Merge branch 'doc_example' into doc

docs/docDIAL.md

@@ -73,15 +73,14 @@ This procedure yields to the following availability of hyperparameter:
 - `--gamma_reg`: ? ($\epsilon$ in the paper)
 - `--lr`: learning rate ($\alpha$ in the paper)
 
-# Examples
+## Examples
 
 ```
 python main_out.py --te_d=0 --task=mnistcolor10 --model=erm --trainer=dial --nname=conv_bn_pool_2
 ```
 
 
 
-## Adversarial images training
 ```shell
 python main_out.py --te_d=0 --task=mnistcolor10 --keep_model --model=erm --trainer=dial --nname=conv_bn_pool_2
 ```

docs/docFishr.md

@@ -72,7 +72,7 @@ For more details, see the reference below or the domainlab code.
 
 
 
-# Examples
+## Examples
 ```
 python main_out.py --te_d=0 --task=mini_vlcs --model=erm --trainer=fishr --nname=alexnet --bs=2 --nocu
 ```

docs/docIRM.md

@@ -26,7 +26,7 @@ where $\lambda$ is a hyperparameter that controls the trade-off between the empi
 In practice, one could simply divide one mini-batch into two subsets, let $i$ and $j$ to index these two subsets, multiply  subset $i$ and subset $j$ forms an unbiased estimation of the L2 norm of gradient.
 In detail: the squared gradient norm via inner product between $\nabla_{w|w=1} \ell(w \circ \Phi(X^{(d, i)}), Y^{(d, i)})$ 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
 ```shell
 python main_out.py --te_d=0 --task=mnistcolor10 --model=erm --trainer=irm --nname=conv_bn_pool_2
 

docs/doc_mldg.md

@@ -4,8 +4,7 @@
 Li, Da, et al. "Learning to generalize: Meta-learning for domain generalization." Proceedings of the AAAI conference on artificial intelligence. Vol. 32. No. 1. 2018.
 
 
-# Examples
-## Meta Learning Domain Generalization
+## Examples
 ```shell
 python main_out.py --te_d=caltech --task=mini_vlcs --debug --bs=8 --model=erm --trainer=mldg --nname=alexnet
 ```