-
Notifications
You must be signed in to change notification settings - Fork 22.3k
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
[inductor][cpu]transformers models static/dynamic quant performance/accuracy crash in 2024-06-17 nightly release #128933
Comments
Hi @ezyang, could you kindly help to take a look? Prepare the script to reproduce this issue: https://gist.github.com/leslie-fang-intel/696041fa7e7352ecb985b04a5e1188de and it starts to fail since 2229884 Here are the version of transformer I used |
vision_maskrcnn and detectron2_fcos_r_50_fpn AMP/float32 single/multiple thread static/dynamic shape default/cpp wrapper meet Repro:
|
Further looking into the why we can't statically known
|
But that guard sounds reasonable to me, no? It's asking that Looking at the error in #128933 (comment), it might suggest that our @ezyang this shows a larger issue that's lurking with the inf treatment: Our bounds are inclusive... unless one of the ends is |
Yean, any suggestions for how to further debug why the guard failed? I am just listing out the difference before and after this commit and maybe there is another potential issue which fails the guard :( ---------- Update for why the new added guard fail ------------
|
This is sort of expected, but what we probably can do is make the constraint violation error more tolerant for this case. The big question I had to answer in #127693 was what I should do if there legitimately was different behavior when s0 == sys.maxsize. Previously, I simply assumed this couldn't happen, because who makes sys.maxsize type tensors. But with int_oo modeling, "just assuming" it doesn't happen is not so convenient. But it's also not a big deal, you just get a guard testing that the int is not maxsize, nbd. Except for the constraint stuff. The constraint violation says "if there is ANY guard, error out". But we can probably make it softer, e.g., a guard that the value is not maxsize shouldn't trigger this. |
🐛 Describe the bug
Versions
SW info
Repro:
Suspected guilty commit: 2229884
text-classification+albert-base-v1-static-quant-accuracy-crash_guilty_commit.log
cc @ezyang @anijain2305 @chauhang @penguinwu @WeizhuoZhang-intel @chuanqi129
The text was updated successfully, but these errors were encountered: