idanmoradarthas / DataScienceUtils / 21644420479

100%

push

github

Handle missing values in plot features interaction (#98)

* Add remove_na parameter to plot_features_interaction for handling missing data

- Introduced a new boolean parameter `remove_na` to the `plot_features_interaction` function, allowing users to choose whether to exclude rows with NA values for the specified features before plotting.
- Updated relevant plotting functions to utilize the `remove_na` parameter, ensuring consistent handling of missing data across various feature types.
- Enhanced documentation to clarify the functionality and usage of the `remove_na` parameter.
- Updated baseline images to reflect changes in plot outputs due to the new parameter.

* Refactor datetime plotting functions for improved clarity

- Renamed `_plot_xy` to `_plot_datetime_vs_numeric` to better reflect its purpose of plotting datetime against numeric features.
- Updated calls to the renamed function within `plot_features_interaction` and `_plot_datetime_feature1` for consistency.
- Enhanced code readability and maintainability by clarifying function names related to datetime visualizations.

* minor reformatting

* Enhance _plot_categorical_vs_numeric function for better handling of missing values

- Updated the docstring to clarify the behavior of the function regarding outlier handling and missing values.
- Implemented logic to add a "Missing" category for missing categorical values.
- Added rug plots for missing numeric values to improve visualization clarity.
- Ensured that the main violin plot only includes non-NA numeric values for accurate representation.

* Enhance _plot_categorical_vs_categorical function to improve missing value handling

- Updated the docstring to clarify how missing values are managed based on the remove_na parameter.
- Implemented logic to add a "Missing" category for NaN values when remove_na is False.
- Adjusted crosstab creation to include or exclude missing values based on the remove_na setting, ensuring accurate representati... (continued)

191 of 191 new or added lines in 1 file covered. (100.0%)

861 of 861 relevant lines covered (100.0%)

12.0 hits per line