https://github.com/kjczarne/shap_adapool
Adaptive Pooling Strategies for Shapley Value Explanations (SHAP)
https://github.com/kjczarne/shap_adapool
Last synced: 3 months ago
JSON representation
Adaptive Pooling Strategies for Shapley Value Explanations (SHAP)
- Host: GitHub
- URL: https://github.com/kjczarne/shap_adapool
- Owner: kjczarne
- Created: 2024-01-29T21:12:56.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-06-25T12:10:30.000Z (12 months ago)
- Last Synced: 2025-01-23T04:11:52.333Z (5 months ago)
- Language: Python
- Size: 327 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# SHAP Adaptive Pooling
This project explores the possibilities of Shapley Value pooling strategies. Since Shapley Values are additive in nature, it is possible to combine input tokens into less granular phrases and sum up the corresponding Shapley Values to obtain phrase-level explanations. Of course the semantic definition of a _phrase_ is loosely defined as a grouping of words that have some syntactic function in a given context.
We explore different strategies of pooling Shapley Values into:
- [x] sentences
- [ ] k-word phrases
- [/] language-syntax-tree-defined phrases
- [ ] adaptively-defined phrases
## Installation
This project uses the [Poetry Package Manager](https://python-poetry.org/) and the recommended way to install the project is to:
1. Build the package: `poetry build`
2. Find the wheel in the `dist` folder and install the wheel with `pip`: `pip install `
3. Because some methods here rely on SpaCy providing the dependency tree, you will need to download the SpaCy pipeline for the English language: `python -m spacy download en_core_web_sm`
> [!warning]
> At least Python 3.10 is required for this package to work. We profusely use functional programming concepts such as structural pattern matching and some of these facilities are only available in Python 3.10 and newer.
> [!warning]
> If you're developing the project, install with `poetry install` instead.
## Usage
### Sentence Pooling
The most naive pipeline is taking all the generated Shapley Values and pooling them together sentence-by-sentence. The pipeline with the provided example file `shap.pkl` can be run as follows:
```bash
python -m shap_adapool.pooling_strategies.sentence_pooling
```
### k-Word Pooling
TODO
### LST Pooling
Language Syntax Tree Pooling relies on the traversal of a syntax tree and formation of phrases defined on that tree:
```bash
python -m shap_adapool.pooling_strategies.syntax_tree_pooling
```
> [!note]
> For now the module only prints out a syntax tree for a predefined sentence. We will soon implement the pooling strategy fully.
### Adaptive Pooling
TODO
## Development
For development:
1. Clone this repo.
2. Install the repo using `poetry install`. This will install the package in editable mode.