whylogs.core.model_performance_metrics

Submodules

• whylogs.core.model_performance_metrics.confusion_matrix
• whylogs.core.model_performance_metrics.model_performance_metrics
• whylogs.core.model_performance_metrics.regression_metrics

Package Contents

Classes

ModelPerformanceMetrics

Container class for various model-related performance metrics

class whylogs.core.model_performance_metrics.ModelPerformanceMetrics(confusion_matrix: Optional[whylogs.core.model_performance_metrics.confusion_matrix.ConfusionMatrix] = None, regression_metrics: Optional[whylogs.core.model_performance_metrics.regression_metrics.RegressionMetrics] = None, metrics: Optional[Dict[str, whylogs.core.metrics.metrics.Metric]] = None, field_metadata: Optional[Dict[str, Set[str]]] = None)

Container class for various model-related performance metrics

Parameters

• confusion_matrix (Optional[whylogs.core.model_performance_metrics.confusion_matrix.ConfusionMatrix]) –

• regression_metrics (Optional[whylogs.core.model_performance_metrics.regression_metrics.RegressionMetrics]) –

• metrics (Optional[Dict[str, whylogs.core.metrics.metrics.Metric]]) –

• field_metadata (Optional[Dict[str, Set[str]]]) –

confusion_matrix

ConfusionMatrix which keeps it track of counts with NumberTracker

Type

ConfusionMatrix

regression_metrics

Regression Metrics keeps track of a common regression metrics in case the targets are continous.

Type

RegressionMetrics

property output_fields: Optional[List[str]]

Return type

Optional[List[str]]

to_protobuf() → whylogs.core.proto.v0.ModelProfileMessage

Return type

whylogs.core.proto.v0.ModelProfileMessage

classmethod from_protobuf(message: whylogs.core.proto.v0.ModelProfileMessage) → ModelPerformanceMetrics

Parameters

message (whylogs.core.proto.v0.ModelProfileMessage) –

Return type

ModelPerformanceMetrics

compute_confusion_matrix(predictions: List[Union[str, int, bool, float]], targets: List[Union[str, int, bool, float]], scores: Optional[List[float]] = None)

computes the confusion_matrix, if one is already present merges to old one.

Parameters

• predictions (List[Union[str, int, bool]]) –

• targets (List[Union[str, int, bool]]) –

• scores (List[float], optional) –

compute_regression_metrics(predictions: List[Union[float, int]], targets: List[Union[float, int]])

Parameters

• predictions (List[Union[float, int]]) –

• targets (List[Union[float, int]]) –

add_metadata_to_field(column_name: str, categories: Set[str]) → None

Parameters

• column_name (str) –

• categories (Set[str]) –

Return type

None

specify_output_fields(column_names: Union[str, Set[str]]) → None

Parameters

column_names (Union[str, Set[str]]) –

Return type

None

merge(other) → ModelPerformanceMetrics

Return type

ModelPerformanceMetrics