Skip to content

Testing implementations of singular value decomposition (SVD) in Python

Notifications You must be signed in to change notification settings

jackbowyer/svd-tests

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

svd-tests

Testing implementations of singular value decomposition (SVD) in Python.

The sklearn.decomposition.PCA module produces spurious results for input data arrays of dtype float32. Converting the same dataset to dtype float64 produces better PCA results (see pca_tests.py). Results for both datasets do not fall within the default tolerances of np.allclose()

The numpy.linalg.svd module performs SVD on both test datasets within the default tolerances of np.allclose() (see svd_tests.py).

Test data:
X_t_test_noise_float32.npy
X_t_test_noise_float64.npy

The datasets are too large to be uploaded directly. Find them here https://figshare.com/articles/dataset/test_data/25218920.

About

Testing implementations of singular value decomposition (SVD) in Python

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages