Enable BF16 Types for DLM Models in Examples

[richagadgil]

No description provided.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.41%. Comparing base (04b82df) to head (8a15a51).

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #3851   +/-   ##
========================================
  Coverage    92.41%   92.41%           
========================================
  Files          520      520           
  Lines        22471    22471           
========================================
  Hits         20766    20766           
  Misses        1705     1705           

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

[migraphx-bot]

Test	Batch	Rate new 069e10	Rate old b19d13	Diff	Compare
torchvision-resnet50	64	3,235.40	3,237.11	-0.05%	✅
torchvision-resnet50_fp16	64	6,880.90	6,877.59	0.05%	✅
torchvision-densenet121	32	2,438.37	2,437.56	0.03%	✅
torchvision-densenet121_fp16	32	4,205.70	4,202.03	0.09%	✅
torchvision-inceptionv3	32	1,612.21	1,614.89	-0.17%	✅
torchvision-inceptionv3_fp16	32	2,680.90	2,677.87	0.11%	✅
cadene-inceptionv4	16	750.65	750.76	-0.01%	✅
cadene-resnext64x4	16	810.29	810.16	0.02%	✅
slim-mobilenet	64	6,659.85	6,665.81	-0.09%	✅
slim-nasnetalarge	64	196.88	196.84	0.02%	✅
slim-resnet50v2	64	3,434.49	3,431.69	0.08%	✅
bert-mrpc-onnx	8	1,144.03	1,141.15	0.25%	✅
bert-mrpc-tf	1	486.88	482.99	0.81%	✅
pytorch-examples-wlang-gru	1	475.59	486.63	-2.27%	✅
pytorch-examples-wlang-lstm	1	447.14	443.83	0.74%	✅
torchvision-resnet50_1	1	808.54	806.76	0.22%	✅
cadene-dpn92_1	1	427.09	430.96	-0.90%	✅
cadene-resnext101_1	1	392.14	392.23	-0.02%	✅
onnx-taau-downsample	1	371.69	371.19	0.14%	✅
dlrm-criteoterabyte	1	31.82	31.79	0.10%	✅
dlrm-criteoterabyte_fp16	1	51.02	51.07	-0.11%	✅
agentmodel	1	8,681.76	8,932.40	-2.81%	✅
unet_fp16	2	58.29	58.29	0.00%	✅
resnet50v1_fp16	1	1,043.99	1,045.16	-0.11%	✅
resnet50v1_int8	1	798.15	804.91	-0.84%	✅
bert_base_cased_fp16	64	1,164.41	1,164.48	-0.01%	✅
bert_large_uncased_fp16	32	361.61	361.61	0.00%	✅
bert_large_fp16	1	200.46	200.06	0.20%	✅
distilgpt2_fp16	16	2,219.70	2,212.16	0.34%	✅
yolov5s	1	520.19	522.38	-0.42%	✅
tinyllama	1	43.61	43.59	0.03%	✅
vicuna-fastchat	1	43.83	43.89	-0.13%	✅
whisper-tiny-encoder	1	412.55	412.16	0.10%	✅
whisper-tiny-decoder	1	409.22	408.89	0.08%	✅
yolov10	1	nan	nan	nan%	❌
llama2_7b	1	nan	nan	nan%	❌
qwen1.5-7b	1	nan	nan	nan%	❌
phi3-3.8b	1	nan	nan	nan%	❌
mask-rcnn	1	nan	nan	nan%	❌
llama3-8b	1	nan	nan	nan%	❌
whisper-large-encoder	1	nan	nan	nan%	❌
whisper-large-decoder	1	nan	nan	nan%	❌
mistral-7b	1	nan	nan	nan%	❌
FLUX.1-schnell	1	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

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 1 256 @attention_mask 1 256

❌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: 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/mask-rcnn/MaskRCNN-10.onnx

❌llama3-8b: 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/llama3-8b/model.onnx

❌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/onnx/onnx_parser.cpp:264: parse_from: PARSE_FROM: Failed reading onnx file: /new-saved-models/whisper-large/encoder_model.onnx

❌whisper-large-decoder: 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/whisper-large/decoder_model.onnx

❌mistral-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/mistral-7b/model.onnx

❌FLUX.1-schnell: 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/FLUX.1-schnell/text_encoder/model.onnx