pp_log — Element-wise logarithmic transform¶
Group: Preprocessing · C ABI: n4m_transform_log_transform_*
Description¶
Element-wise logarithm with optional fit-time auto-offset.
Parameters¶
Name |
Type |
Default |
|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
API and bindings¶
C ABI (ABI 2): n4m_transform_log_transform_create · n4m_transform_log_transform_destroy · n4m_transform_log_transform_fit · n4m_transform_log_transform_is_fitted · n4m_transform_log_transform_transform. Use the linked public header for the exact signature, configuration, and result handles.
Python (verified public re-export):
from n4m.transform.scaling import LogTransform
Source signature: LogTransform(base: float = 0.0, offset: float = 0.0, auto_offset: bool = True, min_value: float = 1e-08).
R: no current source-verified entry point was found for this catalog method.
MATLAB / Octave: no current source-verified entry point was found for this catalog method.
Explanations¶
Bibliographic source¶
No canonical paper: this is the mathematical logarithm with an implementation-defined offset policy.
Mathematical principle¶
Transform evaluates \(\log_b(x+c)\), using the natural logarithm when base=0. With auto_offset, fit chooses an offset that moves the training minimum to at least min_value; the same offset is reused on later data.
Appropriate uses¶
Compressing right-skewed positive intensities or expressing multiplicative changes on an additive scale.
Limits and validation¶
Nonpositive shifted values are invalid. A data-derived offset changes interpretation and must be fit without leakage; logs are not interchangeable with physical absorbance unless the correct sign and base are used.
Implementation¶
n4m.transform.scaling.LogTransform uses n4m_transform_log_transform_*; offset fitting and base conversion are in preprocessing/scaling/log_transform.c.
The ABI-2 implementation is the n4m_transform_log_transform_* lifecycle in libn4m.
Sources and provenance¶
See also: methods index · interactive dashboard