shredx.utils.scaling.inverse_min_max_scale

shredx.utils.scaling.inverse_min_max_scale(scaled_tensor: Float[Tensor, '... dim'], original_min_max: tuple[float, float], feature_range: tuple[float, float] = (0, 1)) → Float[Tensor, '... dim']

Invert min-max scaling and recover the original data scale.

This performs the reverse transformation of min_max_scale() using the original (min, max) values and the feature_range used during scaling.

Parameters:

scaled_tensorFloat[torch.Tensor, “… dim”]

Tensor that was previously scaled with min-max normalization.

original_min_maxtuple[float, float]

Tuple (min, max) values from the original data before scaling.

feature_rangetuple[float, float], optional

Feature range that was used during the original scaling, by default (0, 1).

Returns:

Float[torch.Tensor, “… dim”]

Tensor mapped back to the original scale.

Raises:

ValueError

If feature_range is not a tuple of two elements.

ValueError

If original_min_max is not a tuple of two elements.