utils

Functions

_create_or_update(param, name, value)	Create new parameters dict or add field to existing one.
prepare_test_batch(data, input_size)	Prepare batch with data for forecasting.
prepare_train_batch(data, input_size, ...[, ...])	Prepare batch with training data.
to_tensor(x)	Convert data to tensor and put on default device.

prepare_test_batch(data: List[Dict[str, Any]], input_size: int) → Dict[str, Any]

Prepare batch with data for forecasting.

Parameters

• data (List[Dict[str, Any]]) –

• input_size (int) –

Return type

Dict[str, Any]

prepare_train_batch(data: List[Dict[str, Any]], input_size: int, output_size: int, window_sampling_limit: Optional[int] = None, random_state: Optional[numpy.random.mtrand.RandomState] = None) → Dict[str, Optional[torch.Tensor]]

Prepare batch with training data.

Parameters

• data (List[Dict[str, Any]]) –

• input_size (int) –

• output_size (int) –

• window_sampling_limit (Optional[int]) –

• random_state (Optional[numpy.random.mtrand.RandomState]) –

Return type

Dict[str, Optional[torch.Tensor]]

to_tensor(x: Any) → torch.Tensor

Convert data to tensor and put on default device.

Parameters

x (Any) – Initial data for the tensor. Can be a list, tuple, NumPy ndarray, scalar, and other types.

Returns

Input data as tensor.

Return type

torch.Tensor