cc_tk.relationship package

Submodules

cc_tk.relationship.distribution module

Module for computing distribution of data.

cc_tk.relationship.distribution.categorical_distribution(categorical_features: DataFrame) → DataFrame[source]

Compute the distribution of all categorical features.

Parameters:

categorical_features (pd.DataFrame) – Categorical features to compute the distribution of.

Returns:

Distribution of the features.

Return type:

pd.DataFrame

cc_tk.relationship.distribution.numeric_distribution(numeric_features: DataFrame) → DataFrame[source]

Compute the distribution of all numeric features.

Parameters:

numeric_features (pd.DataFrame) – Numeric features to compute the distribution of.

Returns:

Distribution of the features.

Return type:

pd.DataFrame

cc_tk.relationship.distribution.summary_distribution_by_target(features: DataFrame, target: Series) → Tuple[DataFrame, DataFrame][source]

Compute the distribution of all features by target group.

Parameters:

  • features (pd.DataFrame) – Features to compute the distribution of.

  • target (pd.Series) – Target to group by. It must be categorical.

Returns:

numeric_summary, catecorigal_summary – Distribution of the features by target group.

Return type:

Tuple[pd.DataFrame, pd.DataFrame]

cc_tk.relationship.schema module

Defines the schema for the relationship module.

class cc_tk.relationship.schema.SeriesType(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]

Bases: str, Enum

Defines the type of a series.

CATEGORICAL = 'categorical'
NUMERIC = 'numeric'
cc_tk.relationship.schema.all_columns_categorical(df: DataFrame) → bool[source]

Check if all columns in a DataFrame are categorical.

Parameters:

df (pd.DataFrame) – The DataFrame to check.

Returns:

True if all columns are categorical, False otherwise.

Return type:

bool

cc_tk.relationship.schema.all_columns_numeric(df: DataFrame) → bool[source]

Check if all columns in a DataFrame are numeric.

Parameters:

df (pd.DataFrame) – The DataFrame to check.

Returns:

True if all columns are numeric, False otherwise.

Return type:

bool

cc_tk.relationship.schema.check_input_index(*arg_names: str) → Callable[source]

Check that the specified arguments have the same index.

Parameters:

*arg_names (str) – The names of the arguments to check.

Returns:

The decorator.

Return type:

Callable

cc_tk.relationship.schema.check_input_types(*type_specs: Tuple[str, SeriesType]) → Callable[source]

Check the types of the arguments of the decorated function.

Parameters:

*type_specs (Tuple[str, SeriesType]) – A tuple of tuples, each tuple contains the name of the argument and the expected type of the argument.

Returns:

The decorator.

Return type:

Callable

cc_tk.relationship.schema.check_series_in_signature(func: Callable, *arg_names: str) → Signature[source]

Check that the specified arguments are pd.Series.

Parameters:

  • func (Callable) – The function to check.

  • *arg_names (str) – The names of the arguments to check.

Returns:

The signature.

Return type:

Signature

Raises:

  • ValueError – If an argument does not exist.

  • TypeError – If an argument is not a pd.Series.

cc_tk.relationship.significance module

Significance functions for different input types and significance types.

cc_tk.relationship.significance.get_significance(feature_type: Literal[VariableType.NUMERIC], target_type: Literal[VariableType.NUMERIC], significance_type: SignificanceType) → SignificanceNumericNumeric[source]
cc_tk.relationship.significance.get_significance(feature_type: Literal[VariableType.NUMERIC], target_type: Literal[VariableType.CATEGORICAL], significance_type: SignificanceType) → SignificanceNumericCategorical
cc_tk.relationship.significance.get_significance(feature_type: Literal[VariableType.CATEGORICAL], target_type: Literal[VariableType.NUMERIC], significance_type: SignificanceType) → SignificanceCategoricalNumeric
cc_tk.relationship.significance.get_significance(feature_type: Literal[VariableType.CATEGORICAL], target_type: Literal[VariableType.CATEGORICAL], significance_type: SignificanceType) → SignificanceCategoricalCategorical

Get significance function based on variable types and significance type.

Parameters:

  • feature_type (VariableType) – Input type of the feature.

  • target_type (VariableType) – Input type of the target.

  • significance_type (SignificanceType) – Significance type.

  • **kwargs – Additional keyword arguments passed to the significance initialization.

Returns:

Significance instance.

Return type:

Significance

Raises:

ValueError – If the feature type and target type are not supported or incompatible with the significance type.

cc_tk.relationship.summary module

Significance functions for different input types and significance types.

cc_tk.relationship.significance.get_significance(feature_type: Literal[VariableType.NUMERIC], target_type: Literal[VariableType.NUMERIC], significance_type: SignificanceType) → SignificanceNumericNumeric[source]
cc_tk.relationship.significance.get_significance(feature_type: Literal[VariableType.NUMERIC], target_type: Literal[VariableType.CATEGORICAL], significance_type: SignificanceType) → SignificanceNumericCategorical
cc_tk.relationship.significance.get_significance(feature_type: Literal[VariableType.CATEGORICAL], target_type: Literal[VariableType.NUMERIC], significance_type: SignificanceType) → SignificanceCategoricalNumeric
cc_tk.relationship.significance.get_significance(feature_type: Literal[VariableType.CATEGORICAL], target_type: Literal[VariableType.CATEGORICAL], significance_type: SignificanceType) → SignificanceCategoricalCategorical

Get significance function based on variable types and significance type.

Parameters:

  • feature_type (VariableType) – Input type of the feature.

  • target_type (VariableType) – Input type of the target.

  • significance_type (SignificanceType) – Significance type.

  • **kwargs – Additional keyword arguments passed to the significance initialization.

Returns:

Significance instance.

Return type:

Significance

Raises:

ValueError – If the feature type and target type are not supported or incompatible with the significance type.

cc_tk.relationship.utils module

Utility functions to perform caracterisation.

cc_tk.relationship.utils.cut_influence(influence_values: Series) → Series[source]

Cut influence values into categories.

Influence values define the impact of a given feature on another one, this function helps categorizing these influences into 6 categories: - –: Strong negative influence. - -: Weak negative influence. - `` or ` : No influence. - `+: Weak positive influence. - ++: Strong positive influence.

Parameters:

influence_values (pd.Series) – Influence values to cut.

Returns:

Categorical influence values.

Return type:

pd.Series

cc_tk.relationship.utils.influence_from_correlation(correlation_value: float) → str[source]

Compute the influence from a correlation value.

Parameters:

correlation_value (float) – Correlation value to compute the influence from.

Returns:

Influence value.

Return type:

str

Module contents

relationship module is dedicated to testing statistical relationship.