v1.1.1

v1.1.1

V1.0.14

PyXAI is available on MacOS via pypi: https://pypi.org/project/pyxai/1.0.14/

• Supports ARM architecture on Linux and MacOS
• Some bugs have been resolved

pip install -U pyxai to update or install.

v1.0.11

To fix the issue #10.

v1.0.10

• To resolve compatibility issues with PyQt6, since V1.0.10, PyXAI’s Graphical Interface is independent and no longer mandatory.
• Remove the PyQt6 dependency and implementation of new methods to display explanations without PyQt6:
  • show_in_notebook()
  • show_on_screen()
  • get_PILImage()
  • save_png()
  • resize_PILimage()
  • More information on a new page of the documentation
• Contrastive for BT classification (binary classes)(documentation in progress)
• Change function name in explainer (unset_specific_features -> unset_excluded_features)
• New procedure installation (github and pypi)
• New visualization for time series
  
  More information at the end of this page.
• Compilation error resolution

The documentation will be updated tomorrow.

v1.0.9

New metrics (documentation page is in progress)

For binary classification:

• accuracy
• precision
• recall
• f1_score
• specificity
• tp, tn, fp, fn

For multiclass classification:

• micro_averaging_accuracy
• micro_averaging_precision
• micro_averaging_recall
• macro_averaging_accuracy
• macro_averaging_precision
• macro_averaging_recall

For regression:

• mean_squared_error
• root_mean_squared_error
• mean_absolute_error

Examples:

labels = [1,1,1,1,1,0,0,0,0,0]
predictions = [1,1,1,1,1,0,0,0,0,0]
metrics = Tools.Metric.compute_metrics_binary_classification(labels, predictions)

learner = Learning.Scikitlearn("tests/dermatology.csv", learner_type=Learning.CLASSIFICATION)
models = learner.evaluate(method=Learning.K_FOLDS, output=Learning.DT, test_size=0.2)
for id, models in enumerate(models):
     metrics = learner.get_details()[id]["metrics"]

v1.0.7

Build and Tests with CI

V1.0.0

This is the release 1.0.0 of PyXAI with new features:

• Regression for Boosted Trees with XGBoost or LightGBM
• Adding Theories (knowledge about the dataset)
• Easier model Import (automatic detection of model types)
• PyXAI's Graphical User Interface (GUI): displaying, loading and saving explanations.
• Supports Multiple Image Formats for imaging datasets
• Supports Data Pre-Processing (tool for preparing and cleansing a dataset)
• Unit Tests with the Unittest module

V0.8.7

This is the first release of PyXAI.