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

mxr to onnx #3682

Merged
merged 35 commits into from
Jan 13, 2025
Merged

mxr to onnx #3682

merged 35 commits into from
Jan 13, 2025

Conversation

richagadgil
Copy link
Contributor

@richagadgil richagadgil commented Dec 4, 2024

Changes to migraphx_py to expose information about instruction_ref to engineer ONNX file.

Copy link

codecov bot commented Dec 4, 2024

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.16%. Comparing base (6d02806) to head (a72564f).
Report is 2 commits behind head on develop.

Additional details and impacted files
@@           Coverage Diff            @@
##           develop    #3682   +/-   ##
========================================
  Coverage    92.16%   92.16%           
========================================
  Files          515      515           
  Lines        21978    21978           
========================================
  Hits         20256    20256           
  Misses        1722     1722           

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

@richagadgil richagadgil self-assigned this Dec 4, 2024
@richagadgil richagadgil marked this pull request as ready for review December 6, 2024 23:58
@richagadgil richagadgil requested a review from causten as a code owner December 6, 2024 23:58
src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
tools/check_stamped.py Outdated Show resolved Hide resolved
src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
@migraphx-bot
Copy link
Collaborator

Test Batch Rate new
56feec
Rate old
f56b1b
Diff Compare
torchvision-resnet50 64 3,254.45 3,255.67 -0.04%
torchvision-resnet50_fp16 64 6,981.91 6,983.58 -0.02%
torchvision-densenet121 32 2,430.24 2,431.65 -0.06%
torchvision-densenet121_fp16 32 4,075.77 4,074.03 0.04%
torchvision-inceptionv3 32 1,627.80 1,628.91 -0.07%
torchvision-inceptionv3_fp16 32 2,743.53 2,746.14 -0.10%
cadene-inceptionv4 16 764.69 764.54 0.02%
cadene-resnext64x4 16 812.54 813.45 -0.11%
slim-mobilenet 64 7,463.84 7,469.86 -0.08%
slim-nasnetalarge 64 209.03 209.05 -0.01%
slim-resnet50v2 64 3,438.14 3,440.80 -0.08%
bert-mrpc-onnx 8 1,146.66 1,145.22 0.13%
bert-mrpc-tf 1 475.75 476.55 -0.17%
pytorch-examples-wlang-gru 1 455.11 422.09 7.82% 🔆
pytorch-examples-wlang-lstm 1 389.97 394.94 -1.26%
torchvision-resnet50_1 1 786.65 769.40 2.24%
cadene-dpn92_1 1 398.53 398.97 -0.11%
cadene-resnext101_1 1 382.07 383.77 -0.44%
onnx-taau-downsample 1 346.25 345.22 0.30%
dlrm-criteoterabyte 1 33.30 33.32 -0.08%
dlrm-criteoterabyte_fp16 1 52.71 52.72 -0.02%
agentmodel 1 8,254.71 8,109.71 1.79%
unet_fp16 2 58.81 58.87 -0.11%
resnet50v1_fp16 1 928.26 930.54 -0.25%
resnet50v1_int8 1 1,007.23 1,002.61 0.46%
bert_base_cased_fp16 64 1,169.68 1,168.63 0.09%
bert_large_uncased_fp16 32 363.31 363.25 0.02%
bert_large_fp16 1 199.68 198.22 0.74%
distilgpt2_fp16 16 2,195.14 2,197.80 -0.12%
yolov5s 1 523.87 532.80 -1.68%
tinyllama 1 43.39 43.43 -0.10%
vicuna-fastchat 1 178.58 174.20 2.52%
whisper-tiny-encoder 1 418.40 418.04 0.09%
whisper-tiny-decoder 1 427.93 433.15 -1.21%

Check results before 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

src/py/migraphx_py.cpp Outdated Show resolved Hide resolved
@causten causten merged commit c87d983 into develop Jan 13, 2025
43 of 45 checks passed
@causten causten deleted the mxr_to_onnx branch January 13, 2025 20:07
causten pushed a commit that referenced this pull request Jan 24, 2025
torch-migraphx sometimes expects default python object comparison in if-else trees when working with instruction_refs. This should keep the correct behavior added by #3682 but add fallback for when not comparing 2 instruction_refs
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.

None yet

5 participants