Metadata-Version: 2.2
Name: ir_explain
Version: 0.1
Summary: A Python Library for Explainable IR methods
Home-page: https://github.com/souravsaha/ir_explain/
Author: Sourav Saha
Author-email: souravsaha.juit@gmail.com
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: ipython==8.12.3
Requires-Dist: matplotlib==3.9.4
Requires-Dist: nltk==3.9.1
Requires-Dist: numpy==1.24.4
Requires-Dist: pandas==2.2.3
Requires-Dist: rank_bm25==0.2.2
Requires-Dist: scikit_learn==1.6.0
Requires-Dist: scipy==1.10.1
Requires-Dist: sentence_transformers==3.3.1
Requires-Dist: scikit-image
Requires-Dist: torch==2.2.0
Requires-Dist: ir_datasets
Requires-Dist: tqdm==4.67.1
Requires-Dist: pyserini==0.21.0
Requires-Dist: gensim==4.3.1
Requires-Dist: torchtext==0.17.0
Requires-Dist: h5py
Requires-Dist: captum==0.7.0
Requires-Dist: genosolver==0.1.0.6
Requires-Dist: cvxpy==1.3.2
Requires-Dist: pytorch-lightning
Requires-Dist: faiss-cpu==1.8.0
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# ir_explain

## Installing requirements

```
git clone https://github.com/souravsaha/ir_explain.git
```
```
conda create -n ir_explain python=3.9
```
```
pip install .
```
Note that for now we have tested ir_explain on Python 3.9 and Java 11. we will test on newer versions of Python soon and update.

Before running the ir_explain library, set the JAVA HOME as follows (replace the path appropriately):
```
export JAVA_HOME="/usr/lib/jvm/java-11-openjdk-amd64/"
```

## Usage

As of now, we have established a single pipeline with pointwise component. 

### To get started 

You can run the programs test_point_wise.py, test_pair_wise.py, and test_list_wise.py to get the pipeline. 

## Contributing

IR Explain library is open-source, and we are open to all the contributions from IR and NLP community. If you find a bug please report to the issue tracker, even better to send us a pull-request on Github. 
