Fix: Driver --batch option sets Window Dimensions.

[lakhinderwalia]

No description provided.

[causten]

Take a look at the models that are failing in CI. You likely have caught some input parameter assumptions.

[codecov]

Codecov Report

Attention: Patch coverage is 84.61538% with 2 lines in your changes missing coverage. Please review.

Project coverage is 92.28%. Comparing base (976ae75) to head (5966118).
Report is 5 commits behind head on develop.

Files with missing lines	Patch %	Lines
src/onnx/onnx_parser.cpp	87.50%	1 Missing ⚠️
src/program.cpp	50.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #3770   +/-   ##
========================================
  Coverage    92.28%   92.28%           
========================================
  Files          519      519           
  Lines        22222    22224    +2     
========================================
+ Hits         20507    20510    +3     
+ Misses        1715     1714    -1     

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

[migraphx-bot]

Test	Batch	Rate new 596611	Rate old 43488c	Diff	Compare
torchvision-resnet50	64	3,251.92	3,254.25	-0.07%	✅
torchvision-resnet50_fp16	64	6,924.08	6,916.65	0.11%	✅
torchvision-densenet121	32	2,453.97	2,455.95	-0.08%	✅
torchvision-densenet121_fp16	32	4,168.73	4,183.26	-0.35%	✅
torchvision-inceptionv3	32	1,629.16	1,629.97	-0.05%	✅
torchvision-inceptionv3_fp16	32	2,715.80	2,715.98	-0.01%	✅
cadene-inceptionv4	16	763.07	762.93	0.02%	✅
cadene-resnext64x4	16	811.88	812.98	-0.13%	✅
slim-mobilenet	64	7,461.43	7,459.33	0.03%	✅
slim-nasnetalarge	64	208.58	208.65	-0.03%	✅
slim-resnet50v2	64	3,448.03	3,445.08	0.09%	✅
bert-mrpc-onnx	8	1,148.73	1,147.48	0.11%	✅
bert-mrpc-tf	1	475.79	481.97	-1.28%	✅
pytorch-examples-wlang-gru	1	481.59	478.37	0.67%	✅
pytorch-examples-wlang-lstm	1	437.19	439.41	-0.50%	✅
torchvision-resnet50_1	1	797.68	803.28	-0.70%	✅
cadene-dpn92_1	1	426.58	427.49	-0.21%	✅
cadene-resnext101_1	1	384.85	376.23	2.29%	✅
onnx-taau-downsample	1	372.32	373.50	-0.32%	✅
dlrm-criteoterabyte	1	33.32	33.31	0.03%	✅
dlrm-criteoterabyte_fp16	1	52.65	52.64	0.02%	✅
agentmodel	1	8,582.66	8,546.35	0.42%	✅
unet_fp16	2	58.31	58.43	-0.21%	✅
resnet50v1_fp16	1	1,042.59	1,033.48	0.88%	✅
resnet50v1_int8	1	1,043.18	1,018.13	2.46%	✅
bert_base_cased_fp16	64	1,181.36	1,181.00	0.03%	✅
bert_large_uncased_fp16	32	365.01	365.36	-0.10%	✅
bert_large_fp16	1	200.00	201.35	-0.67%	✅
distilgpt2_fp16	16	2,226.54	2,228.65	-0.09%	✅
yolov5s	1	534.84	527.51	1.39%	✅
tinyllama	1	43.60	43.62	-0.05%	✅
vicuna-fastchat	1	nan	173.15	nan%	❌
whisper-tiny-encoder	1	418.05	416.80	0.30%	✅
whisper-tiny-decoder	1	428.13	429.87	-0.41%	✅

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: ERROR - check error output

Traceback (most recent call last):
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 340, in
main()
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 205, in main
model = migraphx.parse_onnx(model_name, default_dim_value=batch)
RuntimeError: /src/AMDMIGraphX/src/onnx/onnx_parser.cpp:654: operator(): Batch inserted at index 1 of lS_i

     ✅ 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: ERROR - check error output

Traceback (most recent call last):
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 340, in
main()
File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 205, in main
model = migraphx.parse_onnx(model_name, default_dim_value=batch)
RuntimeError: /src/AMDMIGraphX/src/onnx/onnx_parser.cpp:654: operator(): Batch inserted at index 1 of input_ids

     ✅ whisper-tiny-encoder: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-decoder: PASSED: MIGraphX meets tolerance

     ✅ distilgpt2_fp16: PASSED: MIGraphX meets tolerance

[CharlieL7]

Why check default_dim_value and idx? If default_dim_value is 0 that would also be an error since it's would be setting a dimenion to zero?

[CharlieL7]

This should probably default to 1.