facebookresearch / facebookresearch/sam3
Resuming Training From a Checkpoint
Open
- Dominant language
- Python
- Stars
- 11.7k
- Forks
- 1.8k
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
I am trying to finetune SAM3 on a different dataset, halfway it cuts off due to some issues, is there a way to resume training halfway
Contributor guide
Research direction
The issue does not name a training file, test, or entry point. Start by locating the finetuning workflow and its model-checkpoint handling, then determine whether interrupted training can restore the required state; done means a documented or tested way to resume finetuning from a checkpoint.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100