nannyml.performance_calculation.metrics.regression module

Performance Calculation Regression Metrics Module.

class nannyml.performance_calculation.metrics.regression.MAE(y_true: str, y_pred: str, threshold: Threshold, y_pred_proba: Optional[str] = None, **kwargs)[source]

Bases: Metric

Mean Absolute Error metric.

Creates a new MAE instance.

Parameters:

y_true (str) – The name of the column containing target values.
y_pred (str) – The name of the column containing your model predictions.
threshold (Threshold) – The Threshold instance that determines how the lower and upper threshold values will be calculated.
y_pred_proba (Optional[str], default=None) – Name of the column containing your model output.

__str__()[source]: Get string representation of metric.

y_pred: str

class nannyml.performance_calculation.metrics.regression.MAPE(y_true: str, y_pred: str, threshold: Threshold, y_pred_proba: Optional[str] = None, **kwargs)[source]

Bases: Metric

Mean Absolute Percentage Error metric.

Creates a new MAPE instance.

Parameters:

y_true (str) – The name of the column containing target values.
y_pred (str) – The name of the column containing your model predictions.
threshold (Threshold) – The Threshold instance that determines how the lower and upper threshold values will be calculated.
y_pred_proba (Optional[str], default=None) – Name of the column containing your model output.

__str__()[source]: Get string representation of metric.

y_pred: str

class nannyml.performance_calculation.metrics.regression.MSE(y_true: str, y_pred: str, threshold: Threshold, y_pred_proba: Optional[str] = None, **kwargs)[source]

Bases: Metric

Mean Squared Error metric.

Creates a new MSE instance.

Parameters:

y_true (str) – The name of the column containing target values.
y_pred (str) – The name of the column containing your model predictions.
threshold (Threshold) – The Threshold instance that determines how the lower and upper threshold values will be calculated.
y_pred_proba (Optional[str], default=None) – Name of the column containing your model output.

__str__()[source]: Get string representation of metric.

y_pred: str

class nannyml.performance_calculation.metrics.regression.MSLE(y_true: str, y_pred: str, threshold: Threshold, y_pred_proba: Optional[str] = None, **kwargs)[source]

Bases: Metric

Mean Squared Logarithmic Error metric.

Creates a new MSLE instance.

Parameters:

y_true (str) – The name of the column containing target values.
y_pred (str) – The name of the column containing your model predictions.
threshold (Threshold) – The Threshold instance that determines how the lower and upper threshold values will be calculated.
y_pred_proba (Optional[str], default=None) – Name of the column containing your model output.

__str__()[source]: Get string representation of metric.

y_pred: str

class nannyml.performance_calculation.metrics.regression.RMSE(y_true: str, y_pred: str, threshold: Threshold, y_pred_proba: Optional[str] = None, **kwargs)[source]

Bases: Metric

Root Mean Squared Error metric.

Creates a new RMSE instance.

Parameters:

y_true (str) – The name of the column containing target values.
y_pred (str) – The name of the column containing your model predictions.
threshold (Threshold) – The Threshold instance that determines how the lower and upper threshold values will be calculated.
y_pred_proba (Optional[str], default=None) – Name of the column containing your model output.

__str__()[source]: Get string representation of metric.

y_pred: str

class nannyml.performance_calculation.metrics.regression.RMSLE(y_true: str, y_pred: str, threshold: Threshold, y_pred_proba: Optional[str] = None, **kwargs)[source]

Bases: Metric

Root Mean Squared Logarithmic Error metric.

Creates a new RMSLE instance.

Parameters:

y_true (str) – The name of the column containing target values.
y_pred (str) – The name of the column containing your model predictions.
threshold (Threshold) – The Threshold instance that determines how the lower and upper threshold values will be calculated.
y_pred_proba (Optional[str], default=None) – Name of the column containing your model output.

__str__()[source]: Get string representation of metric.

y_pred: str