Public/automatic
1
0
Fork 0
mirror of https://github.com/vladmandic/automatic synced 2026-09-03 11:30:46 +02:00
Code Issues Packages Projects Releases Wiki Activity
14,589 Commits 3 Branches 34 Tags
dev
Commit Graph

2 Commits

Author SHA1 Message Date
CalamitousFelicitousness db4a345a1b fix(lora): set GLora dim so alpha scaling is applied 
Without self.dim, NetworkModule.calc_scale returns 1.0 even when
alpha is present, silently dropping the trained scaling. Use
w1b.shape[0] as the rank, matching the network_lora and
network_hada conventions.
 2026-05-09 22:04:42 +01:00
AI-Casanova cb561fa486 Major lora refactor: works on my machine edition 2024-11-23 21:57:03 -06:00