[reland] Add int1 to int7 dtypes#137928
Conversation
Summary: Similar to #117208, we want to add int1 to int7 for edge use cases for weight quantization Test Plan: python test/test_quantization.py -k test_uint4_int4_dtype Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/137928
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 6cf187b with merge base a1b22e3 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
@jerryzh168 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
Are these to represent signed dtypes? (trying to understand the semantic difference between uint1 and int1 - or maybe it's just for simplicity in coding to have also "signed" dtypes?) |
Summary: Similar to #117208, we want to add int1 to int7 for edge use cases for weight quantization Test Plan: python test/test_quantization.py -k test_uint4_int4_dtype Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D64344944](https://our.internmc.facebook.com/intern/diff/D64344944) [ghstack-poisoned]
|
@jerryzh168 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
yes, these are for signed dtypes |
malfet
left a comment
There was a problem hiding this comment.
Sure, though please consider dropping Int1, not a very useful type imo
yeah I thought about this as well, seems like edge wants to add it still, cc @metascroy and @kimishpatel |
Unless int1 significantly adds maintenance burden, I think we should have it for completeness given we have uint1 as well. |
|
@pytorchbot rebase |
|
@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here |
|
Successfully rebased |
|
cc @malfet I will need to resolve the errors in the diff before landing btw, still fixing some of these |
Summary: Similar to #117208, we want to add int1 to int7 for edge use cases for weight quantization Test Plan: python test/test_quantization.py -k test_uint4_int4_dtype Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D64344944](https://our.internmc.facebook.com/intern/diff/D64344944) [ghstack-poisoned]
|
@jerryzh168 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
Summary: Similar to #117208, we want to add int1 to int7 for edge use cases for weight quantization Test Plan: python test/test_quantization.py -k test_uint4_int4_dtype Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D64344944](https://our.internmc.facebook.com/intern/diff/D64344944) [ghstack-poisoned]
|
@jerryzh168 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@pytorchbot merge |
Merge failedReason: This PR has internal changes and must be landed via Phabricator! Please try reimporting/rexporting the PR! Details for Dev Infra teamRaised by workflow job |
|
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Stack from ghstack (oldest at bottom):
Summary:
Similar to #117208, we want to add int1 to int7 for edge use cases
for weight quantization
Test Plan:
python test/test_quantization.py -k test_uint4_int4_dtype
Reviewers:
Subscribers:
Tasks:
Tags:
Differential Revision: D64344944