conheaven comments

Repositories
Issues
Comments

Results 5 comments of


                                            conheaven

[Usage] How to load fine-tuned lora checkpoint for CLI inference

> > @Nomiluks could you produce a slightly more full example than what is found in https://github.com/haotian-liu/LLaVA/blob/main/docs/Finetune_Custom_Data.md ? It looks like you were successful at training a LoRA and saving...

[Usage] How to load fine-tuned lora checkpoint for CLI inference

> Hi @conheaven, > > I got the same error, did you find a solution? sorry, i didn't find any feasible solution

[Question] inference after first-stage pretraining

> Hi @yinincanada , 你好, > > I happen to have the same question as yours. I find that `load_pretrained_model` from `llava/model` is able to load the model with a...

refcoco_eval.json

thx

[Question] after finetune i run cli.py

> try to upgrade transformers my transformers is 4.37.2, should i upgrade it？