Added LayerDeepLiftShap and LayerGradientShap

Merge pull request #16 from aj280192/main
Final pull request for release 1.0.2

README.md

@@ -144,6 +144,9 @@ Layer Integrated Gradients (`lig`) | [`.attr.LayerIntegratedGradients`](https://
 LIME (`lime`) | [`.attr.LimeBase`](https://captum.ai/api/lime.html) | # samples = 25, <br>mask prob = 0.3
 Occlusion (`occ`) | [`.attr.Occlusion`](https://captum.ai/api/occlusion.html) | sliding window = 3
 Shapley Value Sampling (`svs`) | [`.attr.ShapleyValueSampling`](https://captum.ai/api/shapley_value_sampling.html) | # samples = 25
+Layer DeepLiftShap (`lds`) | [`.attr.LayerDeepLiftShap`](https://captum.ai/api/layer.html#layer-deepliftshap) |
+Layer GradientShap (`lgs`) | [`.attr.LayerGradientShap`](https://captum.ai/api/layer.html#layer-gradientshap) | # samples = 5
+
 
 
 
@@ -161,53 +164,53 @@ Shapley Value Sampling (`svs`) | [`.attr.ShapleyValueSampling`](https://captum.a
 [`imdb`](https://huggingface.co/datasets/viewer/?dataset=imdb) is a sentiment analysis dataset with 2 classes (`pos` and `neg`). The available split is the `test` subset containing 25k examples.  
 Example configuration: `imdb-xlnet-lig`
 
-Name | 🤗 | `lgxa` | `lig` | `lime` | `occ` | `svs`  
---- | --- | --- | --- | --- | --- | ---
-ALBERT (`albert`) | [`textattack/albert-base-v2-imdb`](https://huggingface.co/textattack/albert-base-v2-imdb) | ✅ | ✅ | ✅ | ✅ | ✅
-BERT (`bert`) | [`textattack/bert-base-uncased-imdb`](https://huggingface.co/textattack/bert-base-uncased-imdb) | ✅ | ✅ | ✅ | ✅ | ✅
-ELECTRA (`electra`) | [`monologg/electra-small-finetuned-imdb`](https://huggingface.co/monologg/electra-small-finetuned-imdb) | ✅ | ✅ | ✅ | ✅ | ✅
-RoBERTa (`roberta`) | [`textattack/roberta-base-imdb`](https://huggingface.co/textattack/roberta-base-imdb) | ✅ | ✅ | ✅ | ✅ | ✅
-XLNet (`xlnet`) | [`textattack/xlnet-base-cased-imdb`](https://huggingface.co/textattack/xlnet-base-cased-imdb) | ✅ | ✅ | ✅ | ✅ | ✅
+Name | 🤗 | `lgxa` | `lig` | `lime` | `occ` | `svs` | `lds` | `lgs` 
+--- | --- | --- | --- | --- | --- | --- | --- | ---
+ALBERT (`albert`) | [`textattack/albert-base-v2-imdb`](https://huggingface.co/textattack/albert-base-v2-imdb) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+BERT (`bert`) | [`textattack/bert-base-uncased-imdb`](https://huggingface.co/textattack/bert-base-uncased-imdb) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+ELECTRA (`electra`) | [`monologg/electra-small-finetuned-imdb`](https://huggingface.co/monologg/electra-small-finetuned-imdb) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+RoBERTa (`roberta`) | [`textattack/roberta-base-imdb`](https://huggingface.co/textattack/roberta-base-imdb) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+XLNet (`xlnet`) | [`textattack/xlnet-base-cased-imdb`](https://huggingface.co/textattack/xlnet-base-cased-imdb) | ✅ | ✅ | ✅ | ✅ | ✅ | ⚠️ | ⚠️
 
 
 ### MultiNLI
 
 [`multi_nli`](https://huggingface.co/datasets/viewer/?dataset=multi_nli) is a textual entailment dataset. The available split is the `validation_matched` subset containing 9815 examples.  
 Example configuration: `multi_nli-roberta-lime`
 
-Name | 🤗 | `lgxa` | `lig` | `lime` | `occ` | `svs`
---- | --- | --- | --- | --- | --- | ---
-ALBERT (`albert`) | [`prajjwal1/albert-base-v2-mnli`](https://huggingface.co/prajjwal1/albert-base-v2-mnli) | ✅ | ✅ | ✅ | ✅ | ✅
-BERT (`bert`) | [`textattack/bert-base-uncased-MNLI`](https://huggingface.co/textattack/bert-base-uncased-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅
-ELECTRA (`electra`) | [`howey/electra-base-mnli`](https://huggingface.co/howey/electra-base-mnli) | ✅ | ✅ | ✅ | ✅ | ✅
-RoBERTa (`roberta`) | [`textattack/roberta-base-MNLI`](https://huggingface.co/textattack/roberta-base-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅
-XLNet (`xlnet`) | [`textattack/xlnet-base-cased-MNLI`](https://huggingface.co/textattack/xlnet-base-cased-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅
+Name | 🤗 | `lgxa` | `lig` | `lime` | `occ` | `svs` | `lds` | `lgs`
+--- | --- | --- | --- | --- | --- | --- | --- | ---
+ALBERT (`albert`) | [`prajjwal1/albert-base-v2-mnli`](https://huggingface.co/prajjwal1/albert-base-v2-mnli) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+BERT (`bert`) | [`textattack/bert-base-uncased-MNLI`](https://huggingface.co/textattack/bert-base-uncased-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+ELECTRA (`electra`) | [`howey/electra-base-mnli`](https://huggingface.co/howey/electra-base-mnli) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+RoBERTa (`roberta`) | [`textattack/roberta-base-MNLI`](https://huggingface.co/textattack/roberta-base-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+XLNet (`xlnet`) | [`textattack/xlnet-base-cased-MNLI`](https://huggingface.co/textattack/xlnet-base-cased-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅ | ⚠️ | ⚠️
 
 
 ### XNLI
 
 [`xnli`](https://huggingface.co/datasets/viewer/?dataset=xnli) is a textual entailment dataset. It provides the test set of MultiNLI through the "en" configuration. The fine-tuned models used here are the same as the MultiNLI ones. The available split is the `test` subset containing 5010 examples.  
 Example configuration: `xnli-roberta-lime`
 
-Name | 🤗 | `lgxa` | `lig` | `lime` | `occ` | `svs`
---- | --- | --- | --- | --- | --- | ---
-ALBERT (`albert`) | [`prajjwal1/albert-base-v2-mnli`](https://huggingface.co/prajjwal1/albert-base-v2-mnli) | ✅ | ✅ | ✅ | ✅ | ✅
-BERT (`bert`) | [`textattack/bert-base-uncased-MNLI`](https://huggingface.co/textattack/bert-base-uncased-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅
-ELECTRA (`electra`) | [`howey/electra-base-mnli`](https://huggingface.co/howey/electra-base-mnli) | ✅ | ✅ | ✅ | ✅ | ✅
-RoBERTa (`roberta`) | [`textattack/roberta-base-MNLI`](https://huggingface.co/textattack/roberta-base-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅
-XLNet (`xlnet`) | [`textattack/xlnet-base-cased-MNLI`](https://huggingface.co/textattack/xlnet-base-cased-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅
+Name | 🤗 | `lgxa` | `lig` | `lime` | `occ` | `svs` | `lds` | `lgs`
+--- | --- | --- | --- | --- | --- | --- | --- | ---
+ALBERT (`albert`) | [`prajjwal1/albert-base-v2-mnli`](https://huggingface.co/prajjwal1/albert-base-v2-mnli) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+BERT (`bert`) | [`textattack/bert-base-uncased-MNLI`](https://huggingface.co/textattack/bert-base-uncased-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+ELECTRA (`electra`) | [`howey/electra-base-mnli`](https://huggingface.co/howey/electra-base-mnli) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+RoBERTa (`roberta`) | [`textattack/roberta-base-MNLI`](https://huggingface.co/textattack/roberta-base-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+XLNet (`xlnet`) | [`textattack/xlnet-base-cased-MNLI`](https://huggingface.co/textattack/xlnet-base-cased-MNLI) | ✅ | ✅ | ✅ | ✅ | ✅ | ⚠️ | ⚠️
 
 
 ### AG News
 
 [`ag_news`](https://huggingface.co/datasets/viewer/?dataset=ag_news) is a news topic classification dataset. The available split is the `test` subset containing 7600 examples.  
 Example configuration: `ag_news-albert-svs`
 
-Name | 🤗 | `lgxa` | `lig` | `lime` | `occ` | `svs`
---- | --- | --- | --- | --- | --- | ---
-ALBERT (`albert`) | [`textattack/albert-base-v2-ag-news`](https://huggingface.co/textattack/albert-base-v2-ag-news) | ✅ | ✅ | ✅ | ✅ | ✅
-BERT (`bert`) | [`textattack/bert-base-uncased-ag-news`](https://huggingface.co/textattack/bert-base-uncased-ag-news) | ✅ | ✅ | ✅ | ✅ | ✅
-RoBERTa (`roberta`) | [`textattack/roberta-base-ag-news`](https://huggingface.co/textattack/roberta-base-ag-news) | ✅ | ✅ | ✅ | ✅ | ✅
+Name | 🤗 | `lgxa` | `lig` | `lime` | `occ` | `svs` | `lds` | `lgs`
+--- | --- | --- | --- | --- | --- | --- | --- | ---
+ALBERT (`albert`) | [`textattack/albert-base-v2-ag-news`](https://huggingface.co/textattack/albert-base-v2-ag-news) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+BERT (`bert`) | [`textattack/bert-base-uncased-ag-news`](https://huggingface.co/textattack/bert-base-uncased-ag-news) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
+RoBERTa (`roberta`) | [`textattack/roberta-base-ag-news`](https://huggingface.co/textattack/roberta-base-ag-news) | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅
 
 ---