dimensions match in find_pointwise_reduce

[bpickrel]

Updates the find_pointwise_reduce matcher in the fuse_reduce compiler pass, to not attempt to fuse a multibroadcast instruction if the shape has an incompatible number of dimensions (rank). This cures the exception causing a torch benchmark to fail, resolves #3007

[migraphx-bot]

Test	Batch	Rate new c136f5	Rate old 36c3eb	Diff	Compare
torchvision-resnet50	64	2,959.74	2,960.87	-0.04%	:white_check_mark:
torchvision-resnet50_fp16	64	6,558.70	6,563.04	-0.07%	:white_check_mark:
torchvision-densenet121	32	2,425.27	2,425.42	-0.01%	:white_check_mark:
torchvision-densenet121_fp16	32	3,962.17	3,960.85	0.03%	:white_check_mark:
torchvision-inceptionv3	32	1,660.91	1,661.60	-0.04%	:white_check_mark:
torchvision-inceptionv3_fp16	32	2,599.94	2,595.55	0.17%	:white_check_mark:
cadene-inceptionv4	16	779.11	777.18	0.25%	:white_check_mark:
cadene-resnext64x4	16	740.86	740.90	-0.01%	:white_check_mark:
slim-mobilenet	64	6,917.45	6,920.33	-0.04%	:white_check_mark:
slim-nasnetalarge	64	177.13	177.26	-0.07%	:white_check_mark:
slim-resnet50v2	64	2,877.35	2,878.57	-0.04%	:white_check_mark:
bert-mrpc-onnx	8	1,064.90	1,066.24	-0.13%	:white_check_mark:
bert-mrpc-tf	1	464.14	494.05	-6.05%	:red_circle:
pytorch-examples-wlang-gru	1	432.63	524.81	-17.56%	:red_circle:
pytorch-examples-wlang-lstm	1	342.71	519.08	-33.98%	:red_circle:
torchvision-resnet50_1	1	797.47	791.51	0.75%	:white_check_mark:
cadene-dpn92_1	1	437.38	401.51	8.93%	:high_brightness:
cadene-resnext101_1	1	366.65	367.36	-0.19%	:white_check_mark:
onnx-taau-downsample	1	349.60	348.99	0.17%	:white_check_mark:
dlrm-criteoterabyte	1	33.59	33.54	0.16%	:white_check_mark:
dlrm-criteoterabyte_fp16	1	56.62	56.60	0.04%	:white_check_mark:
agentmodel	1	7,748.81	7,522.70	3.01%	:high_brightness:
unet_fp16	2	57.35	57.28	0.11%	:white_check_mark:
resnet50v1_fp16	1	875.20	955.51	-8.41%	:red_circle:
resnet50v1_int8	1	778.90	836.30	-6.86%	:red_circle:
bert_base_cased_fp16	64	1,012.46	1,013.15	-0.07%	:white_check_mark:
bert_large_uncased_fp16	32	316.64	316.80	-0.05%	:white_check_mark:
bert_large_fp16	1	nan	nan	nan%	:x:
distilgpt2_fp16	16	1,992.98	1,993.30	-0.02%	:white_check_mark:
yolov5s	1	515.34	513.06	0.44%	:white_check_mark:
tinyllama	1	45.03	44.98	0.13%	:white_check_mark:
vicuna-fastchat	1	178.43	177.15	0.72%	:white_check_mark:
whisper-tiny-encoder	1	405.22	405.22	-0.00%	:white_check_mark:
whisper-tiny-decoder	1	426.72	425.41	0.31%	:white_check_mark:

This build is not recommended to merge :red_circle:

[migraphx-bot]

:x:bert-mrpc-onnx: 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:264: parse_from: PARSE_FROM: Failed reading onnx file: /new-saved-models/huggingface-transformers/bert_mrpc1.onnx

     :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

:x:cadene-resnext101_1: ERROR - check error output

2024-05-16 21:31:39.118626127 [W:onnxruntime:, model.cc:183 Model] ONNX Runtime only guarantees support for models stamped with opset version 7 or above for opset domain 'ai.onnx'. Please upgrade your model to opset 7 or higher. For now, this opset 6 model may run depending upon legacy support of some older opset version operators.
2024-05-16 21:31:39.124780041 [W:onnxruntime:, transpose_optimizer.cc:28 ApplyImpl] Transpose optimizer failed: Unsupported ONNX opset: 6
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 267, in main
sess = ort.InferenceSession(model_name,
File "/usr/local/lib/python3.8/dist-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 419, in init
self._create_inference_session(providers, provider_options, disabled_optimizers)
File "/usr/local/lib/python3.8/dist-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 463, in _create_inference_session
sess.initialize_session(providers, provider_options, disabled_optimizers)
onnxruntime.capi.onnxruntime_pybind11_state.NotImplemented: [ONNXRuntimeError] : 9 : NOT_IMPLEMENTED : Could not find an implementation for BatchNormalization(6) node with name ''

     :white_check_mark: dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance

     :white_check_mark: agentmodel: PASSED: MIGraphX meets tolerance

:x:unet: 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 207, in main
model = migraphx.parse_onnx(model_name,
RuntimeError: /src/AMDMIGraphX/src/onnx/onnx_parser.cpp:264: parse_from: PARSE_FROM: Failed reading onnx file: /new-saved-models/unet/model.onnx

     :white_check_mark: resnet50v1: PASSED: MIGraphX meets tolerance

:red_circle:bert_base_cased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output

:red_circle:bert_large_uncased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output

:x:bert_large: 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:264: parse_from: PARSE_FROM: Failed reading onnx file: /new-saved-models/bert/model.onnx

     :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

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 85.71429% with 1 line in your changes missing coverage. Please review.

Project coverage is 91.82%. Comparing base (77a5597) to head (c136f5b). Report is 151 commits behind head on develop.

Files	Patch %	Lines
src/fuse_reduce.cpp	85.71%	1 Missing :warning:

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #3038   +/-   ##
========================================
  Coverage    91.82%   91.82%           
========================================
  Files          486      486           
  Lines        18951    18955    +4     
========================================
+ Hits         17401    17405    +4     
  Misses        1550     1550           

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

[bpickrel]

Thanks. I made the last change Paul requested--the only remaining open conversation was just a question.

[bpickrel]

@CharlieL7 , I think I have to re-request a review to get this accepted.