Re: Error trying to download and convert yolo-v4-tf model

Jitendra_S_Intel · ‎03-16-2023

I tried in a jupyter notebook and directly in a terminal. Terminal output is pasted below

(openvino_2022.3.0) u73522@s099-n001:~$ omz_downloader --name yolo-v4-tf -o ~/Downloads/models
################|| Downloading yolo-v4-tf ||################

========== Downloading /home/u73522/Downloads/models/public/yolo-v4-tf/yolov4.weights
... 100%, 251677 KB, 67352 KB/s, 3 seconds passed

========== Downloading /home/u73522/Downloads/models/public/yolo-v4-tf/keras-YOLOv3-model-set/tools/model_converter/convert.py
... 100%, 16 KB, 39499 KB/s, 0 seconds passed

========== Downloading /home/u73522/Downloads/models/public/yolo-v4-tf/keras-YOLOv3-model-set/yolo4/models/layers.py
... 100%, 15 KB, 53637 KB/s, 0 seconds passed

========== Downloading /home/u73522/Downloads/models/public/yolo-v4-tf/keras-YOLOv3-model-set/cfg/yolov4.cfg
... 100%, 11 KB, 61731 KB/s, 0 seconds passed

========== Replacing text in /home/u73522/Downloads/models/public/yolo-v4-tf/keras-YOLOv3-model-set/yolo4/models/layers.py
========== Replacing text in /home/u73522/Downloads/models/public/yolo-v4-tf/keras-YOLOv3-model-set/yolo4/models/layers.py

(openvino_2022.3.0) u73522@s099-n001:~$ omz_converter --name yolo-v4-tf -o ~/Downloads/models
========== Running pre-convert script for yolo-v4-tf
Pre-convert command: /data/venv/openvino_2022.3.0/bin/python3.8 -- /data/venv/openvino_2022.3.0/lib/python3.8/site-packages/openvino/model_zoo/models/public/yolo-v4-tf/pre-convert.py -- /home/u73522/public/yolo-v4-tf /home/u73522/Downloads/models/public/yolo-v4-tf

/data/venv/openvino_2022.3.0/bin/python3.8: can't open file '/home/u73522/public/yolo-v4-tf/keras-YOLOv3-model-set/tools/model_converter/convert.py': [Errno 2] No such file or directory
Traceback (most recent call last):
  File "/data/venv/openvino_2022.3.0/lib/python3.8/site-packages/openvino/model_zoo/models/public/yolo-v4-tf/pre-convert.py", line 40, in <module>
    main()
  File "/data/venv/openvino_2022.3.0/lib/python3.8/site-packages/openvino/model_zoo/models/public/yolo-v4-tf/pre-convert.py", line 30, in main
    subprocess.run([sys.executable, '--',
  File "/usr/lib/python3.8/subprocess.py", line 512, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['/data/venv/openvino_2022.3.0/bin/python3.8', '--', '/home/u73522/public/yolo-v4-tf/keras-YOLOv3-model-set/tools/model_converter/convert.py', '/home/u73522/public/yolo-v4-tf/keras-YOLOv3-model-set/cfg/yolov4.cfg', '/home/u73522/public/yolo-v4-tf/yolov4.weights', '/home/u73522/Downloads/models/public/yolo-v4-tf/yolo-v4.savedmodel', '--yolo4_reorder']' returned non-zero exit status 2.

FAILED:
yolo-v4-tf
(openvino_2022.3.0) u73522@s099-n001:~$

Jitendra_S_Intel · ‎03-16-2023

I switched to openvino_2022.1.0.643 virtual env and was able to download and convert yolo-v4-tf using following steps:

omz_downloader --name yolo-v4-tf -o models
omz_converter --name yolo-v4-tf --output_dir models\irs --precisions FP16 -d models
omz_converter --name yolo-v4-tf --output_dir models\irs --precisions FP32 -d models
(openvino_2022.1.0.643) u73522@s099-n001:~/Downloads$

Hairul_Intel · ‎03-19-2023

Hi Jitendra_S_Intel,

Thank you for reaching out to us.

For your information, I've validated from my end using OpenVINO 2022.3.0 and OpenVINO Jupyter Notebook. Both were successful in downloading and converting the yolo-v4-tf model. Here are the results:

Converting using openvino-dev[tensorflow2]:

Converting using openvino jupyter notebook:

On another note, the error log you've provided seems to indicate that 'yolo-v4-tf/keras-YOLOv3-model-set/tools/model_converter/convert.py' file was not found.

Please ensure the following file is available once you've ran the "omz_downloader --name yolo-v4-tf" command:

Regards,

Hairul

Hairul_Intel · ‎03-26-2023

Hi Jitendra_S_Intel,

This thread will no longer be monitored since we have provided a solution. If you need any additional information from Intel, please submit a new question.

Regards,

Hairul