ROCm / AMDMIGraphX

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

Windows fixes for generic_float #3607

Closed pfultz2 closed 5 days ago

pfultz2 commented 1 week ago

This is similar to #3599, but passes the tests. There is no macro, and all_ones function doesn't change, so there is less code changes here.

codecov[bot] commented 1 week ago

Codecov Report

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

Project coverage is 92.17%. Comparing base (f5df004) to head (1015d51). Report is 2 commits behind head on develop.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## develop #3607 +/- ## ======================================== Coverage 92.17% 92.17% ======================================== Files 513 513 Lines 21536 21537 +1 ======================================== + Hits 19851 19852 +1 Misses 1685 1685 ```

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

migraphx-bot commented 6 days ago
Test Batch Rate new
1015d5
Rate old
f5df00
Diff Compare
torchvision-resnet50 64 3,260.13 3,258.22 0.06% :white_check_mark:
torchvision-resnet50_fp16 64 6,985.56 6,994.86 -0.13% :white_check_mark:
torchvision-densenet121 32 2,435.89 2,438.08 -0.09% :white_check_mark:
torchvision-densenet121_fp16 32 4,104.17 4,082.99 0.52% :white_check_mark:
torchvision-inceptionv3 32 1,637.41 1,637.00 0.03% :white_check_mark:
torchvision-inceptionv3_fp16 32 2,766.04 2,762.55 0.13% :white_check_mark:
cadene-inceptionv4 16 776.73 776.12 0.08% :white_check_mark:
cadene-resnext64x4 16 810.66 811.97 -0.16% :white_check_mark:
slim-mobilenet 64 7,541.02 7,534.11 0.09% :white_check_mark:
slim-nasnetalarge 64 211.43 211.46 -0.01% :white_check_mark:
slim-resnet50v2 64 3,501.82 3,503.66 -0.05% :white_check_mark:
bert-mrpc-onnx 8 1,151.06 1,150.99 0.01% :white_check_mark:
bert-mrpc-tf 1 466.22 494.60 -5.74% :red_circle:
pytorch-examples-wlang-gru 1 528.74 428.98 23.25% :high_brightness:
pytorch-examples-wlang-lstm 1 406.37 385.66 5.37% :high_brightness:
torchvision-resnet50_1 1 784.50 766.70 2.32% :white_check_mark:
cadene-dpn92_1 1 408.99 433.66 -5.69% :red_circle:
cadene-resnext101_1 1 383.37 383.99 -0.16% :white_check_mark:
onnx-taau-downsample 1 342.55 343.15 -0.17% :white_check_mark:
dlrm-criteoterabyte 1 33.32 33.33 -0.03% :white_check_mark:
dlrm-criteoterabyte_fp16 1 52.74 52.72 0.03% :white_check_mark:
agentmodel 1 8,310.41 8,297.10 0.16% :white_check_mark:
unet_fp16 2 58.88 58.77 0.19% :white_check_mark:
resnet50v1_fp16 1 1,044.55 948.38 10.14% :high_brightness:
resnet50v1_int8 1 997.70 1,046.77 -4.69% :red_circle:
bert_base_cased_fp16 64 1,172.03 1,171.13 0.08% :white_check_mark:
bert_large_uncased_fp16 32 363.86 363.68 0.05% :white_check_mark:
bert_large_fp16 1 198.34 197.66 0.34% :white_check_mark:
distilgpt2_fp16 16 2,201.55 2,203.57 -0.09% :white_check_mark:
yolov5s 1 521.20 549.13 -5.09% :red_circle:
tinyllama 1 43.47 43.43 0.09% :white_check_mark:
vicuna-fastchat 1 172.45 169.54 1.71% :white_check_mark:
whisper-tiny-encoder 1 418.19 418.17 0.00% :white_check_mark:
whisper-tiny-decoder 1 426.78 428.69 -0.44% :white_check_mark:

This build is not recommended to merge :red_circle:

migraphx-bot commented 6 days 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