search

migraphx-benchmark / AMDMIGraphX

AMD's graph optimization engine.
https://rocmsoftwareplatform.github.io/AMDMIGraphX/doc/html/
MIT License
0 stars 1 forks source link

Test new model implementation #44

Closed IgorMirosavljevicHTEC closed 1 year ago

migraphx-bot commented 1 year ago
Test Batch Rate new
0b0a6d		 Rate old
a63ee2		 Diff Compare
torchvision-resnet50 64 2,266.74 2,264.84 0.08% :white_check_mark:
torchvision-resnet50_fp16 64 5,275.65 5,275.96 -0.01% :white_check_mark:
torchvision-alexnet 64 9,462.89 9,453.05 0.10% :white_check_mark:
torchvision-alexnet_fp16 64 30,488.68 30,544.70 -0.18% :white_check_mark:
torchvision-densenet121 32 1,783.22 1,781.69 0.09% :white_check_mark:
torchvision-densenet121_fp16 32 3,269.27 3,248.65 0.63% :white_check_mark:
torchvision-inceptionv3 32 1,320.36 1,315.05 0.40% :white_check_mark:
torchvision-inceptionv3_fp16 32 2,323.77 2,317.97 0.25% :white_check_mark:
torchvision-vgg16 16 894.70 894.25 0.05% :white_check_mark:
torchvision-vgg16_fp16 16 2,126.60 2,132.78 -0.29% :white_check_mark:
cadene-inceptionv4 16 673.62 673.21 0.06% :white_check_mark:
cadene-resnext64x4 16 592.92 593.03 -0.02% :white_check_mark:
slim-mobilenet 64 7,179.44 7,156.20 0.32% :white_check_mark:
slim-nasnetalarge 64 228.79 230.18 -0.60% :white_check_mark:
slim-resnet50v2 64 2,351.30 2,348.81 0.11% :white_check_mark:
bert-mrpc-onnx 8 nan nan nan% :x:
bert-mrpc-tf 1 306.20 306.34 -0.05% :white_check_mark:
pytorch-examples-wlang-gru 1 299.09 316.30 -5.44% :red_circle:
pytorch-examples-wlang-lstm 1 313.48 315.41 -0.61% :white_check_mark:
torchvision-resnet50_1 1 541.54 529.99 2.18% :white_check_mark:
torchvision-inceptionv3_1 1 300.84 300.18 0.22% :white_check_mark:
torchvision-vgg16_1 1 462.39 462.23 0.04% :white_check_mark:
cadene-dpn92_1 1 329.54 328.35 0.36% :white_check_mark:
cadene-resnext101_1 1 228.80 229.97 -0.51% :white_check_mark:
slim-vgg16_1 1 158.43 158.37 0.04% :white_check_mark:
slim-mobilenet_1 1 1,505.95 1,526.73 -1.36% :white_check_mark:
slim-inceptionv4_1 1 192.40 193.03 -0.32% :white_check_mark:
onnx-taau-downsample 1 323.99 323.49 0.16% 0
dlrm-criteoterabyte 1 19.61 19.61 0.01% 0
dlrm-criteoterabyte_fp16 1 35.79 35.82 -0.08% 0
agentmodel 1 4,907.28 5,091.83 -3.62% 0
agentmodel2 1 5,059.78 nan nan% :x:

This build is not recommended to merge :red_circle:

migraphx-bot commented 1 year ago


    :white_check_mark:bert-mrpc-onnx: PASSED: MIGraphX meets tolerance
:x:bert-mrpc-tf: ERROR - check error outputusage: accuracy_checker.py [-h] --onnx ONNX [--provider PROVIDER]
[--batch BATCH] [--fill1] [--verbose]
[--tolerance TOLERANCE]
accuracy_checker.py: error: the following arguments are required: --onnx

:x:pytorch-examples-wlang-gru: ERROR - check error outputusage: accuracy_checker.py [-h] --onnx ONNX [--provider PROVIDER]
[--batch BATCH] [--fill1] [--verbose]
[--tolerance TOLERANCE]
accuracy_checker.py: error: unrecognized arguments: --fill0

:x:pytorch-examples-wlang-lstm: ERROR - check error outputusage: accuracy_checker.py [-h] --onnx ONNX [--provider PROVIDER]
[--batch BATCH] [--fill1] [--verbose]
[--tolerance TOLERANCE]
accuracy_checker.py: error: unrecognized arguments: --fill0

    :white_check_mark:torchvision-resnet50_1: PASSED: MIGraphX meets tolerance
    :white_check_mark:torchvision-inceptionv3_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
:x:slim-vgg16_1: ERROR - check error outputusage: accuracy_checker.py [-h] --onnx ONNX [--provider PROVIDER]
[--batch BATCH] [--fill1] [--verbose]
[--tolerance TOLERANCE]
accuracy_checker.py: error: the following arguments are required: --onnx

:x:slim-mobilenet_1: ERROR - check error outputusage: accuracy_checker.py [-h] --onnx ONNX [--provider PROVIDER]
[--batch BATCH] [--fill1] [--verbose]
[--tolerance TOLERANCE]
accuracy_checker.py: error: the following arguments are required: --onnx

:x:slim-inceptionv4_1: ERROR - check error outputusage: accuracy_checker.py [-h] --onnx ONNX [--provider PROVIDER]
[--batch BATCH] [--fill1] [--verbose]
[--tolerance TOLERANCE]
accuracy_checker.py: error: the following arguments are required: --onnx

    :white_check_mark:dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance
    :white_check_mark:dlrm-criteoterabyte_fp16: PASSED: MIGraphX meets tolerance
    :white_check_mark:agenmodel: PASSED: MIGraphX meets tolerance
    :white_check_mark:agenmodel2: PASSED: MIGraphX meets tolerance