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

Save request outputs and add eval accuracy support #8

Merged
merged 6 commits into from
Mar 8, 2024

Conversation

FanhaiLu1
Copy link
Collaborator

This PR add two features:

  • Save request outputs to a local file path with Json format, it is additional function in benchmark_serving, please use --save-request-outputs to enable it.
  • Measure accuracy of inference generated text accuracy.

Below are command line examples to use above features:

python JetStream/benchmarks/benchmark_serving.py \
--tokenizer /home/{username}/maxtext/assets/tokenizer \
--num-prompts 10  \
--dataset ~/data/ShareGPT_V3_unfiltered_cleaned_split.json \
--save-request-outputs 

python JetStream/benchmarks/eval_accuracy.py

Copy link
Collaborator

@JoeZijunZhou JoeZijunZhou left a comment

Choose a reason for hiding this comment

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

LGTM! Do you mind adding a README under /benchmarks to introduce the eval script, thanks!

requirements.in Outdated Show resolved Hide resolved
benchmarks/README.md Outdated Show resolved Hide resolved
benchmarks/README.md Outdated Show resolved Hide resolved
@JoeZijunZhou JoeZijunZhou merged commit 41ad033 into AI-Hypercomputer:main Mar 8, 2024
3 checks passed
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