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 finetune #10

Open
CDchenlin opened this issue Dec 12, 2023 · 3 comments
Open

How to finetune #10

CDchenlin opened this issue Dec 12, 2023 · 3 comments

Comments

@CDchenlin
Copy link

Great job! I’m wondering if there are any scripts available that I can use to fine-tune this model for my specific task.

@shenyunhang
Copy link
Owner

Hi @CDchenlin,
Our training script should be able to fine-tune your task with some preparation.

  1. register new data to detectron2
  2. maybe need to modify the data mapper
  3. add the data to model configs
  4. maybe need to modify the loss function

@skylning
Copy link

微调对设备有什么要求吗?一张rtx4090,只有少量数据可以不?

@shenyunhang
Copy link
Owner

我们是在V100上训练。
RTX4090的显存可能不太够,可以考虑冻结部分层,或者语言模型离线提取prompt,可以节省些内存。
我们也在训练ViT-Ti的小模型。

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

3 participants