Yolov5 Convert onnx to NEF model error !!

yolov5之模型是在自己訓練的,並且利用yolov5官方之export.py將pt檔轉換成onnx檔

由於我在kneron的mmdetection文件中只找到yolox說明步驟,因此onnx轉nef我是參考這個網址的step5來操作https://github.com/kneron/kneron-mmdetection/blob/main/docs_kneron/yolox_step_by_step.md

配合官方影片為: https://www.youtube.com/watch?v=6JOyNonmpfY ( 時間為 30:05 ~ 34:05 )

但在跑 python python_api_workflow.py 指令時,出現如下錯誤:

terminate called after throwing an instance of 'std::runtime_error'

 what(): CSim only support CPU node in the end of model and write data to output buffer.

./compilerIpevaluator_720.sh: line 32:  38 Aborted         (core dumped) $LIBS_FOLDER/compiler/compile 720 $model $TMP_FOLDER/config_compiler.json warning compile.log

Traceback (most recent call last):

 File "python_api_workflow.py", line 14, in <module>

  eval_result = km.evaluate()

 File "/workspace/miniconda/lib/python3.7/site-packages/ktc/toolchain.py", line 134, in evaluate

  subprocess.run(['./compilerIpevaluator_720.sh', input_model_path], check=True)

 File "/workspace/miniconda/lib/python3.7/subprocess.py", line 512, in run

  output=stdout, stderr=stderr)

subprocess.CalledProcessError: Command '['./compilerIpevaluator_720.sh', '/data1/input.onnx']' returned non-zero exit status 1.


下圖為 python_api_workflow.py :



我的onnx檔:

Comments

The discussion has been closed due to inactivity. To continue with the topic, please feel free to post a new discussion.