Skip to content

Tags: VertexCherry/vector-quantize-pytorch

Tags

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.

1.6.23

Toggle 1.6.23's commit message
loosen restriction around needing an separate optimizer for learnable…

… codebook

1.6.22

Toggle 1.6.22's commit message
address straight through lucidrains#59

1.6.21

Toggle 1.6.21's commit message
patch

1.6.20

Toggle 1.6.20's commit message
release inplace codebook optimizer, thanks to @theAdamColton

1.6.19

Toggle 1.6.19's commit message
fix mps detection

1.6.18

Toggle 1.6.18's commit message
fix memory leak issue, thanks to @fauzanardh

1.6.17

Toggle 1.6.17's commit message
detach the quantize commit target if not using learnable codebook

1.6.16

Toggle 1.6.16's commit message
fix inplace ema

1.6.15

Toggle 1.6.15's commit message
fix inplace ema