driver verify bisect

[richagadgil]

No description provided.

[codecov]

Codecov Report

Attention: Patch coverage is 66.66667% with 3 lines in your changes missing coverage. Please review.

Project coverage is 92.03%. Comparing base (f694f25) to head (b906842).
Report is 15 commits behind head on develop.

Files with missing lines	Patch %	Lines
src/dom_info.cpp	66.66%	3 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #3460      +/-   ##
===========================================
- Coverage    92.05%   92.03%   -0.02%     
===========================================
  Files          506      506              
  Lines        20837    20878      +41     
===========================================
+ Hits         19181    19215      +34     
- Misses        1656     1663       +7     

Flag	Coverage Δ
	92.03% <66.66%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[pfultz2]

You need to skip the instructions that are const foldable.

[pfultz2]

You can probably create a vector to do the mapping:

std::vector<std::size_t> trim_map;
std::size_t i = 0;
for(const auto& ins:*mm)
{
    if(not ins.can_eval())
        trim_map.push_back(i);
    i++;
}

Then do the bisection over trim_map and then use it to get the actual trim value: verify_bisect(p, trim_map[mid], t, options, vo, inputs, tols).

[pfultz2]

Let me revert these changes to the dominator since its not needed anymore.

[migraphx-bot]

Test	Batch	Rate new b90684	Rate old 9f0187	Diff	Compare
torchvision-resnet50	64	3,249.79	3,257.99	-0.25%	✅
torchvision-resnet50_fp16	64	6,985.44	7,007.13	-0.31%	✅
torchvision-densenet121	32	2,434.48	2,433.76	0.03%	✅
torchvision-densenet121_fp16	32	4,111.59	4,104.98	0.16%	✅
torchvision-inceptionv3	32	1,636.08	1,637.02	-0.06%	✅
torchvision-inceptionv3_fp16	32	2,738.58	2,743.63	-0.18%	✅
cadene-inceptionv4	16	776.05	779.52	-0.45%	✅
cadene-resnext64x4	16	807.87	808.16	-0.04%	✅
slim-mobilenet	64	7,454.59	7,461.23	-0.09%	✅
slim-nasnetalarge	64	208.25	208.15	0.05%	✅
slim-resnet50v2	64	3,434.82	3,439.71	-0.14%	✅
bert-mrpc-onnx	8	1,152.98	1,153.59	-0.05%	✅
bert-mrpc-tf	1	306.55	314.92	-2.66%	✅
pytorch-examples-wlang-gru	1	412.63	419.42	-1.62%	✅
pytorch-examples-wlang-lstm	1	345.77	380.88	-9.22%	🔴
torchvision-resnet50_1	1	779.02	807.68	-3.55%	🔴
cadene-dpn92_1	1	399.34	401.60	-0.56%	✅
cadene-resnext101_1	1	382.53	380.85	0.44%	✅
onnx-taau-downsample	1	344.59	344.22	0.11%	✅
dlrm-criteoterabyte	1	35.09	35.07	0.07%	✅
dlrm-criteoterabyte_fp16	1	58.07	58.16	-0.16%	✅
agentmodel	1	7,983.78	7,954.63	0.37%	✅
unet_fp16	2	58.03	57.95	0.14%	✅
resnet50v1_fp16	1	949.01	964.88	-1.64%	✅
resnet50v1_int8	1	1,005.32	983.11	2.26%	✅
bert_base_cased_fp16	64	1,152.37	1,172.44	-1.71%	✅
bert_large_uncased_fp16	32	355.84	362.89	-1.94%	✅
bert_large_fp16	1	210.17	211.97	-0.85%	✅
distilgpt2_fp16	16	2,160.86	2,204.93	-2.00%	✅
yolov5s	1	534.17	530.94	0.61%	✅
tinyllama	1	43.47	43.43	0.08%	✅
vicuna-fastchat	1	172.02	174.78	-1.58%	✅
whisper-tiny-encoder	1	418.26	417.49	0.19%	✅
whisper-tiny-decoder	1	434.83	423.71	2.62%	✅

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