spectrochempy.log_transform๏ƒ

log_transform(dataset, method='log1p', eps=1e-10, inplace=False)[source]๏ƒ

Apply a logarithmic transform.

Parameters:

  • dataset (NDDataset) โ€“ The input data.

  • method (str, optional, default:โ€™log1pโ€™) โ€“ Transform to apply:

    • 'log1p' โ€” compute log(1 + x) (stable for small or zero values).

    • 'log' โ€” compute log(x). If the data contain values \(\le 0\), a small offset eps is added automatically.

  • eps (float, optional, default:1e-10) โ€“ Offset added when method='log' and non-positive values are present.

  • inplace (bool, optional, default:False) โ€“ If True, the transform is performed in place.

Returns:

NDDataset โ€“ The log-transformed dataset.

Examples

>>> dataset = scp.read("irdata/nh4y-activation.spg")
>>> nd = dataset.log_transform(method="log1p")