DataModule.inverse_transform

DataModule.inverse_transform(x: ndarray | torch.Tensor, y: ndarray | torch.Tensor, out_type: str = 'numpy') → Tuple[ndarray | torch.Tensor, ndarray | torch.Tensor][source]

Unnormalize the input and output data. Takes care of the case where x and y are numpy arrays or torch tensors.

Parameters:

  • x (Union[np.ndarray, torch.Tensor]) – Input data to be un-transformed, either a numpy array or a torch tensor.

  • y (Union[np.ndarray, torch.Tensor]) – Output data to be un-transformed, either a numpy array or a torch tensor.

  • out_type (str, optional) – The type of the output. Can be “numpy”, “torch”, or “same”. The default is “numpy”.

Returns:

Tuple[Union[np.ndarray, torch.Tensor], Union[np.ndarray, torch.Tensor]] – A tuple containing the un-transformed input and output data.