Dynamic ConstantOfShape

[CharlieL7]

• Depends on Dynamic allocate #2079 and fill ref operator #2087
• Makes ConstantOfShape work with a variable dimensions input

[codecov]

Codecov Report

Merging #2111 (84ec289) into develop (4188c38) will decrease coverage by 0.02%.
Report is 3 commits behind head on develop.
The diff coverage is 100.00%.

❗ Current head 84ec289 differs from pull request most recent head 3471135. Consider uploading reports for the commit 3471135 to get more accurate results

@@             Coverage Diff             @@
##           develop    #2111      +/-   ##
===========================================
- Coverage    91.49%   91.48%   -0.02%     
===========================================
  Files          430      427       -3     
  Lines        16129    15942     -187     
===========================================
- Hits         14758    14585     -173     
+ Misses        1371     1357      -14     

Files	Coverage Δ
src/onnx/parse_constant_of_shape.cpp	92.85% <100.00%> (+1.19%)	⬆️

... and 17 files with indirect coverage changes

[migraphx-bot]

Test	Batch	Rate new 70574e	Rate old c2e01b	Diff	Compare
torchvision-resnet50	64	2,284.89	2,284.87	0.00%	✅
torchvision-resnet50_fp16	64	5,356.59	5,357.62	-0.02%	✅
torchvision-densenet121	32	1,836.93	1,829.81	0.39%	✅
torchvision-densenet121_fp16	32	3,397.29	3,378.73	0.55%	✅
torchvision-inceptionv3	32	1,344.28	1,335.78	0.64%	✅
torchvision-inceptionv3_fp16	32	2,593.34	2,585.25	0.31%	✅
cadene-inceptionv4	16	676.87	679.39	-0.37%	✅
cadene-resnext64x4	16	590.94	590.19	0.13%	✅
slim-mobilenet	64	7,218.61	7,216.76	0.03%	✅
slim-nasnetalarge	64	236.84	236.57	0.11%	✅
slim-resnet50v2	64	2,528.81	2,530.71	-0.08%	✅
bert-mrpc-onnx	8	721.33	721.25	0.01%	✅
bert-mrpc-tf	1	390.56	391.28	-0.18%	✅
pytorch-examples-wlang-gru	1	297.95	303.90	-1.96%	✅
pytorch-examples-wlang-lstm	1	312.83	311.77	0.34%	✅
torchvision-resnet50_1	1	558.84	555.70	0.56%	✅
torchvision-inceptionv3_1	1	309.22	306.97	0.73%	✅
cadene-dpn92_1	1	354.84	351.70	0.89%	✅
cadene-resnext101_1	1	220.73	220.52	0.10%	✅
slim-vgg16_1	1	224.68	224.37	0.14%	✅
slim-mobilenet_1	1	1,508.64	1,468.80	2.71%	✅
slim-inceptionv4_1	1	219.10	221.26	-0.97%	✅
onnx-taau-downsample	1	322.83	322.42	0.13%	✅
dlrm-criteoterabyte	1	21.67	21.67	-0.01%	✅
dlrm-criteoterabyte_fp16	1	40.65	40.54	0.28%	✅
agentmodel	1	5,669.64	5,779.11	-1.89%	✅
unet_fp16	2	55.09	55.07	0.04%	✅

This build is OK for merge ✅

[migraphx-bot]

    :white_check_mark:bert-mrpc-onnx: PASSED: MIGraphX meets tolerance

    :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

🔴torchvision-inceptionv3_1: FAILED: MIGraphX is not within tolerance - check verbose output

🔴cadene-dpn92_1: FAILED: MIGraphX is not within tolerance - check verbose output

    :white_check_mark:cadene-resnext101_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:slim-vgg16_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:slim-mobilenet_1: PASSED: MIGraphX meets tolerance

🔴slim-inceptionv4_1: FAILED: MIGraphX is not within tolerance - check verbose output

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

    :white_check_mark:agentmodel: PASSED: MIGraphX meets tolerance

    :white_check_mark:unet: PASSED: MIGraphX meets tolerance

[TedThemistokleous]

Fix CI but LGTM here. @pfultz2 / @umangyadav ?

[umangyadav]

should we add tests inside verify_onnx.cpp as well ?

[TedThemistokleous]

marked as high priority as depending PRs merged in already