Merge pull request #823 from dianna-ai/756-add-notebook-for-eulaw

add eulaw with lime notebook (refs #756)
dianna-ai · Oct 11, 2024 · 116ffeb · 116ffeb
2 parents f3cda01 + e4980d8
commit 116ffeb
Show file tree

Hide file tree

Showing 6 changed files with 1,094 additions and 34 deletions.
diff --git a/README.md b/README.md
@@ -200,7 +200,7 @@ Explore the explanations of your trained model using the DIANNA dashboard (for n
 _Dianna dashboard screenshot here_
 <!-- <a href="https://github.com/dianna-ai/dianna/tree/main/dianna/dashboard" target="_blank">
    <img width="1000" align="center" alt="Dianna dashboard screenshot" src="https://raw.githubusercontent.com/dianna-ai/dianna/main/dianna/dashboard/dashboard-screenshot.png">
-</a> 
+</a>
 -->
 
 ## Datasets
@@ -260,9 +260,10 @@ And here are links to notebooks showing how we created our models on the benchma
 
 ### Text
 
-| Models                                                | Generation                                                                                                                                                              |
-| :---------------------------------------------------- | :---------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| [Movie reviews model](https://zenodo.org/record/5910598) | [Stanford sentiment treebank model generation](https://github.com/dianna-ai/dianna-exploration/blob/main/example_data/model_generation/movie_reviews/generate_model.ipynb) |
+| Models                                                               | Generation                                                                                                                                                                 |
+|:---------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| [Movie reviews model](https://zenodo.org/record/5910598)             | [Stanford sentiment treebank model generation](https://github.com/dianna-ai/dianna-exploration/blob/main/example_data/model_generation/movie_reviews/generate_model.ipynb) |
+| [Regalatory statement classifier](https://zenodo.org/record/8200001) | [EU-law regulatory-statement-classification](https://github.com/nature-of-eu-rules/regulatory-statement-classification)                                                    |
 
 ### Time series
 
@@ -306,13 +307,13 @@ Also [GradCAM](https://openaccess.thecvf.com/content_ICCV_2017/papers/Selvaraju_
 Our goal is that the scientific community embrases XAI as a source for novel and unexplored perspectives on scientific problems.
 Here, we offer [tutorials](./tutorials) on specific scientific use-cases of uisng XAI:
 
-| Use-case (data) \ XAI                                            | [RISE](http://bmvc2018.org/contents/papers/1064.pdf) | [LIME](https://www.kdd.org/kdd2016/papers/files/rfp0573-ribeiroA.pdf) | [KernelSHAP](https://proceedings.neurips.cc/paper/2017/file/8a20a8621978632d76c43dfd28b67767-Paper.pdf) |
-| :----------------------------------------------------------------| :----------------------------------------------------| :---------------------------------------------------------------------| :-------------------------------------------------------------------------------------------------------|
-| Biology (Phytomorphology): Tree Leaves classification (images)   |                                                      |            ✅                                                        |                                                                                                         |
-| Astronomy: Fast Radio Burst detection (timeseries)               | ✅                                                  |                                                                       |                                                                                                         |
-| Land-atmosphere modeling: Latent heat flux prediction (tabular)                                        |                                               |                                                                    | ✅                                                                                                     |
-| Social sciences (text)                                           | work in progress                                     |  ...                                                                  |...                                                                                                     |
-| Climate                                                          | planned                                              |   ...                                                                 |       ...                                                                                              |
+| Use-case (data) \ XAI                                              | [RISE](http://bmvc2018.org/contents/papers/1064.pdf) | [LIME](https://www.kdd.org/kdd2016/papers/files/rfp0573-ribeiroA.pdf) | [KernelSHAP](https://proceedings.neurips.cc/paper/2017/file/8a20a8621978632d76c43dfd28b67767-Paper.pdf) |
+|:-------------------------------------------------------------------|:-----------------------------------------------------| :---------------------------------------------------------------------| :-------------------------------------------------------------------------------------------------------|
+| Biology (Phytomorphology): Tree Leaves classification (images)     |                                                      |            ✅                                                        |                                                                                                         |
+| Astronomy: Fast Radio Burst detection (timeseries)                 | ✅                                                    |                                                                       |                                                                                                         |
+| Land-atmosphere modeling: Latent heat flux prediction (tabular)    |                                                      |                                                                    | ✅                                                                                                     |
+| Social sciences: EU-law regulatory statement classification (text) |                                                      |  ✅                                                                  |                                                                                                     |
+| Climate                                                            | planned                                              |   ...                                                                 |       ...                                                                                              |
 
 ## Reference documentation
 

diff --git a/dianna/utils/downloader.py b/dianna/utils/downloader.py
@@ -59,6 +59,10 @@
         "doi:10.5281/zenodo.10656613/apertif_frb_dynamic_spectrum_model.onnx",
         "sha256:3c87db3c6257d7f251a7bdceb3197d5bb482b8edc19870219fb7ca7c204dd257"
     ],
+    "inlegal_bert_xgboost_classifier.json": [
+        "doi:10.5281/zenodo.8200001/inlegal_bert_xgboost_classifier.json",
+        "68a672f29aac4a19c404c24f4c5da82a1ce7f704ccce701b0a1073c63730e127"
+    ],
     "stemmus_scope_emulator_model_LEtot.onnx": [
         "doi:10.5281/zenodo.12623256/stemmus_scope_emulator_model_LEtot.onnx",
         "sha256:8c8f34ad5a2c519b1f3c67a4eb0c645c96cac1de166277bfb24e7887c2ce83be"

diff --git a/docs/tutorials/9-lime_text_eulaw.nblink b/docs/tutorials/9-lime_text_eulaw.nblink
@@ -0,0 +1,3 @@
+{
+    "path": "../../tutorials/explainers/LIME/lime_text_eulaw.ipynb"
+}
diff --git a/setup.cfg b/setup.cfg
@@ -112,6 +112,9 @@ notebooks =
     torch
     torchvision
     ipywidgets
+    freetype-py
+    transformers
+    xgboost
 
 [options.entry_points]
 console_scripts =