Scalar multiplication and followed by conv

[aarushjain29]

No description provided.

[codecov]

Codecov Report

Attention: Patch coverage is 72.97297% with 10 lines in your changes missing coverage. Please review.

Project coverage is 91.98%. Comparing base (e2e9219) to head (69457bd).
Report is 3 commits behind head on develop.

Files with missing lines	Patch %	Lines
src/simplify_algebra.cpp	71.42%	10 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #3478      +/-   ##
===========================================
- Coverage    92.02%   91.98%   -0.04%     
===========================================
  Files          509      509              
  Lines        21005    21034      +29     
===========================================
+ Hits         19330    19349      +19     
- Misses        1675     1685      +10     

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

[pfultz2]

What operators have a scalar attribute?

[pfultz2]

Unit tests needs to be added.

[migraphx-bot]

Test	Batch	Rate new 69457b	Rate old fc26f0	Diff	Compare
torchvision-resnet50	64	3,259.24	3,261.06	-0.06%	✅
torchvision-resnet50_fp16	64	6,967.28	6,993.61	-0.38%	✅
torchvision-densenet121	32	2,433.46	2,434.05	-0.02%	✅
torchvision-densenet121_fp16	32	4,073.77	4,095.98	-0.54%	✅
torchvision-inceptionv3	32	1,637.22	1,638.20	-0.06%	✅
torchvision-inceptionv3_fp16	32	2,757.50	2,759.48	-0.07%	✅
cadene-inceptionv4	16	780.34	775.34	0.64%	✅
cadene-resnext64x4	16	808.13	808.43	-0.04%	✅
slim-mobilenet	64	7,473.13	7,536.82	-0.85%	✅
slim-nasnetalarge	64	208.51	211.47	-1.40%	✅
slim-resnet50v2	64	nan	3,501.50	nan%	❌
bert-mrpc-onnx	8	1,148.66	1,149.92	-0.11%	✅
bert-mrpc-tf	1	320.36	465.24	-31.14%	🔴
pytorch-examples-wlang-gru	1	412.23	413.68	-0.35%	✅
pytorch-examples-wlang-lstm	1	377.12	379.94	-0.74%	✅
torchvision-resnet50_1	1	754.52	786.00	-4.00%	🔴
cadene-dpn92_1	1	nan	400.42	nan%	❌
cadene-resnext101_1	1	380.37	381.26	-0.23%	✅
onnx-taau-downsample	1	367.87	343.07	7.23%	🔆
dlrm-criteoterabyte	1	35.04	33.36	5.02%	🔆
dlrm-criteoterabyte_fp16	1	58.18	52.73	10.33%	🔆
agentmodel	1	7,997.95	8,550.08	-6.46%	🔴
unet_fp16	2	58.96	58.85	0.19%	✅
resnet50v1_fp16	1	931.17	941.58	-1.11%	✅
resnet50v1_int8	1	972.20	970.00	0.23%	✅
bert_base_cased_fp16	64	1,171.27	1,171.45	-0.02%	✅
bert_large_uncased_fp16	32	363.67	363.62	0.01%	✅
bert_large_fp16	1	210.95	200.57	5.18%	🔆
distilgpt2_fp16	16	2,202.17	2,203.25	-0.05%	✅
yolov5s	1	546.60	540.96	1.04%	✅
tinyllama	1	43.49	43.52	-0.08%	✅
vicuna-fastchat	1	177.37	174.66	1.55%	✅
whisper-tiny-encoder	1	418.40	418.66	-0.06%	✅
whisper-tiny-decoder	1	425.50	428.20	-0.63%	✅

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: 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 227, in main
model.compile(
RuntimeError: /src/AMDMIGraphX/src/include/migraphx/check_shapes.hpp:220: same_dims: mul: Dimensions do not match

     ✅ 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

[pfultz2]

This should be checking for slice -> mul -> add. There is no reason to check for convolution.

[pfultz2]

Also, in the apply method you need to check the other slices go into add or mul -> add.