Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

$model_path in the video evaluation script. #66

Open
gauravsh0812 opened this issue Dec 11, 2024 · 1 comment
Open

$model_path in the video evaluation script. #66

gauravsh0812 opened this issue Dec 11, 2024 · 1 comment

Comments

@gauravsh0812
Copy link

I have trained and fine-tuned the ChatUniVi v1.5 model. For training, I set the parameter --model_name_or_path to /data/gs/weights/vicuna-7b-v1.5 and used my custom datasets specified with --dataset as "New" or "FINETUNE."

Now, I am trying to run the evaluation script as described in the [Chat-UniVi documentation](https://github.com/PKU-YuanGroup/Chat-UniVi/blob/main/TRAIN_AND_VALIDATE.md#step-1-load-the-model-to-generate-results-1).

During training, I set the --output_dir to outputs/, where I obtained the mm_projection.bin file along with other .bin and .model files.

My question is: For the evaluation script, what should I set as the $model_path? Should it point to vicuna-7b-v1.5, or should it refer to the files in the outputs/ directory?

@jpthu17
Copy link
Member

jpthu17 commented Dec 11, 2024

The $model_path should point to the files in the outputs/ directory.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants