extraction/

Feature extraction and dimensionality reduction methods.

Feature extraction transforms the original feature space into a new, lower-dimensional space that preserves as much information as possible. This is essential for visualization, noise reduction, and improving computational efficiency.

Algorithms

PCAExtractor: Projects data onto the axes of maximum variance.
RandomProjectionExtractor: Projects data onto a random subspace

(preserves distances).

SparseRandomProjectionExtractor: Optimized random projection

using sparse matrices.

Reducing dimensionality with PCA:

python

>>> from tuiml.features.extraction import PCAExtractor
>>> import numpy as np
>>> X = np.random.randn(50, 20)
>>> pca = PCAExtractor(n_components=5)
>>> X_reduced = pca.fit_transform(X)

Using Random Projection for fast reduction:

python

>>> from tuiml.features.extraction import RandomProjectionExtractor
>>> rp = RandomProjectionExtractor(n_components='auto')
>>> X_rp = rp.fit_transform(X)

Modules

pca

Principal Component Analysis (PCAExtractor) for feature extraction....

1 class · 0 functions

random_projection

Random Projection for dimensionality reduction....

2 classes · 0 functions