Skip to content

Latest commit

 

History

History
81 lines (62 loc) · 2.34 KB

README.md

File metadata and controls

81 lines (62 loc) · 2.34 KB

INVASE: Instance-wise Variable Selection

Tests Downloads arXiv Test In Colab License: MIT

image

Authors: Jinsung Yoon, James Jordon, Mihaela van der Schaar

Paper: Jinsung Yoon, James Jordon, Mihaela van der Schaar, "IINVASE: Instance-wise Variable Selection using Neural Networks," International Conference on Learning Representations (ICLR), 2019. (https://openreview.net/forum?id=BJg_roAcK7)

🚀 Installation

The library can be installed from PyPI using

$ pip install invase

or from source, using

$ pip install .

💥 Sample Usage

import pandas as pd

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression

from invase import INVASE

X, y = load_iris(return_X_y=True, as_frame = True)

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

## Load the model
model = LogisticRegression()

model.fit(X_train, y_train)

## Load INVASE
explainer = INVASE(
    model, 
    X_train, 
    y_train, 
    n_epoch=1000, 
    prefit = True, # the model is already trained
)

## Explain
explainer.explain(X_test.head(5))

🔨 Tests

Install the testing dependencies using

pip install .[testing]

The tests can be executed using

pytest -vsx

Citing

If you use this code, please cite the associated paper:

@inproceedings{
    yoon2018invase,
    title={{INVASE}: Instance-wise Variable Selection using Neural Networks},
    author={Jinsung Yoon and James Jordon and Mihaela van der Schaar},
    booktitle={International Conference on Learning Representations},
    year={2019},
    url={https://openreview.net/forum?id=BJg_roAcK7},
}