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')
else:
raise ValueError('Unsupported loss type {}'.format(
self.loss[TYPE]))
def _setup_loss(self):
if self.loss[TYPE] == "mean_squared_error":
self.train_loss_function = MSELoss()
elif self.loss[TYPE] == "mean_absolute_error":
self.train_loss_function = MAELoss()
elif self.loss[TYPE] == "root_mean_squared_error":
self.train_loss_function = RMSELoss()
elif self.loss[TYPE] == "root_mean_squared_percentage_error":
self.train_loss_function = RMSPELoss()
else:
raise ValueError(
"Unsupported loss type {}".format(self.loss[TYPE])
)
self.eval_loss_function = self.train_loss_function
def test_model_weights_match_training(tmpdir, csv_filename):
np.random.seed(1)
input_features = [number_feature()]
output_features = [number_feature()]
output_feature_name = output_features[0][NAME]
# Generate test data
data_csv_path = generate_data(input_features,
output_features,
os.path.join(tmpdir, csv_filename),
num_examples=100)
config = {
"input_features": input_features,
"output_features": output_features,
"trainer": {
"epochs": 5,
"batch_size": 32
},
}
model = LudwigModel(config=config, )
training_stats, _, _ = model.train(training_set=data_csv_path,
random_seed=1919)
# generate predicitons from training data
df = pd.read_csv(data_csv_path)
predictions = model.predict(df)
# compute loss on predictions from training data
loss_function = MSELoss()
loss = loss_function(
torch.tensor(predictions[0][output_feature_name +
"_predictions"].values), # predictions
torch.tensor(df[output_feature_name].values), # target
).type(torch.float32)
# get last loss value from training
last_training_loss = torch.tensor(
training_stats[TRAINING][output_feature_name][LOSS][-1])
# loss from predictions should match last loss value recorded during training
assert torch.isclose(loss, last_training_loss), (
"Model predictions on training set did not generate same loss value as in training. "
"Need to confirm that weights were correctly captured in model.")
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]))