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
Overload vec::dequantize to eliminate rounding error for quantized sigmoid #114098
Conversation
…gmoid [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/114098
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit a7ba41b with merge base 5a96a42 (): This comment was automatically generated by Dr. CI and updates every 15 minutes. |
…gmoid ghstack-source-id: ae0fc902a817baaf4da14943587808f572bf78bc Pull Request resolved: #114098
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Maybe we need to revisit other callers and fix them too?
Yes. Maybe we can do it one by one if the same issue occurs for other ops so that we can minimize the impact. |
Hi @jerryzh168 @salilsdesai @kimishpatel @digantdesai @jianyuh Could you please review this PR? Thanks! |
@pytorchbot merge |
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 |
…gmoid (pytorch#114098) **Description** Fix pytorch#107030 Dequantize X by `(x_val - zp) * scale` instead of `x_val * scale + (-zp * scale)` to eliminate rounding error. Now this overload is used for sigmoid only. Performance impact: ![image](https://github.com/pytorch/pytorch/assets/12522207/655abd16-7d9d-4a9a-8c59-327ebf39157a) Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) **Test plan** `python test_quantization.py TestQuantizedOps.test_sigmoid_dequantize_rounding_error` Pull Request resolved: pytorch#114098 Approved by: https://github.com/jgong5, https://github.com/jerryzh168
Why were we doing |
I guess it was about making use of |
It is bound by mem access, not sure if |
Stack from ghstack (oldest at bottom):
Description
Fix #107030
Dequantize X by
(x_val - zp) * scale
instead ofx_val * scale + (-zp * scale)
to eliminate rounding error.Now this overload is used for sigmoid only.
Performance impact:
Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake)
Test plan
python test_quantization.py TestQuantizedOps.test_sigmoid_dequantize_rounding_error
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10