Add FP8 ONNX tests

[ahsan-ca]

This PR adds ONNX tests for FP8.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 91.98%. Comparing base (771f294) to head (8b8a2ab).
Report is 137 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #3041      +/-   ##
===========================================
+ Coverage    91.97%   91.98%   +0.01%     
===========================================
  Files          489      489              
  Lines        19398    19398              
===========================================
+ Hits         17841    17844       +3     
+ Misses        1557     1554       -3     

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

[migraphx-bot]

Test	Batch	Rate new 8b8a2a	Rate old 5fcf86	Diff	Compare
torchvision-resnet50	64	1,751.04	1,751.74	-0.04%	✅
torchvision-resnet50_fp16	64	4,085.59	4,084.34	0.03%	✅
torchvision-densenet121	32	1,467.12	1,467.39	-0.02%	✅
torchvision-densenet121_fp16	32	2,526.53	2,525.45	0.04%	✅
torchvision-inceptionv3	32	889.64	889.64	0.00%	✅
torchvision-inceptionv3_fp16	32	1,483.78	1,483.57	0.01%	✅
cadene-inceptionv4	16	412.53	412.40	0.03%	✅
cadene-resnext64x4	16	419.75	419.50	0.06%	✅
slim-mobilenet	64	4,006.66	4,006.71	-0.00%	✅
slim-nasnetalarge	64	101.04	101.01	0.02%	✅
slim-resnet50v2	64	1,681.26	1,680.60	0.04%	✅
bert-mrpc-onnx	8	615.10	618.22	-0.51%	✅
bert-mrpc-tf	1	278.45	279.81	-0.49%	✅
pytorch-examples-wlang-gru	1	322.08	319.57	0.79%	✅
pytorch-examples-wlang-lstm	1	325.66	289.36	12.55%	🔆
torchvision-resnet50_1	1	467.58	471.89	-0.91%	✅
cadene-dpn92_1	1	246.40	247.09	-0.28%	✅
cadene-resnext101_1	1	204.36	204.23	0.06%	✅
onnx-taau-downsample	1	206.42	206.24	0.09%	✅
dlrm-criteoterabyte	1	22.92	22.90	0.06%	✅
dlrm-criteoterabyte_fp16	1	42.72	42.73	-0.01%	✅
agentmodel	1	6,406.15	6,323.67	1.30%	✅
unet_fp16	2	34.20	34.21	-0.01%	✅
resnet50v1_fp16	1	605.68	589.36	2.77%	✅
resnet50v1_int8	1	577.80	573.50	0.75%	✅
bert_base_cased_fp16	64	646.25	646.33	-0.01%	✅
bert_large_uncased_fp16	32	199.01	198.99	0.01%	✅
bert_large_fp16	1	117.21	117.54	-0.28%	✅
distilgpt2_fp16	16	1,212.11	1,211.40	0.06%	✅
yolov5s	1	301.61	301.25	0.12%	✅
tinyllama	1	23.33	23.34	-0.05%	✅
vicuna-fastchat	1	133.71	133.65	0.05%	✅
whisper-tiny-encoder	1	244.07	244.39	-0.13%	✅
whisper-tiny-decoder	1	256.25	256.66	-0.16%	✅

Check results before 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

[umangyadav]

As a next part we should cover more operators for the "verification".

e.g. "reduction" operations like reduce_max, reduce_sum, avgpool, maxpool
math operations like sine, cosine, sqrt etc.
binary ops : division and multiplication

[CharlieL7]

Address Umang's comment and should be good to go