Update benchmark to run openorca dataset #21
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
morgandu
requested review from
JoeZijunZhou,
patemotter,
FanhaiLu1 and
vipannalla
and removed request for
vipannalla
morgandu
force-pushed
the
mor--golden-openorca-dataset
branch
2 times, most recently
from
5f5540f
to
54194d3
Compare
FanhaiLu1
approved these changes
Mar 29, 2024
JoeZijunZhou
approved these changes
Mar 29, 2024
benchmarks/benchmark_serving.py
Outdated
| @@ -42,7 +42,7 @@ | |||
| python -m benchmarks.benchmark_serving \ | |||
| --request-rate 1 | |||
|
|
|||
| e2e example: python3 benchmark_serving.py --tokenizer /home/rwitten/maxtext/assets/tokenizer --num-prompts 100 --dataset ~/ShareGPT_V3_unfiltered_cleaned_split.json | |||
| e2e example: python3 benchmark_serving.py --tokenizer /home/rwitten/maxtext/assets/tokenizer --num-prompts 100 --dataset ~/ShareGPT_V3_unfiltered_cleaned_split.json | |||
There was a problem hiding this comment.
We can also update the example here with your change. And also the README in /benchmark.
morgandu
force-pushed
the
mor--golden-openorca-dataset
branch
from
54194d3
to
7477e24
Compare
morgandu
force-pushed
the
mor--golden-openorca-dataset
branch
from
7477e24
to
71c111c
Compare
morgandu
force-pushed
the
mor--golden-openorca-dataset
branch
from
e9e3ac7
to
4f41058
Compare
Comment on lines
+206
to
+211
| # Tokenize the prompts and completions. | ||
| prompts = dataset["prompts"] | ||
| outputs = dataset["results"] | ||
| n = len(prompts) | ||
| prompt_token_ids = tokenizer.tokenize(prompts) | ||
| output_token_ids = tokenizer.tokenize(outputs) |
There was a problem hiding this comment.
I feel we could extract this part out as a func for different dataset, and the rest are identical and thus we could keep them in the sample_request function?
There was a problem hiding this comment.
Discussed offline, there are some existing data processing that may not be necessary. Will revisit and refactor the data preprocessing part if needed.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.