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

train from scratch? #4075

Closed
fangzhangmnm opened this issue Jun 5, 2024 · 5 comments
Closed

train from scratch? #4075

fangzhangmnm opened this issue Jun 5, 2024 · 5 comments
Labels
solved This problem has been already solved

Comments

@fangzhangmnm
Copy link

Is there a function to pretrain a base model from scratch? With tinystories dataset it is something can be done on a potato computer in a night.

hiyouga added a commit that referenced this issue Jun 5, 2024
@hiyouga
Copy link
Owner

hiyouga commented Jun 5, 2024

use train_from_scratch: true

@hiyouga hiyouga added the solved This problem has been already solved label Jun 5, 2024
@hiyouga hiyouga closed this as completed Jun 5, 2024
@fangzhangmnm
Copy link
Author

I'm sorry. here are two additional requirements

  1. can you add to the webui the train_from_scratch commands? What I'm thinking of is to pick a model structure, specify the number of heads and layers for example, and then train a model from scratch.
  2. can you add a tinystories and tinystories-chinese example into datasets?

thank you very much

@wangqh1996
Copy link

I'm sorry. here are two additional requirements

  1. can you add to the webui the train_from_scratch commands? What I'm thinking of is to pick a model structure, specify the number of heads and layers for example, and then train a model from scratch.
  2. can you add a tinystories and tinystories-chinese example into datasets?

thank you very much

同样需求这个功能,特别是第一个train_from_scratch,thanks。

@HuangZhen02
Copy link

can you add to the webui the train_from_scratch commands? What I'm thinking of is to pick a model structure, specify the number of heads and layers for example, and then train a model from scratch.

The same requirement.

@Mr-lonely0
Copy link

I'm sorry. here are two additional requirements

  1. can you add to the webui the train_from_scratch commands? What I'm thinking of is to pick a model structure, specify the number of heads and layers for example, and then train a model from scratch.
  2. can you add a tinystories and tinystories-chinese example into datasets?

thank you very much

同样需要第一个功能,可以自己定义一个全新的模型,完全的train from scratch, thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
solved This problem has been already solved
Projects
None yet
Development

No branches or pull requests

5 participants