doc

Define preprocessors used in documentation tutorials.

Classes:

Name	Description
Scaler	Air quality, ECG, HAR and Gas Price preprocessors.

Scaler(input_size: tuple[int, ...], mean: torch.Tensor, scale: torch.Tensor)

Air quality, ECG, HAR and Gas Price preprocessors.

Methods:

Name	Description
transform	Transform.
inverse_transform	Apply inverse transform.

Source code in src/xpdeep/dataset/preprocessor/zoo/doc.py

def __init__(self, input_size: tuple[int, ...], mean: torch.Tensor, scale: torch.Tensor):
    super().__init__(input_size=input_size)
    # Saved as buffer for torch.export: saved loaded with `state_dict` but not optimized with `optimizer.step()
    self.register_buffer("mean", mean)
    self.register_buffer("scale", scale)

transform(inputs: torch.Tensor) -> torch.Tensor

Transform.

Source code in src/xpdeep/dataset/preprocessor/zoo/doc.py

def transform(self, inputs: torch.Tensor) -> torch.Tensor:
    """Transform."""
    return (inputs - self.mean) / self.scale  # type: ignore[operator]

inverse_transform(output: torch.Tensor) -> torch.Tensor

Apply inverse transform.

Source code in src/xpdeep/dataset/preprocessor/zoo/doc.py

def inverse_transform(self, output: torch.Tensor) -> torch.Tensor:
    """Apply inverse transform."""
    return output * self.scale + self.mean  # type: ignore[operator]