from aitoolbox.experiment.core_metrics.abstract_metric import AbstractBaseMetric
from sklearn.metrics import mean_squared_error, mean_absolute_error
[docs]class MeanSquaredErrorMetric(AbstractBaseMetric):
def __init__(self, y_true, y_predicted):
"""Model prediction MSE
Args:
y_true (numpy.ndarray or list): ground truth targets
y_predicted (numpy.ndarray or list): predicted targets
"""
AbstractBaseMetric.__init__(self, y_true, y_predicted, metric_name='Mean_squared_error')
[docs] def calculate_metric(self):
return mean_squared_error(self.y_true, self.y_predicted)
[docs]class MeanAbsoluteErrorMetric(AbstractBaseMetric):
def __init__(self, y_true, y_predicted):
"""Model prediction MAE
Args:
y_true (numpy.ndarray or list): ground truth targets
y_predicted (numpy.ndarray or list): predicted targets
"""
AbstractBaseMetric.__init__(self, y_true, y_predicted, metric_name='Mean_absolute_error')
[docs] def calculate_metric(self):
return mean_absolute_error(self.y_true, self.y_predicted)