def _setup_loss(self):
if self.loss[TYPE] == 'mean_squared_error':
self.train_loss_function = MSELoss()
self.eval_loss_function = MSEMetric(name='eval_loss')
elif self.loss[TYPE] == 'mean_absolute_error':
self.train_loss_function = MAELoss()
self.eval_loss_function = MAEMetric(name='eval_loss')
elif self.loss[TYPE] == SOFTMAX_CROSS_ENTROPY:
self.train_loss_function = SoftmaxCrossEntropyLoss(
num_classes=self.vector_size,
feature_loss=self.loss,
name='train_loss')
self.eval_loss_function = SoftmaxCrossEntropyMetric(
num_classes=self.vector_size,
feature_loss=self.loss,
name='eval_loss')
else:
raise ValueError('Unsupported loss type {}'.format(
self.loss[TYPE]))
def _setup_metrics(self):
self.metric_functions = {} # needed to shadow class variable
if self.loss[TYPE] == "mean_squared_error":
self.metric_functions[LOSS] = MSEMetric(name="eval_loss")
elif self.loss[TYPE] == "mean_absolute_error":
self.metric_functions[LOSS] = MAEMetric(name="eval_loss")
elif self.loss[TYPE] == "root_mean_squared_error":
self.metric_functions[LOSS] = RMSEMetric(name="eval_loss")
elif self.loss[TYPE] == "root_mean_squared_percentage_error":
self.metric_functions[LOSS] = RMSPEMetric(name="eval_loss")
self.metric_functions[MEAN_SQUARED_ERROR] = MeanSquaredErrorMetric(
name="metric_mse"
)
self.metric_functions[MEAN_ABSOLUTE_ERROR] = MeanAbsoluteErrorMetric(
name="metric_mae"
)
self.metric_functions[
ROOT_MEAN_SQUARED_ERROR
] = RootMeanSquaredErrorMetric(name="metric_rmse")
self.metric_functions[
ROOT_MEAN_SQUARED_PERCENTAGE_ERROR
] = RMSPEMetric(name="metric_rmspe")
self.metric_functions[R2] = R2Score(name="metric_r2")