microsoft / microsoft/dp-transformers
How can I load model after lora fine tune？
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 149
- Forks
- 33
- Avg merge
- 22h 6m
- Merged PRs (30d)
- 1
Description
I do actually as you said that I add --save_safetensors=False to training.
#52
However the checkpoint file is like
There is no adapter.bin and the model and I actually use lora to fine tune
When I want to load that model it says that
OSError: examples/nlg-reddit/sample-level-dp/finetune_dp_private/checkpoint-2250 does not appear to have a file named config.json. Checkout 'https://huggingface.co/examples/nlg-reddit/sample-level-dp/finetune_dp_private/checkpoint-2250/tree/None' for available files.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the checkpoint-2250 contents shown in the issue and review the context from issue #52. Compare the saved files with the files required by the reported loading path, including config.json and the expected adapter file. Done means the LoRA-fine-tuned checkpoint can be loaded without the reported OSError.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100