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

How to training again GOT-OCR2.0 in Stage 2, Stage 3 #238

Open
C0NGTRI123 opened this issue Dec 6, 2024 · 1 comment
Open

How to training again GOT-OCR2.0 in Stage 2, Stage 3 #238

C0NGTRI123 opened this issue Dec 6, 2024 · 1 comment

Comments

@C0NGTRI123
Copy link

Hello, I want to ask again to confirm. After I training again in stage 1 in Vary_tiny, they return me the weight .pth. The weight from Huggingface you provide, it may be the weight from pretrain. Because I read your paper after stage 1, you get vision encoder continue training with Qwen-0.5B in stage 2 and remove old LLM(OPT-125m) in stage 1. But I don't see this code. Can you explain me that

@Ucas-HaoranWei
Copy link
Owner

You should extract the encoder weights in stage-1 .pth, and load it to the next stage models.

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