mlrl.testbed.data_characteristics module¶
Author: Michael Rapp (michael.rapp.ml@gmail.com)
Provides classes for printing certain characteristics of multi-label data sets. The characteristics can be written to one or several outputs, e.g., to the console or to a file.
- class mlrl.testbed.data_characteristics.DataCharacteristicsWriter(sinks: List[Sink])¶
Bases:
OutputWriter
Allows to write the characteristics of a data set to one or severals sinks.
- class CsvSink(output_dir: str, options: ~mlrl.common.options.Options = <mlrl.common.options.Options object>)¶
Bases:
CsvSink
Allows to write the characteristics of a data set to a CSV file.
- class DataCharacteristics(feature_characteristics: FeatureCharacteristics, label_characteristics: LabelCharacteristics)¶
Bases:
Formattable
,Tabularizable
Stores characteristics of a feature matrix and a label matrix.
- class mlrl.testbed.data_characteristics.FeatureCharacteristics(meta_data: MetaData, x)¶
Bases:
object
Stores characteristics of a feature matrix.
- property feature_density¶
The feature density.
- property feature_sparsity¶
The feature sparsity.
- property num_features¶
The total number of features.
- property num_nominal_features¶
The total number of nominal features.
- property num_numerical_features¶
The total number of numerical features.
- property num_ordinal_features¶
The total number of ordinal features.