-
Notifications
You must be signed in to change notification settings - Fork 3
Pull requests: EvolvingLMMs-Lab/OneVision-Encoder
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
perf: zero_grad(set_to_none=True) and reduce checkpoint I/O
#83
opened Feb 7, 2026 by
Luodian
Loading…
fix: correct gradient accumulation off-by-one and lr_scheduler over-stepping
#82
opened Feb 7, 2026 by
Luodian
Loading…
perf: disable find_unused_parameters for faster DDP training
#68
opened Jan 10, 2026 by
Luodian
Loading…
fix: use closed-form LR calculation to fix polynomial decay formula bug
#67
opened Jan 10, 2026 by
Luodian
Loading…
fix: filter weight decay for LayerNorm, biases, and special tokens
#66
opened Jan 10, 2026 by
Luodian
Loading…
fix: save and restore AdamW optimizer state for proper training resume
#65
opened Jan 10, 2026 by
Luodian
Loading…
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.