Module preprocessing 

Data preprocessing utilities for QML

Functions

l1_normalize

L1 normalization - normalize each sample (row) to unit L1 norm Each row will sum to 1, useful for probability-like features

l2_normalize

L2 normalization - normalize each sample (row) to unit L2 norm Each row will have length 1, useful for cosine similarity

max_abs_scale

Max absolute scaling - scales features by maximum absolute value Useful for sparse data where centering would destroy sparsity

min_max_normalize

Min-max normalization to [0, 1] range

quantile_normalize

Quantile normalization - forces features to have the same distribution Useful when features should be on the same scale but have different distributions

robust_scale

Robust scaling using median and IQR (Interquartile Range) More robust to outliers than standardization

standardize

Normalize features to zero mean and unit variance