fixing scripts #395
Merged
fixing scripts #395
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/395
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 3fa636f with merge base cb3bd8c ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
Summary: a few bugfixes for scripts 1) convert_hf_checkpoint.py had a gpt-fast dependency that wasn't caught due to it being in the path 2) eval.py had a bug due to the switch to aqt apis 3) generate.py had a bug due to the deprecation of the old quant apis Test Plan: python eval.py python eval.py -q int8dq --compile --limit 2 python eval.py -q int8wo --compile --limit 2 python eval.py -q int4wo-64 --compile --limit 2 python eval.py -q int4wo-64-gptq --compile sh benchmarks.sh (going to add the output results once they finish Reviewers: Subscribers: Tasks: Tags:
HDCharles
force-pushed
the
071_scripts_fix
branch
from
be3f668
to
3fa636f
Compare
|
@vayuda this PR still had testing in progress and has some issues, please review->let owner merge if possible, rather than review->merge, just so any final changes can be made. |
dbyoung18
pushed a commit
to dbyoung18/ao
that referenced
this pull request
Jul 31, 2024
Summary: a few bugfixes for scripts 1) convert_hf_checkpoint.py had a gpt-fast dependency that wasn't caught due to it being in the path 2) eval.py had a bug due to the switch to aqt apis 3) generate.py had a bug due to the deprecation of the old quant apis Test Plan: python eval.py python eval.py -q int8dq --compile --limit 2 python eval.py -q int8wo --compile --limit 2 python eval.py -q int4wo-64 --compile --limit 2 python eval.py -q int4wo-64-gptq --compile sh benchmarks.sh (going to add the output results once they finish Reviewers: Subscribers: Tasks: Tags:
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.
Summary:
a few bugfixes for scripts
convert_hf_checkpoint.py had a gpt-fast dependency that wasn't caught
due to it being in the path. (caught here: Issues with the gptfast setup scripts #393)
eval.py had a bug due to the switch to aqt apis
Test Plan:
python eval.py
wikitext: {'word_perplexity,none': 12.212490471702079, 'word_perplexity_stderr,none': 'N/A', 'byte_perplexity,none': 1.59675331009031, 'byte_perplexity_stderr,none': 'N/A', 'bits_per_byte,none': 0.6751414412399839, 'bits_per_byte_stderr,none': 'N/A', 'alias': 'wikitext'}
python eval.py -q int8dq --compile
wikitext: {'word_perplexity,none': 12.253535794972834, 'word_perplexity_stderr,none': 'N/A', 'byte_perplexity,none': 1.597755518273052, 'byte_perplexity_stderr,none': 'N/A', 'bits_per_byte,none': 0.6760466701188526, 'bits_per_byte_stderr,none': 'N/A', 'alias': 'wikitext'}
python eval.py -q int8wo --compile
wikitext: {'word_perplexity,none': 12.210618271188853, 'word_perplexity_stderr,none': 'N/A', 'byte_perplexity,none': 1.5967075310980392, 'byte_perplexity_stderr,none': 'N/A', 'bits_per_byte,none': 0.6751000785125764, 'bits_per_byte_stderr,none': 'N/A', 'alias': 'wikitext'}
python eval.py -q int4wo-64 --compile
python eval.py -q int4wo-64-gptq --compile
Reviewers:
Subscribers:
Tasks:
Tags: