https://github.com/SelfExplainML/Aletheia
A Python package for unwrapping ReLU DNNs
https://github.com/SelfExplainML/Aletheia
deep-neural-networks explainable-ai local-interpretation local-linear-models self-explanatory-ml
Last synced: 20 days ago
JSON representation
A Python package for unwrapping ReLU DNNs
- Host: GitHub
- URL: https://github.com/SelfExplainML/Aletheia
- Owner: SelfExplainML
- Created: 2020-11-15T10:15:48.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2024-01-19T11:18:13.000Z (about 2 years ago)
- Last Synced: 2024-04-19T17:00:01.317Z (almost 2 years ago)
- Topics: deep-neural-networks, explainable-ai, local-interpretation, local-linear-models, self-explanatory-ml
- Homepage:
- Size: 68.8 MB
- Stars: 68
- Watchers: 7
- Forks: 20
- Open Issues: 6
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-machine-learning-interpretability - Aletheia -  | "A Python package for unwrapping ReLU DNNs.” | (Technical Resources / Open Source/Access Responsible AI Software Packages)
README
# Aletheia
A Python package for unwrapping ReLU Neural Networks
## Installation
The following environments are required:
- Python 3.6, 3.7, 3.8, 3.9 (Try Google Colab)
- matplotlib>=3.1.3
- numpy>=1.17
- pandas>=1.1.2
- seaborn>=0.9.0
- scikit-learn>=0.23.0
- statsmodels>=0.12.2
```shell
pip install aletheia-dnn
```
## Usage
**Load data**
```python
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.datasets import make_circles
from sklearn.model_selection import train_test_split
random_state = 0
x, y = make_circles(n_samples=2000, noise=0.1, random_state=random_state)
train_x, test_x, train_y, test_y = train_test_split(x, y, test_size=0.2, random_state=random_state)
plt.figure(figsize=(10,8))
scatter = plt.scatter(x[:, 0], x[:, 1], c=y)
plt.legend(*scatter.legend_elements(), loc="upper right")
plt.show()
```
**Train a ReLU Net**
```python
from sklearn.neural_network import MLPClassifier
mlp = MLPClassifier(hidden_layer_sizes=[40] * 4, max_iter=2000, early_stopping=True,
n_iter_no_change=100, validation_fraction=0.2,
solver='adam', activation="relu", random_state=random_state,
learning_rate_init=0.001)
mlp.fit(train_x, train_y)
```
**UnwrapperClassifier**
```python
from aletheia import *
clf = UnwrapperClassifier(mlp.coefs_, mlp.intercepts_)
clf.fit(train_x, train_y)
clf.summary()
```
Partitioned regions
```python
clf.visualize2D_regions(figsize=(8, 8), meshsize=300, show_label=False)
```
**Simplification**
```python
from sklearn.metrics import make_scorer, roc_auc_score
from sklearn.model_selection import GridSearchCV, PredefinedSplit
from sklearn.linear_model import LogisticRegressionCV, LogisticRegression
datanum = train_x.shape[0]
indices = np.arange(datanum)
idx1, idx2 = train_test_split(indices, test_size=0.2, random_state=random_state)
val_fold = np.ones((len(indices)))
val_fold[idx1] = -1
grid = GridSearchCV(MergerClassifier(unwrapper=None,
weights=mlp.coefs_,
biases=mlp.intercepts_,
min_samples=30,
n_neighbors=np.round(clf.nllms * 0.01).astype(int),
refit_model=LogisticRegression()),
param_grid={"n_clusters": [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20]},
scoring={"auc": make_scorer(roc_auc_score, needs_proba=True)},
cv=PredefinedSplit(val_fold), refit="auc", n_jobs=10, error_score=np.nan)
grid.fit(train_x, train_y)
clf_merge = grid.best_estimator_
clf_merge.summary()
```
**Local Inference**
```python
tmpid = 0
clf_merge.visualize2D_one_line(tmpid, figsize=(8, 8))
clf_merge.local_inference_wald(tmpid).round(4)
```
## Citations
Agus Sudjianto, William Knauth, Rahul Singh, Zebin Yang and Aijun Zhang. 2020. Unwrapping The Black Box of Deep ReLU Networks: Interpretability, Diagnostics, and Simplification. [arXiv:2011.04041](https://arxiv.org/abs/2011.04041)
```latex
@article{sudjianto2020unwrapping,
title={Unwrapping The Black Box of Deep ReLU Networks: Interpretability, Diagnostics, and Simplification},
author={Sudjianto, Agus and Knauth, William and Singh, Rahul and Yang, Zebin and Zhang, Aijun},
journal={arXiv:2011.04041},
year={2020}
}
```