I tried integrate my trained detectnet_v2 model (onnx) with Deepstream according to (BYOM Converter - NVIDIA Docs),
My command is:
tao_byom -m ./tao-getting-started_v5.3.0/notebooks/tao_launcher_starter_kit/detectnet_v2/experiment_dir_final/resnet18_detector.onnx -r ./tao-getting-started_v5.3.0/notebooks/tao_launcher_starter_kit/detectnet_v2/experiment_dir_final/byom_resnet18 -n resnet18 --verbose
But got following error message:
ValueError: TAO BYOM currently only supports single input and single output ONNX models while there are 1 inputs and 2 outputs in the provided ONNX model
What’s problem? Thank you.
Complete log:
(byom) xxxx@XXXX-3090:/data/workspace/TAO$ tao_byom -m ./tao-getting-started_v5.3.0/notebooks/tao_launcher_starter_kit/detectnet_v2/experiment_dir_final/resnet18_detector.onnx -r ./tao-getting-started_v5.3.0/notebooks/tao_launcher_starter_kit/detectnet_v2/experiment_dir_final/byom_resnet18 -n resnet --verbose
2024-07-01 18:38:46.902108: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable TF_ENABLE_ONEDNN_OPTS=0.
2024-07-01 18:38:48.583946: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-07-01 18:38:48.649046: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library ‘libcudnn.so.8’; dlerror: libcudnn.so.8: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-12.2/lib64:
2024-07-01 18:38:48.649072: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at Install TensorFlow with pip for how to download and setup the required libraries for your platform.
Skipping registering GPU devices…
INFO: Converter is called.
2024-07-01 18:38:48.697589: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1850] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at GPU 支持 | TensorFlow for how to download and setup the required libraries for your platform.
Skipping registering GPU devices…
Traceback (most recent call last):
File “/home/theo/anaconda3/envs/byom/bin/tao_byom”, line 8, in
sys.exit(main())
File “/home/theo/anaconda3/envs/byom/lib/python3.8/site-packages/tao_byom/entrypoint/entrypoint.py”, line 167, in main
k_model, layer_mappings, lambda_dicts = tao_byom_converter(onnx_model,
File “/home/theo/anaconda3/envs/byom/lib/python3.8/site-packages/tao_byom/converter.py”, line 95, in tao_byom_converter
raise ValueError(“TAO BYOM currently only supports single input and single output ONNX models”
ValueError: TAO BYOM currently only supports single input and single output ONNX models while there are 1 inputs and 2 outputs in the provided ONNX model
2 posts - 2 participants