Expand description
Feature preprocessing (scalers, PCA).
Modules§
- binarizer
- cca
- Canonical Correlation Analysis.
- fast_
ica - FastICA — fixed-point Independent Component Analysis with deflation.
- kbins_
discretizer - kernel_
pca - Kernel PCA.
- label_
encoder - max_
abs_ scaler - minmax_
scaler - mutual_
information - nmf
- Non-negative Matrix Factorisation.
- normalizer
- one_
hot_ encoder - ordinal_
encoder - pca
- pls
- Partial Least Squares Regression (PLS1).
- polynomial_
features - power_
transformer - quantile_
transformer - rfe
- Recursive Feature Elimination (RFE).
- robust_
scaler - select_
from_ model - select_
k_ best - simple_
imputer - standard_
scaler - truncated_
svd - Truncated SVD (a.k.a. LSA when applied to a term-document matrix).
- variance_
threshold
Structs§
- Binarizer
- Parameters for Binarizer (unfitted state).
- Cca
- FastIca
- Fitted
Binarizer - Fitted Binarizer — stateless, stores only the threshold.
- Fitted
Cca - Fitted
Fast Ica - FittedK
Bins Discretizer - Fitted KBinsDiscretizer – holds bin edges per feature.
- Fitted
Kernel Pca - Fitted
Label Encoder - Fitted LabelEncoder — holds the learned vocabulary and mapping.
- Fitted
MaxAbs Scaler - Fitted MaxAbsScaler — holds the maximum absolute value per feature.
- Fitted
MinMax Scaler - Fitted MinMaxScaler — holds learned min/max per feature.
- Fitted
Mutual Information Selector - Fitted
MutualInformationSelector— holds per-feature MI scores and the indices of the selected top-k features. - Fitted
Nmf - Fitted
Normalizer - Fitted Normalizer — stateless (fit is a validation-only no-op).
- Fitted
OneHot Encoder - Fitted OneHotEncoder — holds the number of unique categories per column.
- Fitted
Ordinal Encoder - Fitted OrdinalEncoder — holds per-column vocabularies and mappings.
- Fitted
Pca - Fitted PCA — holds learned principal components, explained variance, and mean.
- Fitted
PlsRegression - Fitted
Polynomial Features - Fitted PolynomialFeatures — stores the number of input features.
- Fitted
Power Transformer - Fitted PowerTransformer – holds learned lambdas per feature and optional standardization parameters (mean and std).
- Fitted
Quantile Transformer - Fitted QuantileTransformer – holds quantile references per feature.
- Fitted
Rfe - Fitted
Robust Scaler - Fitted RobustScaler — holds learned median and IQR per feature.
- Fitted
Select From Model - Fitted
SelectFromModel– holds the original importances and the indices of the selected features. - Fitted
SelectK Best - Fitted
SelectKBest– holds per-feature scores and the indices of the selected top-k features. - Fitted
Sequential Feature Selector - Fitted
Simple Imputer - Fitted
SimpleImputer— holds one fill value per column. - Fitted
Standard Scaler - Fitted StandardScaler — holds learned mean and std per feature.
- Fitted
Truncated Svd - Fitted
Variance Threshold - Fitted
VarianceThreshold— holds learned per-feature variances and the indices of features that exceeded the threshold. - KBins
Discretizer - Parameters for KBinsDiscretizer (unfitted state).
- Kernel
Pca - Label
Encoder - Encodes string labels as integer indices.
- MaxAbs
Scaler - Parameters for MaxAbsScaler (unfitted state).
- MinMax
Scaler - Parameters for MinMaxScaler (unfitted state).
- Mutual
Information Selector - Parameters for
MutualInformationSelector(unfitted state). - Nmf
- Normalizer
- Parameters for Normalizer (unfitted state).
- OneHot
Encoder - One-hot encoder for integer-encoded categorical features.
- Ordinal
Encoder - Encodes string categories as ordinal (integer) values per column.
- Pca
- Parameters for PCA (unfitted state).
- PlsRegression
- Polynomial
Features - Generates polynomial and interaction features.
- Power
Transformer - Parameters for PowerTransformer (unfitted state).
- Quantile
Transformer - Parameters for QuantileTransformer (unfitted state).
- Rfe
- Robust
Scaler - Parameters for RobustScaler (unfitted state).
- Select
From Model - Parameters for
SelectFromModelfeature selector (unfitted state). - SelectK
Best - Parameters for
SelectKBestfeature selector (unfitted state). - Sequential
Feature Selector - Simple
Imputer - Parameters for
SimpleImputer(unfitted state). - Standard
Scaler - Parameters for StandardScaler (unfitted state).
- Truncated
Svd - Variance
Threshold - Parameters for
VarianceThresholdfeature selector (unfitted state).
Enums§
- BinStrategy
- Strategy for computing bin edges.
- Encode
Strategy - Encoding strategy for transformed output.
- Impute
Strategy - Strategy used to compute the fill value for missing (NaN) entries.
- Kpca
Kernel - Norm
Type - The type of norm used to normalize each sample (row).
- Output
Distribution - Output distribution for the quantile transformer.