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

add npu_group_size for transformers_int4_npu_win in all-in-one benchmark api #12316

Merged
merged 2 commits into from
Nov 1, 2024

Conversation

ch1y0q
Copy link
Contributor

@ch1y0q ch1y0q commented Nov 1, 2024

small bugfix

Description

1. Why the change?

2. User API changes

3. Summary of the change

4. How to test?

  • N/A
  • Unit test: Please manually trigger the PR Validation here by inputting the PR number (e.g., 1234). And paste your action link here once it has been successfully finished.
  • Application test
  • Document test
  • ...

5. New dependencies

  • New Python dependencies
    - Dependency1
    - Dependency2
    - ...
  • New Java/Scala dependencies and their license
    - Dependency1 and license1
    - Dependency2 and license2
    - ...

@@ -2193,6 +2198,8 @@ def run_pipeline_parallel_gpu(repo_id,
optimize_model = conf['optimize_model']
if 'group_size' in conf:
group_size = conf['group_size']
Copy link
Contributor

Choose a reason for hiding this comment

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

remove above if.

Copy link
Contributor

@rnwang04 rnwang04 left a comment

Choose a reason for hiding this comment

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

others LGTM.

@rnwang04 rnwang04 changed the title add npu_group_size for transformers_int4_npu_win add npu_group_size for transformers_int4_npu_win in all-in-one benchmark api Nov 1, 2024
@@ -214,7 +214,8 @@ def run_model(repo_id, test_api, in_out_pairs, local_model_hub=None, warm_up=1,
round(result[in_out_pair][-1][5], 2),
result[in_out_pair][-1][6] if any(keyword in test_api for keyword in ['int4_gpu', 'int4_fp16_gpu_win', 'int4_loadlowbit_gpu', 'int4_fp16_loadlowbit_gpu', 'fp16_gpu', 'deepspeed_optimize_model_gpu']) and not lookahead else 'N/A',
streaming if 'win' in test_api else 'N/A',
use_fp16_torch_dtype if 'pipeline_parallel_gpu' in test_api else 'N/A'],
use_fp16_torch_dtype if 'pipeline_parallel_gpu' in test_api else 'N/A',
group_size],
Copy link
Contributor

Choose a reason for hiding this comment

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

Should we make group_size value "N/A" for other test_api? Maybe confusing

Copy link
Contributor

Choose a reason for hiding this comment

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

Good point, we will fix this in next PR . 😊

@rnwang04 rnwang04 merged commit 48123af into intel-analytics:main Nov 1, 2024
1 check passed
@ch1y0q ch1y0q deleted the npu_group_size branch November 4, 2024 02:09
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.

4 participants