Releases: lucidrains/vector-quantize-pytorch
Releases · lucidrains/vector-quantize-pytorch
1.20.2
allow for an MLP implicitly parameterized codebook instead of just a …
1.20.1
still need comit loss from quantize to input to optimize linear proje…
1.20.0a
Merge pull request #172 from lucidrains/simvq SimVQ
1.19.5
move the rotation operation from rotation trick to a function for pot…
1.19.4
take care of a small pain point
1.19.3
only sync seed during training
1.19.2
Full Changelog: 1.19.1...1.19.2
1.19.1
Full Changelog: 1.18.8...1.19.1
1.19.0
Full Changelog: 1.18.8...1.19.0
1.18.8
Full Changelog: 1.18.7...1.18.8