 29|0.288|TTT|5184.740| 0.000|5712.031
- 4| 0s|runner|Au8 |unknown | -16.751| 8|0.281|TTT|2863.348| 0.000|1575.733
- 5| 0s|runner|Au16 |unknown | -37.166| 16|0.131|TTT|3923.762| 0.000|3151.465
- 6| 0s|runner|Au4 |unknown | -11.599| 4|1.869|TTT| 156.940| 0.000| 787.866
- 7| 0s|runner|Au10 |unknown | -21.713| 10|0.701|TTT|4112.512| 0.000|1969.666
- 8| 0s|runner|Au4 |unknown | -11.902| 4|0.302|TTT| 156.940| 0.000| 787.866
- 9| 0s|runner|Au36 |unknown | -93.701| 36|0.067|TTT|5663.627| 0.000|7090.796
-10| 0s|runner|Au37 |unknown | -92.625| 37|0.254|TTT|5771.976| 0.000|7287.763
-11| 0s|runner|Au2 |unknown | -0.473| 2|0.000|TTT| 208.559| 0.000| 393.933
-12| 0s|runner|Au29 |unknown | -69.638| 29|0.554|TTT|5184.740| 0.000|5712.031
-13| 0s|runner|Au19 |unknown | -46.081| 19|0.145|TTT|4101.856| 0.000|3742.365
-14| 0s|runner|Au14 |unknown | -31.015| 14|0.418|TTT|3847.524| 0.000|2757.532
-15| 0s|runner|Au40 |unknown |-103.328| 40|0.402|TTT|5868.580| 0.000|7878.663
-16| 0s|runner|Au11 |unknown | -24.226| 11|0.079|TTT|3414.029| 0.000|2166.632
-17| 0s|runner|Au35 |unknown | -90.988| 35|0.058|TTT|5504.869| 0.000|6893.830
-18| 0s|runner|Au18 |unknown | -39.521| 18|0.900|TTT|4250.202| 0.000|3545.398
-19| 0s|runner|Au12 |unknown | -27.037| 12|0.230|TTT|3571.095| 0.000|2363.599
-20| 0s|runner|Au41 |unknown |-106.786| 41|0.272|TTT|6012.425| 0.000|8075.629
+id|age|user |formula|calculator| energy|natoms| fmax|pbc| volume|charge| mass
+ 1| 0s|runner|Au52 |unknown |-137.735| 52|0.199|TTT| 6744.510| 0.000|10242.262
+ 2| 0s|runner|Au10 |unknown | -20.920| 10|0.721|TTT| 4112.512| 0.000| 1969.666
+ 3| 0s|runner|Au4 |unknown | -12.096| 4|0.187|TTT| 143.096| 0.000| 787.866
+ 4| 0s|runner|Au5 |unknown | -9.207| 5|1.052|TTT| 2575.618| 0.000| 984.833
+ 5| 0s|runner|Au7 |unknown | -13.236| 7|0.392|TTT| 3511.808| 0.000| 1378.766
+ 6| 0s|runner|Au119 |unknown |-327.888| 119|0.621|TTT|12284.672| 0.000|23439.022
+ 7| 0s|runner|Au4 |unknown | -11.527| 4|0.424|TTT| 156.940| 0.000| 787.866
+ 8| 0s|runner|Au12 |unknown | -27.463| 12|0.151|TTT| 3066.950| 0.000| 2363.599
+ 9| 0s|runner|Au52 |unknown |-137.905| 52|0.158|TTT| 6744.510| 0.000|10242.262
+10| 0s|runner|Au27 |unknown | -67.450| 27|0.407|TTT| 4914.709| 0.000| 5318.097
+11| 0s|runner|Au4 |unknown | -11.872| 4|0.574|TTT| 156.940| 0.000| 787.866
+12| 0s|runner|Au36 |unknown | -93.692| 36|0.121|TTT| 5663.627| 0.000| 7090.796
+13| 0s|runner|Au45 |unknown |-117.721| 45|0.155|TTT| 6185.400| 0.000| 8863.496
+14| 0s|runner|Au16 |unknown | -35.355| 16|1.151|TTT| 3923.762| 0.000| 3151.465
+15| 0s|runner|Au48 |unknown |-124.508| 48|0.377|TTT| 6440.631| 0.000| 9454.395
+16| 0s|runner|Au41 |unknown |-103.140| 41|0.838|TTT| 6012.425| 0.000| 8075.629
+17| 0s|runner|Au13 |unknown | -29.197| 13|0.109|TTT| 3635.026| 0.000| 2560.565
+18| 0s|runner|Au17 |unknown | -39.838| 17|0.058|TTT| 4121.531| 0.000| 3348.432
+19| 0s|runner|Au28 |unknown | -86.927| 28|0.019|TTT| 526.183| 0.000| 5515.064
+20| 0s|runner|Au16 |unknown | -37.467| 16|0.323|TTT| 3923.762| 0.000| 3151.465
Rows: 50 (showing first 20)
-Keys: NEB, bulk, calc_time, cluster, config, diffusion, ediff, ediffg, encut, factor, fermi, gga, group, image, kpt1, kpt2, kpt3, miller, neural_energy, reax_energy, strain, structure, surf, train_set, traj, vacuum, volume, xc
+Keys: NEB, bulk, calc_time, cluster, concentration, config, diffusion, ediff, ediffg, encut, fermi, gga, group, image, kpt1, kpt2, kpt3, lattice, miller, neural_energy, reax_energy, site, structure, surf, train_set, traj, type, vacuum, volume, xc
@@ -881,7 +866,7 @@ Fast batched inference
-2024-07-09 14:17:28 (INFO): Total time taken: 6.688116550445557
+2024-07-10 16:34:13 (INFO): Total time taken: 8.430647373199463
@@ -950,7 +935,7 @@ Fast batched inference
-
+
@@ -999,104 +984,101 @@ The ASE calculator way0it [00:00, ?it/s]
-/tmp/ipykernel_2741/402181013.py:7: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2603/402181013.py:7: DeprecationWarning: Please use atoms.calc = calc
atoms.set_calculator(calc)
-1it [00:00, 4.34it/s]
-
-
-2it [00:00, 5.39it/s]
+1it [00:00, 2.79it/s]
-3it [00:00, 5.67it/s]
+4it [00:01, 2.71it/s]
-5it [00:00, 8.90it/s]
+6it [00:01, 3.36it/s]
-7it [00:00, 8.30it/s]
+7it [00:02, 3.77it/s]
-8it [00:01, 7.05it/s]
+8it [00:02, 3.93it/s]
-9it [00:01, 6.84it/s]
+9it [00:02, 3.77it/s]
-10it [00:01, 7.38it/s]
+11it [00:02, 4.29it/s]
-12it [00:01, 6.92it/s]
+12it [00:03, 4.21it/s]
-14it [00:01, 7.17it/s]
+14it [00:03, 5.78it/s]
-16it [00:02, 8.57it/s]
+15it [00:03, 4.68it/s]
-17it [00:02, 6.91it/s]
+17it [00:03, 5.72it/s]
-18it [00:02, 6.91it/s]
+18it [00:04, 5.75it/s]
-21it [00:02, 9.55it/s]
+19it [00:04, 5.51it/s]
-23it [00:03, 7.96it/s]
+20it [00:04, 4.42it/s]
-25it [00:03, 8.24it/s]
+21it [00:04, 4.47it/s]
-27it [00:03, 7.53it/s]
+22it [00:05, 4.10it/s]
-28it [00:03, 6.57it/s]
+23it [00:05, 3.83it/s]
-29it [00:03, 6.74it/s]
+24it [00:05, 3.51it/s]
-30it [00:04, 6.10it/s]
+25it [00:05, 4.01it/s]
-31it [00:04, 3.48it/s]
+26it [00:06, 4.01it/s]
-32it [00:05, 3.31it/s]
+27it [00:06, 4.05it/s]
-33it [00:05, 3.89it/s]
+29it [00:06, 6.21it/s]
-34it [00:05, 4.21it/s]
+30it [00:06, 5.13it/s]
-35it [00:05, 5.03it/s]
+32it [00:07, 5.42it/s]
-36it [00:05, 5.10it/s]
+34it [00:07, 5.38it/s]
-37it [00:06, 3.18it/s]
+35it [00:07, 5.20it/s]
-38it [00:06, 3.06it/s]
+37it [00:08, 4.93it/s]
-39it [00:06, 3.68it/s]
+40it [00:08, 7.21it/s]
-41it [00:07, 5.73it/s]
+41it [00:08, 6.77it/s]
-41it [00:07, 5.82it/s]
+41it [00:08, 4.74it/s]
-Elapsed time 7e+00 seconds
+Elapsed time 9e+00 seconds
@@ -1130,7 +1112,7 @@ Comparing ASE calculator and main.py
-52.66483077912157
+62.14755616196559
@@ -1142,7 +1124,7 @@ Comparing ASE calculator and main.py
-(-236.94300270080566, 236.9430046081543)
+(-328.9178466796875, 328.9178466796875)
@@ -1154,7 +1136,7 @@ Comparing ASE calculator and main.py
-
+
Here we see many of the differences are very small. 0.0078125 = 1 / 128, and these errors strongly suggest some kind of mixed precision is responsible for these differences. It is an open issue to remove them and identify where the cause is.
@@ -1165,19 +1147,19 @@ Comparing ASE calculator and main.py
-array([[ 0.00000000e+00, -3.04521866e+01, -2.49373627e+01, ...,
- -3.49432831e+01, -8.12342224e+01, -6.72080231e+01],
- [ 3.04521866e+01, 0.00000000e+00, 5.51482391e+00, ...,
- -4.49109650e+00, -5.07820358e+01, -3.67558365e+01],
- [ 2.49373627e+01, -5.51482391e+00, 0.00000000e+00, ...,
- -1.00059204e+01, -5.62968597e+01, -4.22706604e+01],
+array([[ 0. , -123.06163406, -131.50551605, ..., -127.80545235,
+ -95.427742 , -58.97216797],
+ [ 123.06163406, 0. , -8.44388199, ..., -4.74381828,
+ 27.63389206, 64.08946609],
+ [ 131.50551605, 8.44388199, 0. , ..., 3.70006371,
+ 36.07777405, 72.53334808],
...,
- [ 3.49432831e+01, 4.49109650e+00, 1.00059204e+01, ...,
- 0.00000000e+00, -4.62909393e+01, -3.22647400e+01],
- [ 8.12342224e+01, 5.07820358e+01, 5.62968597e+01, ...,
- 4.62909393e+01, 0.00000000e+00, 1.40261993e+01],
- [ 6.72080154e+01, 3.67558289e+01, 4.22706528e+01, ...,
- 3.22647324e+01, -1.40262070e+01, -7.62939453e-06]])
+ [ 127.80545235, 4.74381828, -3.70006371, ..., 0. ,
+ 32.37771034, 68.83328438],
+ [ 95.427742 , -27.63389206, -36.07777405, ..., -32.37771034,
+ 0. , 36.45557404],
+ [ 58.97216797, -64.08946609, -72.53334808, ..., -68.83328438,
+ -36.45557404, 0. ]])
diff --git a/core/lmdb_dataset_creation.html b/core/lmdb_dataset_creation.html
index d23cb3c6e..890f113b9 100644
--- a/core/lmdb_dataset_creation.html
+++ b/core/lmdb_dataset_creation.html
@@ -687,7 +687,7 @@ Generate toy dataset: Relaxation of CO on Cu
-/tmp/ipykernel_2828/901556023.py:8: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2689/901556023.py:8: DeprecationWarning: Please use atoms.calc = calc
adslab.set_calculator(EMT())
@@ -902,46 +902,49 @@ Structure to Energy and Forces (S2EF) LMDBs 0%| | 0/1001 [00:00<?, ?it/s]
- 7%|▋ | 74/1001 [00:00<00:01, 738.46it/s]
+ 6%|▌ | 59/1001 [00:00<00:01, 582.40it/s]
- 15%|█▌ | 152/1001 [00:00<00:01, 755.68it/s]
+ 13%|█▎ | 131/1001 [00:00<00:01, 657.50it/s]
- 23%|██▎ | 229/1001 [00:00<00:01, 759.99it/s]
+ 20%|██ | 202/1001 [00:00<00:01, 678.46it/s]
- 30%|███ | 305/1001 [00:00<00:00, 743.81it/s]
+ 27%|██▋ | 271/1001 [00:00<00:01, 678.63it/s]
- 38%|███▊ | 380/1001 [00:00<00:00, 742.81it/s]
+ 34%|███▍ | 342/1001 [00:00<00:00, 687.75it/s]
- 45%|████▌ | 455/1001 [00:00<00:00, 742.96it/s]
+ 41%|████▏ | 415/1001 [00:00<00:00, 701.72it/s]
- 53%|█████▎ | 530/1001 [00:00<00:00, 744.99it/s]
+ 49%|████▉ | 491/1001 [00:00<00:00, 717.42it/s]
- 61%|██████ | 607/1001 [00:00<00:00, 752.60it/s]
+ 56%|█████▌ | 563/1001 [00:00<00:00, 713.63it/s]
- 68%|██████▊ | 683/1001 [00:00<00:00, 747.17it/s]
+ 63%|██████▎ | 635/1001 [00:00<00:00, 711.60it/s]
- 76%|███████▌ | 758/1001 [00:01<00:00, 738.12it/s]
+ 71%|███████ | 707/1001 [00:01<00:00, 704.80it/s]
- 83%|████████▎ | 832/1001 [00:01<00:00, 738.66it/s]
+ 78%|███████▊ | 779/1001 [00:01<00:00, 708.09it/s]
- 91%|█████████ | 906/1001 [00:01<00:00, 722.54it/s]
+ 85%|████████▌ | 851/1001 [00:01<00:00, 711.56it/s]
- 98%|█████████▊| 979/1001 [00:01<00:00, 721.01it/s]
+ 92%|█████████▏| 923/1001 [00:01<00:00, 707.86it/s]
-100%|██████████| 1001/1001 [00:01<00:00, 737.14it/s]
+100%|█████████▉| 996/1001 [00:01<00:00, 713.40it/s]
+
+
+100%|██████████| 1001/1001 [00:01<00:00, 700.98it/s]
diff --git a/core/ocpapi.html b/core/ocpapi.html
index 59ed19d24..e9e9205d8 100644
--- a/core/ocpapi.html
+++ b/core/ocpapi.html
@@ -664,7 +664,7 @@ InstallationERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
-fairchem-demo-ocpapi 0.1.dev1+gbf2b26e requires requests==2.32.0, but you have requests 2.31.0 which is incompatible.
+fairchem-demo-ocpapi 0.1.dev1+g36a17ad requires requests==2.32.0, but you have requests 2.31.0 which is incompatible.
pymatgen 2024.6.10 requires requests>=2.32, but you have requests 2.31.0 which is incompatible.
diff --git a/core/quickstart.html b/core/quickstart.html
index 61418213d..d2a5c4bf3 100644
--- a/core/quickstart.html
+++ b/core/quickstart.html
@@ -701,60 +701,60 @@ Using pre-trained models in ASEWARNING:root:No seed has been set in modelcheckpoint or OCPCalculator! Results may not be reproducible on re-run
-/tmp/ipykernel_2885/3462928393.py:13: DeprecationWarning: Please use atoms.calc = calc
+/tmp/ipykernel_2745/3462928393.py:13: DeprecationWarning: Please use atoms.calc = calc
slab.set_calculator(calc)
Step Time Energy fmax
-BFGS: 0 14:18:03 -110.398659 1.630680
+BFGS: 0 16:34:50 -110.398659 1.630680
-BFGS: 1 14:18:03 -110.545036 0.925020
+BFGS: 1 16:34:50 -110.545036 0.925020
-BFGS: 2 14:18:03 -110.622719 0.562895
+BFGS: 2 16:34:50 -110.622719 0.562895
-BFGS: 3 14:18:04 -110.613007 0.583018
+BFGS: 3 16:34:51 -110.613007 0.583018
-BFGS: 4 14:18:04 -110.641937 0.461219
+BFGS: 4 16:34:51 -110.641937 0.461219
-BFGS: 5 14:18:05 -110.691620 0.362131
+BFGS: 5 16:34:52 -110.691620 0.362131
-BFGS: 6 14:18:05 -110.723862 0.592763
+BFGS: 6 16:34:52 -110.723862 0.592763
-BFGS: 7 14:18:05 -110.764206 0.712022
+BFGS: 7 16:34:52 -110.764206 0.712022
-BFGS: 8 14:18:06 -110.869545 0.522792
+BFGS: 8 16:34:53 -110.869545 0.522792
-BFGS: 9 14:18:06 -110.938713 0.170553
+BFGS: 9 16:34:53 -110.938713 0.170553
-BFGS: 10 14:18:07 -110.977699 0.083886
+BFGS: 10 16:34:53 -110.977699 0.083886
-BFGS: 11 14:18:07 -110.992668 0.071652
+