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.

Sign up for GitHub

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

Jump to bottom

Normalize standard input shapes to hipBLASLt gemms #3712

Open
wants to merge 1 commit into
base: develop
Choose a base branch
from
Open

Normalize standard input shapes to hipBLASLt gemms #3712

wants to merge 1 commit into from

Conversation

ahsan-ca
Copy link
Contributor

This change normalizes standard input shapes to hipBLASLt gemms.

@ahsan-ca ahsan-ca added the bugfix Fixes a bug found in the code. label Dec 13, 2024
@ahsan-ca ahsan-ca self-assigned this Dec 13, 2024
@ahsan-ca ahsan-ca requested a review from causten as a code owner December 13, 2024 17:22
@ahsan-ca ahsan-ca linked an issue Dec 13, 2024 that may be closed by this pull request
Error in fuse_ops for hip_gemm_pointwise #3690
Open
@TedThemistokleous TedThemistokleous added the simple small or simple changes label Dec 13, 2024
@migraphx-bot
Copy link
Collaborator

Test Batch Rate new
b216f5		 Rate old
79a256		 Diff Compare
torchvision-resnet50 64 3,255.71 3,254.68 0.03% ✅
torchvision-resnet50_fp16 64 6,988.36 6,988.98 -0.01% ✅
torchvision-densenet121 32 2,429.51 2,435.20 -0.23% ✅
torchvision-densenet121_fp16 32 4,089.98 4,089.35 0.02% ✅
torchvision-inceptionv3 32 1,628.47 1,629.15 -0.04% ✅
torchvision-inceptionv3_fp16 32 2,743.25 2,750.39 -0.26% ✅
cadene-inceptionv4 16 764.64 765.66 -0.13% ✅
cadene-resnext64x4 16 813.24 812.33 0.11% ✅
slim-mobilenet 64 7,466.35 7,465.36 0.01% ✅
slim-nasnetalarge 64 208.97 209.02 -0.03% ✅
slim-resnet50v2 64 3,439.86 3,439.27 0.02% ✅
bert-mrpc-onnx 8 1,149.40 1,145.81 0.31% ✅
bert-mrpc-tf 1 479.28 466.39 2.76% ✅
pytorch-examples-wlang-gru 1 410.98 421.94 -2.60% ✅
pytorch-examples-wlang-lstm 1 383.33 381.01 0.61% ✅
torchvision-resnet50_1 1 785.10 763.70 2.80% ✅
cadene-dpn92_1 1 402.48 434.24 -7.31% 🔴
cadene-resnext101_1 1 383.38 383.59 -0.05% ✅
onnx-taau-downsample 1 345.17 346.01 -0.24% ✅
dlrm-criteoterabyte 1 33.33 33.33 0.00% ✅
dlrm-criteoterabyte_fp16 1 52.73 52.73 0.01% ✅
agentmodel 1 10,304.55 8,229.23 25.22% 🔆
unet_fp16 2 58.89 58.93 -0.06% ✅
resnet50v1_fp16 1 970.37 1,025.60 -5.39% 🔴
resnet50v1_int8 1 1,003.67 1,052.08 -4.60% 🔴
bert_base_cased_fp16 64 1,168.61 1,169.64 -0.09% ✅
bert_large_uncased_fp16 32 363.15 363.34 -0.05% ✅
bert_large_fp16 1 200.57 198.80 0.89% ✅
distilgpt2_fp16 16 2,194.21 2,201.30 -0.32% ✅
yolov5s 1 530.60 529.55 0.20% ✅
tinyllama 1 43.35 43.36 -0.02% ✅
vicuna-fastchat 1 175.42 170.51 2.88% ✅
whisper-tiny-encoder 1 417.97 417.88 0.02% ✅
whisper-tiny-decoder 1 429.58 425.43 0.98% ✅

This build is not recommended to merge 🔴

@migraphx-bot
Copy link
Collaborator


     ✅ bert-mrpc-onnx: PASSED: MIGraphX meets tolerance

     ✅ bert-mrpc-tf: PASSED: MIGraphX meets tolerance

     ✅ pytorch-examples-wlang-gru: PASSED: MIGraphX meets tolerance

     ✅ pytorch-examples-wlang-lstm: PASSED: MIGraphX meets tolerance

     ✅ torchvision-resnet50_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-dpn92_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-resnext101_1: PASSED: MIGraphX meets tolerance

     ✅ dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance

     ✅ agentmodel: PASSED: MIGraphX meets tolerance

     ✅ unet: PASSED: MIGraphX meets tolerance

     ✅ resnet50v1: PASSED: MIGraphX meets tolerance

     ✅ bert_base_cased_fp16: PASSED: MIGraphX meets tolerance

🔴bert_large_uncased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output


     ✅ bert_large: PASSED: MIGraphX meets tolerance

     ✅ yolov5s: PASSED: MIGraphX meets tolerance

     ✅ tinyllama: PASSED: MIGraphX meets tolerance

     ✅ vicuna-fastchat: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-encoder: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-decoder: PASSED: MIGraphX meets tolerance

     ✅ distilgpt2_fp16: PASSED: MIGraphX meets tolerance

@codecov Codecov
Copy link

codecov bot commented Dec 13, 2024

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.23%. Comparing base (79a2561) to head (b216f54).
Report is 1 commits behind head on develop.

Additional details and impacted files 
@@           Coverage Diff            @@
##           develop    #3712   +/-   ##
========================================
  Coverage    92.23%   92.23%           
========================================
  Files          514      514           
  Lines        21746    21746           
========================================
  Hits         20057    20057           
  Misses        1689     1689

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@TedThemistokleous TedThemistokleous TedThemistokleous approved these changes

@shivadbhavsar shivadbhavsar shivadbhavsar approved these changes

@causten causten Awaiting requested review from causten causten is a code owner

@pfultz2 pfultz2 Awaiting requested review from pfultz2

@CharlieL7 CharlieL7 Awaiting requested review from CharlieL7

Assignees

@ahsan-ca ahsan-ca

Labels
bugfix Fixes a bug found in the code. simple small or simple changes
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Error in fuse_ops for hip_gemm_pointwise
6 participants
@ahsan-ca @migraphx-bot @shivadbhavsar @TedThemistokleous @pfultz2 @CharlieL7