dashi v0.3.2

dashi v0.3.2

This release introduces new improvements for temporal map handling, visualization customization, and supervised model training configuration.

What's Changed

Unsupervised characterization

• Adapted trim_data_temporal_map to support MultivariateDataTemporalMaps.

• Added start_date and end_date arguments to:

  • plot_multivariate_data_temporal_map
  • plot_conditional_data_temporal_map

  These arguments allow users to restrict temporal map visualizations to a specific date range.

• Added the color_palette argument to:

  • plot_multivariate_data_temporal_map
  • plot_conditional_data_temporal_map

  This allows users to customize the color scheme of temporal map plots.

Supervised characterization

• Added new training configuration arguments to estimate_multibatch_models:

  • max_iter
  • n_estimators
  • max_depth

  These arguments allow users to customize the supervised models trained across batches.

  Notes:

  • max_iter is only applied to histogram gradient boosting models.
  • n_estimators and max_depth are only applied to random forest models.

Installation

pip install --upgrade dashi

Note: All new function parameters have been implemented with sensible defaults to ensure strict backward compatibility with scripts utilizing dashi v0.3.1.