Disable dot/mul optimizations when there is int4 weights

[pfultz2]

No description provided.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.18%. Comparing base (f11472c) to head (da1eadd).
Report is 12 commits behind head on develop.

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #3645   +/-   ##
========================================
  Coverage    92.18%   92.18%           
========================================
  Files          513      513           
  Lines        21576    21596   +20     
========================================
+ Hits         19889    19908   +19     
- Misses        1687     1688    +1     

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

---

🚨 Try these New Features:

• Flaky Tests Detection - Detect and resolve failed and flaky tests

[lakhinderwalia]

Shouldn't this be any_of? Also, the name of this function/lambda from_int4 should say any or all, just to make its intent clear. Thanks.

[pfultz2]

Shouldn't this be any_of?

No, if its mixed then we cant get int4 out of it.

[migraphx-bot]

Test	Batch	Rate new da1ead	Rate old 0ad807	Diff	Compare
torchvision-resnet50	64	3,257.59	3,239.44	0.56%	✅
torchvision-resnet50_fp16	64	6,983.44	6,989.98	-0.09%	✅
torchvision-densenet121	32	2,434.09	2,435.40	-0.05%	✅
torchvision-densenet121_fp16	32	4,061.07	4,083.28	-0.54%	✅
torchvision-inceptionv3	32	1,627.04	1,630.36	-0.20%	✅
torchvision-inceptionv3_fp16	32	2,747.44	2,743.98	0.13%	✅
cadene-inceptionv4	16	765.05	765.72	-0.09%	✅
cadene-resnext64x4	16	810.37	811.01	-0.08%	✅
slim-mobilenet	64	7,460.35	7,471.20	-0.15%	✅
slim-nasnetalarge	64	208.48	208.50	-0.01%	✅
slim-resnet50v2	64	3,442.10	3,438.88	0.09%	✅
bert-mrpc-onnx	8	1,150.93	1,147.12	0.33%	✅
bert-mrpc-tf	1	493.27	469.98	4.96%	🔆
pytorch-examples-wlang-gru	1	423.16	416.34	1.64%	✅
pytorch-examples-wlang-lstm	1	380.10	463.26	-17.95%	🔴
torchvision-resnet50_1	1	760.87	770.23	-1.21%	✅
cadene-dpn92_1	1	396.47	399.89	-0.85%	✅
cadene-resnext101_1	1	383.09	382.62	0.12%	✅
onnx-taau-downsample	1	345.98	345.09	0.26%	✅
dlrm-criteoterabyte	1	33.33	33.34	-0.04%	✅
dlrm-criteoterabyte_fp16	1	52.78	52.72	0.12%	✅
agentmodel	1	8,219.49	10,261.49	-19.90%	🔴
unet_fp16	2	58.91	58.87	0.07%	✅
resnet50v1_fp16	1	933.46	982.51	-4.99%	🔴
resnet50v1_int8	1	1,003.14	1,007.17	-0.40%	✅
bert_base_cased_fp16	64	1,168.84	1,169.34	-0.04%	✅
bert_large_uncased_fp16	32	363.18	363.46	-0.08%	✅
bert_large_fp16	1	198.68	198.15	0.27%	✅
distilgpt2_fp16	16	2,202.56	2,201.47	0.05%	✅
yolov5s	1	526.76	536.48	-1.81%	✅
tinyllama	1	43.43	43.38	0.12%	✅
vicuna-fastchat	1	176.92	172.97	2.28%	✅
whisper-tiny-encoder	1	418.34	418.04	0.07%	✅
whisper-tiny-decoder	1	427.78	423.83	0.93%	✅

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