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.

Sign up for GitHub

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

add NormalizedConfig support qwen, baichuan, chatglm #1490

Open
wants to merge 8 commits into
base: main
Choose a base branch
from
Open

add NormalizedConfig support qwen, baichuan, chatglm #1490

Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
deda7e3
add models config
changwangss Oct 27, 2023
e0ed336
fix black
changwangss Oct 27, 2023
359b38e
Merge branch 'huggingface:main' into wangchang/optimum-config
changwangss Nov 10, 2023
a178662
Merge branch 'huggingface:main' into wangchang/optimum-config
changwangss Nov 22, 2023
fa8ad73
Update key name
changwangss Nov 22, 2023
927e947
update phi due to susnato/phi-1_dev
changwangss Dec 6, 2023
9e291db
Update normalized_config.py
changwangss Dec 20, 2023
e38d40a
Merge branch 'main' into wangchang/optimum-config
changwangss Feb 27, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions optimum/utils/normalized_config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -262,6 +262,11 @@ class NormalizedConfigManager:
"whisper": WhisperLikeNormalizedTextConfig,
"xlm-roberta": NormalizedTextConfig,
"yolos": NormalizedVisionConfig,
"mpt": MPTNormalizedTextConfig,
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

mpt is already in the list.

"baichuan": NormalizedTextConfig,
"qwen": NormalizedTextConfig,
"chatglm": NormalizedTextConfig.with_args(num_layers="num_layers"),
Copy link
Collaborator

@echarlaix echarlaix Oct 31, 2023 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

what about the vocab size, shouldn't it be padded_vocab_size for ChatGLM models ?
https://huggingface.co/THUDM/chatglm3-6b/blob/main/config.json#L32

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

chatglm has 3 models , I find chatglm is vocab_size, but chatglm2&3 is padded_vocab_size，Could you help me deal with this situation? @echarlaix
https://huggingface.co/THUDM/chatglm-6b/blob/main/config.json#L27

Comment on lines +266 to +268
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

qwen2 is now available in transformers.

baichuan and chatglm are not.


}

@classmethod
Expand Down