PyPI page
Home page
Author:
Alexandre Le Mercier
License:
MIT
Summary:
Security protocols for estimating adversarial robustness of machine learning models for both tabular and image datasets. This package implements a set of evasion attacks based on heuristic optimization algorithms, and complex cost functions to give reliable results for tabular problems.
Latest version:
1.0.0
Required dependencies:
lightgbm
|
matplotlib
|
numpy
|
optuna
|
pandas
|
scikit-learn
|
seaborn
|
tqdm
|
xgboost
Downloads last day:
11
Downloads last week:
35
Downloads last month:
45