iaml.metrics.mean_absolute_error_metric¶
[METRIC] Mean Absolute Error
Classes¶
[METRIC] Mean Absolute Error |
Module Contents¶
- class iaml.metrics.mean_absolute_error_metric.MeanAbsoluteErrorMetric¶
Bases:
iaml.metric.Metric[METRIC] Mean Absolute Error
- name: str = 'Mean Absolute Error'¶
Return the metric formatted name
- Returns:
Formatted name.
- greater_is_better = False¶
Whether larger values are better. Set to False for error metrics.
- refs: list[dict[str, Any]]¶
List of references for this metric.
- __str__() str¶
Metric name
- Returns:
Metric name.
- suitable(X: pandas.DataFrame, y: pandas.DataFrame, type_of_target: str) bool¶
Is this metric usable given X and y ?
- Parameters:
X (pd.DataFrame) – The dataset we try to compute metrics on.
y (pd.DataFrame) – The dataset target we try to compute metrics on
type_of_target (str) – The type of target we are trying to predict
- Returns:
Suitable ?
- compute(y: pandas.DataFrame, y_pred: pandas.DataFrame, **kwargs) float¶
Compute metric given y, y_pred. Must be overridden by children classes
- Parameters:
y (pd.DataFrame) – Ground truth to compute the metric.
y_pred (pd.DataFrame) – Prediction to compute the metric.
**kwargs (dict, optional) – Additional parameters
- Returns:
Computed value