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

new mllm eval #317

Merged
merged 19 commits into from
Nov 15, 2024
Merged

new mllm eval #317

merged 19 commits into from
Nov 15, 2024

Conversation

n1ck-guo
Copy link
Contributor

No description provided.

Signed-off-by: n1ck-guo <[email protected]>
auto_round/__main__.py Outdated Show resolved Hide resolved
auto_round/mllm/eval.py Outdated Show resolved Hide resolved
Signed-off-by: n1ck-guo <[email protected]>
Signed-off-by: n1ck-guo <[email protected]>
auto_round/__main__.py Outdated Show resolved Hide resolved
n1ck-guo and others added 4 commits November 13, 2024 03:30
@wenhuach21 wenhuach21 self-requested a review November 14, 2024 13:09
multiple formats exporting has been supported.

```bash
auto—round-mllm \
--model Qwen/Qwen2-VL-2B-Instruct\
--bits 4 \
--batch_size 1 \
--nsamples 128 \
--gradient_accumulate_steps 4 \
Copy link
Contributor

Choose a reason for hiding this comment

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

add a comment "experimental feature, default hyperparameters may be changed later"

auto_round/mllm/README.md Outdated Show resolved Hide resolved

|Model |Eval Lib |calibration dataset|quant nontext module|
|---------------|-----------|-------------------|--------------------|
|Qwen2-VL |vlmeval |pile/llava |✔ |
Copy link
Contributor

Choose a reason for hiding this comment

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

has the issue of "weight.dtype" in qwen already solved

from .eval import mllm_eval, lmms_eval
Copy link
Contributor

Choose a reason for hiding this comment

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

make sure that the tune/ inference work fine when these two eval library have not been installed

@n1ck-guo n1ck-guo merged commit ee816b3 into main Nov 15, 2024
1 of 10 checks passed
@n1ck-guo n1ck-guo deleted the hengguo/mllm_new_eval branch November 15, 2024 02:48
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

Successfully merging this pull request may close these issues.

2 participants