ROCm / AMDMIGraphX

AMD's graph optimization engine.
https://rocm.docs.amd.com/projects/AMDMIGraphX/en/latest/
MIT License
185 stars 86 forks source link

Small fixes for GQA #3537

Closed turneram closed 4 weeks ago

codecov[bot] commented 1 month ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 92.17%. Comparing base (275f854) to head (3a3707c). Report is 1 commits behind head on develop.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## develop #3537 +/- ## ======================================== Coverage 92.17% 92.17% ======================================== Files 512 512 Lines 21391 21391 ======================================== Hits 19718 19718 Misses 1673 1673 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

migraphx-bot commented 1 month ago
Test Batch Rate new
3a3707
Rate old
275f85
Diff Compare
torchvision-resnet50 64 3,260.56 3,260.33 0.01% :white_check_mark:
torchvision-resnet50_fp16 64 6,987.49 6,996.93 -0.13% :white_check_mark:
torchvision-densenet121 32 2,435.33 2,436.13 -0.03% :white_check_mark:
torchvision-densenet121_fp16 32 4,036.08 4,080.65 -1.09% :white_check_mark:
torchvision-inceptionv3 32 1,639.77 1,639.01 0.05% :white_check_mark:
torchvision-inceptionv3_fp16 32 2,762.50 2,759.60 0.11% :white_check_mark:
cadene-inceptionv4 16 775.73 775.88 -0.02% :white_check_mark:
cadene-resnext64x4 16 810.95 811.11 -0.02% :white_check_mark:
slim-mobilenet 64 7,536.28 7,539.27 -0.04% :white_check_mark:
slim-nasnetalarge 64 211.49 211.54 -0.03% :white_check_mark:
slim-resnet50v2 64 3,507.96 3,505.47 0.07% :white_check_mark:
bert-mrpc-onnx 8 1,151.43 1,151.85 -0.04% :white_check_mark:
bert-mrpc-tf 1 463.74 463.50 0.05% :white_check_mark:
pytorch-examples-wlang-gru 1 408.79 411.56 -0.67% :white_check_mark:
pytorch-examples-wlang-lstm 1 380.17 388.27 -2.09% :white_check_mark:
torchvision-resnet50_1 1 785.28 764.51 2.72% :white_check_mark:
cadene-dpn92_1 1 401.12 402.09 -0.24% :white_check_mark:
cadene-resnext101_1 1 373.98 381.69 -2.02% :white_check_mark:
onnx-taau-downsample 1 342.68 342.54 0.04% :white_check_mark:
dlrm-criteoterabyte 1 33.33 33.35 -0.06% :white_check_mark:
dlrm-criteoterabyte_fp16 1 52.71 52.74 -0.07% :white_check_mark:
agentmodel 1 8,016.80 8,186.92 -2.08% :white_check_mark:
unet_fp16 2 58.87 58.82 0.09% :white_check_mark:
resnet50v1_fp16 1 956.77 953.22 0.37% :white_check_mark:
resnet50v1_int8 1 1,020.16 1,003.11 1.70% :white_check_mark:
bert_base_cased_fp16 64 1,172.74 1,170.57 0.19% :white_check_mark:
bert_large_uncased_fp16 32 363.54 363.63 -0.03% :white_check_mark:
bert_large_fp16 1 198.76 201.70 -1.46% :white_check_mark:
distilgpt2_fp16 16 2,199.66 2,202.28 -0.12% :white_check_mark:
yolov5s 1 537.55 537.82 -0.05% :white_check_mark:
tinyllama 1 43.40 43.44 -0.09% :white_check_mark:
vicuna-fastchat 1 180.70 173.47 4.17% :high_brightness:
whisper-tiny-encoder 1 418.62 418.26 0.09% :white_check_mark:
whisper-tiny-decoder 1 428.79 436.61 -1.79% :white_check_mark:

Check results before merge :high_brightness:

migraphx-bot commented 1 month ago


     :white_check_mark: bert-mrpc-onnx: PASSED: MIGraphX meets tolerance
     :white_check_mark: bert-mrpc-tf: PASSED: MIGraphX meets tolerance
     :white_check_mark: pytorch-examples-wlang-gru: PASSED: MIGraphX meets tolerance
     :white_check_mark: pytorch-examples-wlang-lstm: PASSED: MIGraphX meets tolerance
     :white_check_mark: torchvision-resnet50_1: PASSED: MIGraphX meets tolerance
     :white_check_mark: cadene-dpn92_1: PASSED: MIGraphX meets tolerance
     :white_check_mark: cadene-resnext101_1: PASSED: MIGraphX meets tolerance
     :white_check_mark: dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance
     :white_check_mark: agentmodel: PASSED: MIGraphX meets tolerance
     :white_check_mark: unet: PASSED: MIGraphX meets tolerance
     :white_check_mark: resnet50v1: PASSED: MIGraphX meets tolerance
     :white_check_mark: bert_base_cased_fp16: PASSED: MIGraphX meets tolerance
:red_circle:bert_large_uncased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output

     :white_check_mark: bert_large: PASSED: MIGraphX meets tolerance
     :white_check_mark: yolov5s: PASSED: MIGraphX meets tolerance
     :white_check_mark: tinyllama: PASSED: MIGraphX meets tolerance
     :white_check_mark: vicuna-fastchat: PASSED: MIGraphX meets tolerance
     :white_check_mark: whisper-tiny-encoder: PASSED: MIGraphX meets tolerance
     :white_check_mark: whisper-tiny-decoder: PASSED: MIGraphX meets tolerance
     :white_check_mark: distilgpt2_fp16: PASSED: MIGraphX meets tolerance