https://github.com/tensorly/tensorly
Revision 3cfd3e489b9d5f20027663c3fddbd7759c5831f4 authored by Aaron Meurer on 12 December 2018, 22:51:40 UTC, committed by Aaron Meurer on 12 December 2018, 22:51:40 UTC
This required adding a mask flag to kruskal_to_tensor() and kr() so that the mask could be absorbed into the calculation in a sparse friendly way. In order to be sparse friendly, the mask should be a sparse array with a fill value of 1 (True). In other words, the number of missing values should itself be sparse. The generic kr() algorithm was rewritten to use a generic outer product, and kruskal_to_tensor() now uses sum instead of dot (the NumPy kr() just multiplies the mask).
1 parent f0ef0a1
Tip revision: 3cfd3e489b9d5f20027663c3fddbd7759c5831f4 authored by Aaron Meurer on 12 December 2018, 22:51:40 UTC
Make the parafac mask support sparse friendly
Make the parafac mask support sparse friendly
Tip revision: 3cfd3e4
CHANGELOG.md
# 0.4.0
- Added TensorFlow backend with Eager execution
- Added CuPy backend
- PyTorch backend updated for PyTorch >= '0.4.0'
# 0.3.0
- Added PyTorch backend
# 0.2.0
- Added MXNet backend
# 0.1.4
- Added Robust Tensor PCA
Computing file changes ...