Bf16 Verify Onnx Tests

[richagadgil]

Tests with different ONNX representations than half:

• mean_bf16_test
• instance_norm_bf16 tests

Tests for half that do not seem to have BF16 counterpart:

• mod_test_half

Tests Failing:

• gelu_default_bf16_test

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.19%. Comparing base (cd37826) to head (9d03a89).
Report is 1 commits behind head on develop.

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #3627   +/-   ##
========================================
  Coverage    92.18%   92.19%           
========================================
  Files          513      513           
  Lines        21596    21598    +2     
========================================
+ Hits         19908    19912    +4     
+ Misses        1688     1686    -2     

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

[migraphx-bot]

Test	Batch	Rate new 9d03a8	Rate old bd6006	Diff	Compare
torchvision-resnet50	64	3,255.46	3,255.03	0.01%	✅
torchvision-resnet50_fp16	64	6,976.96	6,995.38	-0.26%	✅
torchvision-densenet121	32	2,434.35	2,434.81	-0.02%	✅
torchvision-densenet121_fp16	32	4,102.68	4,064.36	0.94%	✅
torchvision-inceptionv3	32	1,629.76	1,628.42	0.08%	✅
torchvision-inceptionv3_fp16	32	2,745.01	2,747.11	-0.08%	✅
cadene-inceptionv4	16	764.99	765.07	-0.01%	✅
cadene-resnext64x4	16	809.57	809.98	-0.05%	✅
slim-mobilenet	64	7,464.19	7,462.14	0.03%	✅
slim-nasnetalarge	64	208.46	208.42	0.02%	✅
slim-resnet50v2	64	3,441.42	3,441.19	0.01%	✅
bert-mrpc-onnx	8	1,150.29	1,143.39	0.60%	✅
bert-mrpc-tf	1	466.19	468.19	-0.43%	✅
pytorch-examples-wlang-gru	1	417.68	414.07	0.87%	✅
pytorch-examples-wlang-lstm	1	387.79	389.56	-0.45%	✅
torchvision-resnet50_1	1	773.21	774.11	-0.12%	✅
cadene-dpn92_1	1	397.75	415.75	-4.33%	🔴
cadene-resnext101_1	1	382.95	383.64	-0.18%	✅
onnx-taau-downsample	1	345.75	345.07	0.20%	✅
dlrm-criteoterabyte	1	33.34	33.35	-0.03%	✅
dlrm-criteoterabyte_fp16	1	52.78	52.74	0.07%	✅
agentmodel	1	10,473.23	9,248.55	13.24%	🔆
unet_fp16	2	58.85	58.82	0.04%	✅
resnet50v1_fp16	1	933.80	945.33	-1.22%	✅
resnet50v1_int8	1	1,017.89	1,023.31	-0.53%	✅
bert_base_cased_fp16	64	1,169.96	1,170.96	-0.09%	✅
bert_large_uncased_fp16	32	363.48	363.29	0.05%	✅
bert_large_fp16	1	199.88	198.28	0.81%	✅
distilgpt2_fp16	16	2,200.01	2,196.36	0.17%	✅
yolov5s	1	526.38	528.08	-0.32%	✅
tinyllama	1	43.39	43.42	-0.08%	✅
vicuna-fastchat	1	172.56	174.92	-1.35%	✅
whisper-tiny-encoder	1	418.11	417.89	0.05%	✅
whisper-tiny-decoder	1	427.89	427.86	0.01%	✅

This build is not recommended to merge 🔴

[migraphx-bot]

     ✅ 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