dashi package

Subpackages

• dashi.supervised_characterization package
  • Submodules
  • dashi.supervised_characterization.arrange_metrics module
    • arrange_performance_metrics()
  • dashi.supervised_characterization.estimate_models module
    • estimate_multibatch_models()
  • dashi.supervised_characterization.plot_performance module
    • plot_multibatch_performance()
  • Module contents
• dashi.unsupervised_characterization package
  • Subpackages
    • dashi.unsupervised_characterization.data_temporal_map package
      • Submodules
      • dashi.unsupervised_characterization.data_temporal_map.data_temporal_map module
      • dashi.unsupervised_characterization.data_temporal_map.data_temporal_map_plotter module
      • Module contents
    • dashi.unsupervised_characterization.igt package
      • Submodules
      • dashi.unsupervised_characterization.igt.igt_plotting module
      • dashi.unsupervised_characterization.igt.igt_projection module
      • dashi.unsupervised_characterization.igt.igt_projection_estimator module
      • dashi.unsupervised_characterization.igt.igt_trajectory_estimator module
      • Module contents
  • Module contents

Submodules

dashi.constants module

dashi.utils module

Utils functions

format_data(input_dataframe, *, date_column_name=None, source_column_name=None, date_format='%y/%m/%d', verbose=False, numerical_column_names=None, categorical_column_names=None)

Function to transform dates into ‘Date’ Python format

Parameters:

• input_dataframe (pd.DataFrame) – Pandas dataframe object with at least one columns of dates.

• date_column_name (Optional[str]) – The name of the column containing the dates. If you are not performing a temporal analysis, set this parameter to None.

• source_column_name (Optional[str]) – The name of the column containing the source information. If you are not performing a multi-source analysis, set this parameter to None.

• date_format (Optional[str]) – Structure of date format. By default ‘%y/%m/%d’.

• verbose (bool) – Whether to display additional information during the process. Defaults to False.

• numerical_column_names (Optional[List[str]]) – A list containing all the numerical column names in the dataset. If this parameter is None, the variables types must be managed by the user.

• categorical_column_names (Optional[List[str]]) – A list containing all the categorical column names in the dataset. If this parameter is None, the variables types must be managed by the user.

Returns:

A pandas.DataFrame with each column cast to its correct dtype and any rows containing missing values in the date or source fields dropped.

Return type:

pd.DataFrame

Module contents