Fuse multiple outputs for pointwise ops

[pfultz2]

No description provided.

[codecov]

Codecov Report

Attention: Patch coverage is 97.59615% with 5 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/fuse_concat.cpp	88.24%	2 Missing ⚠️
src/fuse_pointwise.cpp	98.28%	2 Missing ⚠️
src/fuse_pointwise_reduce.cpp	0.00%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #3870      +/-   ##
===========================================
- Coverage    92.41%   92.07%   -0.34%     
===========================================
  Files          522      525       +3     
  Lines        22532    24272    +1740     
===========================================
+ Hits         20822    22348    +1526     
- Misses        1710     1924     +214     

Files with missing lines	Coverage Δ
src/include/migraphx/fuse_pointwise.hpp	100.00% <ø> (ø)
src/include/migraphx/instruction.hpp	100.00% <ø> (ø)
src/include/migraphx/matcher.hpp	96.05% <100.00%> (-1.23%)	⬇️
src/include/migraphx/module.hpp	100.00% <ø> (ø)
src/include/migraphx/shape.hpp	91.18% <ø> (ø)
src/instruction.cpp	88.46% <100.00%> (-0.43%)	⬇️
src/module.cpp	86.68% <100.00%> (-<0.01%)	⬇️
src/param_utils.cpp	97.14% <100.00%> (-2.86%)	⬇️
src/replace_allocate.cpp	100.00% <100.00%> (ø)
src/shape.cpp	92.20% <100.00%> (-0.72%)	⬇️
... and 3 more

... and 351 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[amd-jmacaran]

/AzurePipelines run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[migraphx-bot]

Test	Batch	Rate new 45eb6e	Rate old ecb974	Diff	Compare
torchvision-resnet50	64	3,229.57	3,231.64	-0.06%	✅
torchvision-resnet50_fp16	64	6,861.91	6,867.39	-0.08%	✅
torchvision-densenet121	32	2,431.54	2,432.50	-0.04%	✅
torchvision-densenet121_fp16	32	4,206.60	4,212.33	-0.14%	✅
torchvision-inceptionv3	32	1,612.96	1,613.30	-0.02%	✅
torchvision-inceptionv3_fp16	32	2,698.99	2,696.30	0.10%	✅
cadene-inceptionv4	16	749.73	750.25	-0.07%	✅
cadene-resnext64x4	16	809.50	809.71	-0.03%	✅
slim-mobilenet	64	6,649.39	6,654.55	-0.08%	✅
slim-nasnetalarge	64	196.73	203.03	-3.11%	🔴
slim-resnet50v2	64	3,319.66	3,434.71	-3.35%	🔴
bert-mrpc-onnx	8	1,142.41	1,142.05	0.03%	✅
bert-mrpc-tf	1	463.66	464.19	-0.11%	✅
pytorch-examples-wlang-gru	1	324.94	476.27	-31.77%	🔴
pytorch-examples-wlang-lstm	1	442.33	442.88	-0.12%	✅
torchvision-resnet50_1	1	812.72	813.23	-0.06%	✅
cadene-dpn92_1	1	427.60	421.24	1.51%	✅
cadene-resnext101_1	1	392.16	392.62	-0.12%	✅
onnx-taau-downsample	1	395.93	395.87	0.01%	✅
dlrm-criteoterabyte	1	31.80	31.80	-0.01%	✅
dlrm-criteoterabyte_fp16	1	50.99	50.96	0.05%	✅
agentmodel	1	8,878.79	9,458.32	-6.13%	🔴
unet_fp16	2	58.37	58.33	0.08%	✅
resnet50v1_fp16	1	1,084.91	1,071.23	1.28%	✅
resnet50v1_int8	1	881.90	893.66	-1.31%	✅
bert_base_cased_fp16	64	1,158.60	1,162.33	-0.32%	✅
bert_large_uncased_fp16	32	353.70	353.92	-0.06%	✅
bert_large_fp16	1	194.51	194.83	-0.16%	✅
distilgpt2_fp16	16	2,218.18	2,215.11	0.14%	✅
yolov5s	1	496.67	543.55	-8.62%	🔴
tinyllama	1	43.60	43.59	0.02%	✅
vicuna-fastchat	1	43.81	44.05	-0.55%	✅
whisper-tiny-encoder	1	411.71	411.57	0.03%	✅
whisper-tiny-decoder	1	411.43	411.31	0.03%	✅
llama2_7b	1	nan	nan	nan%	❌
qwen1.5-7b	1	23.41	23.41	-0.03%	✅
phi3-3.8b	1	nan	nan	nan%	❌
mask-rcnn	1	18.44	18.55	-0.58%	✅
llama3-8b	1	21.65	21.65	0.00%	✅
whisper-large-encoder	1	10.17	10.17	-0.01%	✅
whisper-large-decoder	1	99.05	97.78	1.30%	✅
mistral-7b	1	23.65	23.63	0.08%	✅
FLUX.1-schnell	1	894.15	904.60	-1.16%	✅
nan	nan	nan	nan	nan%	❌

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

❌llama2_7b: 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/llama2_7b/decoder_model.onnx

❌#qwen1.5-7b: ERROR - check error output

usage: accuracy_checker.py [-h] [--onnx ONNX] [--tf TF] [--provider PROVIDER]
[--batch BATCH] [--fill1] [--fill0] [--fp16]
[--argmax] [--verbose] [--tolerance TOLERANCE]
[--input-dim INPUT_DIM] [--target TARGET]
[--ort-run] [--ort-logging]
[--disable-offload-copy] [--disable-fast-math]
[--exhaustive_tune]
accuracy_checker.py: error: unrecognized arguments: input_ids attention_mask position_ids 1 256 @attention_mask 1 256 @position_ids 1 256

❌phi3-3.8b: 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/phi3-3.8b/model.onnx

🔴mask-rcnn: FAILED: MIGraphX is not within tolerance - check verbose output

     ✅ llama3-8b: PASSED: MIGraphX meets tolerance

❌#whisper-large-encoder: 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/include/migraphx/op/convolution.hpp:100: normalize_compute_shape: CONVOLUTION: mismatched channel numbers

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

     ✅ mistral-7b: PASSED: MIGraphX meets tolerance

     ✅ FLUX.1-schnell: PASSED: MIGraphX meets tolerance