# Copyright 2018 Amazon.com, Inc. or its affiliates. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License").
# You may not use this file except in compliance with the License.
# A copy of the License is located at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# or in the "license" file accompanying this file. This file is distributed
# on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either
# express or implied. See the License for the specific language governing
# permissions and limitations under the License.
from typing import Dict
import numpy as np
[docs]def num_masked_target_values(data: Dict[str, np.ndarray]) -> np.ndarray:
if np.ma.isMaskedArray(data["label"]):
assert isinstance(data["label"], np.ma.MaskedArray)
return data["label"].mask.astype(float)
else:
return np.zeros(data["label"].shape)
[docs]def absolute_label(data: Dict[str, np.ndarray]) -> np.ndarray:
return np.abs(data["label"])
[docs]def error(data: Dict[str, np.ndarray], forecast_type: str) -> np.ndarray:
return data["label"] - data[forecast_type]
[docs]def absolute_error(
data: Dict[str, np.ndarray], forecast_type: str
) -> np.ndarray:
return np.abs(error(data, forecast_type))
[docs]def squared_error(
data: Dict[str, np.ndarray], forecast_type: str
) -> np.ndarray:
return np.square(error(data, forecast_type))
[docs]def quantile_loss(data: Dict[str, np.ndarray], q: float) -> np.ndarray:
forecast_type = str(q)
prediction = data[forecast_type]
return 2 * np.abs(
error(data, forecast_type) * ((prediction >= data["label"]) - q)
)
[docs]def coverage(data: Dict[str, np.ndarray], q: float) -> np.ndarray:
forecast_type = str(q)
return (data["label"] <= data[forecast_type]).astype(float)
[docs]def absolute_percentage_error(
data: Dict[str, np.ndarray], forecast_type: str
) -> np.ndarray:
return absolute_error(data, forecast_type) / absolute_label(data)
[docs]def symmetric_absolute_percentage_error(
data: Dict[str, np.ndarray], forecast_type: str
) -> np.ndarray:
return (
2
* absolute_error(data, forecast_type)
/ (absolute_label(data) + np.abs(data[forecast_type]))
)
[docs]def scaled_interval_score(
data: Dict[str, np.ndarray], alpha: float
) -> np.ndarray:
lower_quantile = data[str(alpha / 2)]
upper_quantile = data[str(1.0 - alpha / 2)]
label = data["label"]
numerator = (
upper_quantile
- lower_quantile
+ 2.0 / alpha * (lower_quantile - label) * (label < lower_quantile)
+ 2.0 / alpha * (label - upper_quantile) * (label > upper_quantile)
)
return numerator / data["seasonal_error"]
[docs]def absolute_scaled_error(
data: Dict[str, np.ndarray], forecast_type: str
) -> np.ndarray:
return absolute_error(data, forecast_type) / data["seasonal_error"]
[docs]def scaled_quantile_loss(data: Dict[str, np.ndarray], q: float) -> np.ndarray:
return quantile_loss(data, q) / data["seasonal_error"]