Skip to content

imn167/Randomized-linear-algebra

Repository files navigation

SSPM-Project

Data compression by randomized algorithms. Codes in Python used in the project. Common project with https://github.com/mariecmpn/SSPM-project.git

Codes of the project:

rsvd.py: functions used for the RSVD; rsvd_tests.ipynb: tests for the RSVD using the rsvd.py file; rcur.py: code used for the CUR decomposition tests; tensor_tucker.py: functions used for the Tucker decomposition; hosvd_st-hosvd.py: tests for the HOSVD and ST_HOSVD; twopass.py: test for the Two-Pass Sketch method; onepass.py: test for the One-Pass Sketch method. Images used for the tests:

image.png: image used for the HOSVD and ST-HOSVD; image2.webp: image used for the Two-Pass and One_Pass algorithms. Codes were made using the algorithms from the following papers and codes:

LOW-RANK TUCKER APPROXIMATION OF A TENSOR FROM STREAMING DATA, from Yiming Sun, Yang Guo, Charlene Luo, Joel Tropp, and Madeleine Udell. FINDING STRUCTURE WITH RANDOMNESS: PROBABILISTIC ALGORITHMS FOR CONSTRUCTING APPROXIMATE MATRIX DECOMPOSITIONS , from Nathan Halko, Per-Gunnar Martinsson, and Joel A. Tropp. https://github.com/udellgroup/tensorsketch

About

Project on Randomized Linear Algebra. Approximation of tensor by low-rank methods

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published