Skip to content
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

[AutoBump] Merge with 911e7235 (May 13) (36) #269

Merged
merged 6 commits into from
Sep 4, 2024

Conversation

mgehre-amd
Copy link
Collaborator

No description provided.

qingyunqu and others added 6 commits May 11, 2024 15:33
llvm#3328)

…NORMAL

* split lowering of uniform, randn, normal from Basic.cpp into Rng.cpp
This change enables more customization with operand quantization, and
generalizes the patterns QuantizeOperands and QuantizeTransposeOperands
to QuantizeOperandsPastCommutingOps.

This allows for passing quantization through operations which are
functionally unaffected by quantization, such as view-like ops. The
purpose of this change is to address a myriad of quantization issues
seen in quantized onnx models that have some reshape-like operations
sandwiched in between a dequant and something like a matmul (whose other
operand is immediately quantizable).
After running the model tests in SHARK-TestSuite, I noticed a few model
failures due to half-fusion.

Notably, RDN_pytorch_vaiq_int8 had a depth=5 convolution chain with
multiple AtenViewOp's.
@mgehre-amd mgehre-amd requested a review from cferry-AMD August 27, 2024 13:10
Base automatically changed from bump_to_00efec0b to feature/backport_ea1_ops September 3, 2024 20:10
@mgehre-amd mgehre-amd merged commit c17c667 into feature/backport_ea1_ops Sep 4, 2024
4 checks passed
@mgehre-amd mgehre-amd deleted the bump_to_911e7235 branch September 4, 2024 05:01
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants