v0.2.3 (May 13, 2023)#

These are minor changes in v0.2.2 that shows fugs fixed, features and improvements.

  • API change Change performed in watex.utils.funcutils.assert_ratio(). The parameter as_percent does no longer exist. It is depreated and modified to in_percent.

  • Fix Bug fixed in concatening sring argument passed to titles parameter in watex.view.plotLearningInspections(). Henceforth, no need to put only single model in litteral string. However for a single model training, watex.view.plotLearningInspection() is prefered instead. Note the 's' at the end of the former function.

  • Fix Bug fixed in watex.utils.plot_logging() when parameter \(X\) is not passed as a DataFrame. Henceforth, a column of random item is generated and is used to build a temporray DataFrame for plotting.

  • Feature Add watex.datasets.load_mxs() load new datasets for performing the MXS. shuffle parameter is needed to be triggered to True for shuffling data before any splitting. By edfaultt it’s set to False.

  • Feature add transformers watex.transformers.KMeansFeaturizer to transform numeric data into k-means cluster memberships. It runs k-means on the input data and converts each data point into the ID of the closest cluster.

  • Feature add watex.transformers.featurize_X() to transform predictor \(X\) at once. It applies the K-Means featurization approach. Refer to watex.transformers.KMeansFeaturizer

  • Feature watex.utils.plotutils.plot_voronoi() plots the Voronoi diagram of the k-Means clusters overlaid with the data.

  • Feature new function watex.utils.plot_roc_curves() for visualizing the Receiving Operating Characterisctic (ROC) curves either into a single plot ( used by default) or individual plot for each model by setting the parameter all=True.