- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
according to the official documentationDeep Learning accuracy validation framework ,I install the tool accuracy checker and prepare to run the sample to test my installation. There is an ERROR about the main.py file and I dont know WHY. If anyone can help me and thanks.
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ming,
Did you try to run using this sample instruction?
I managed to validate using Windows 10 and Python 3.7.
Would you mind sharing which OS and Python version are you using?
Regards,
Rizal
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have run successfully the sample demo, now I am working on my own model and datasets, and there are some PROBLEMS need to be solved.
In the accuracy checker. I need to describe the Dataset entry in my file of accuracy-checke.yml or we can name the file with the model name. Because I use my own dataset with the format of VOCdevkit2007, I imitate the format and files like voc_detection .But I dont know where to set my dataset and where is the converted dataset.
the ERROR is showed as teh follow.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ming,
Can you share the exact command that you ran and the config file that you have modified?
Regards,
Rizal
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have solved last ERROR,but I am facing another one about accuracy checker.yml,. I'm sure I'm following the YAML file format, I dont know why this ERROR occur????
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
accuracy check -c /opt/intel/openvino 2019.3.334/dep|loyment tools/open model _zoo/models/public/mobilenet-yolo-syg/accuracy-check.yml-m o mz-test-s omz-test/VoCdevkit accuracycheck/VOC2007/-a omz-test/vocdevkit accuracych eck/VOC2007/Annotations/
the file is opened with notepad++
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ming,
You seem to have some format error in your yaml file.
You can try checking it using YAML Lint to verify it.
Furthermore I could not find the properties that you include for the 'tf_object_detection' adapter in the adapters list.
Regards,
Rizal
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
THANKS for applying me , The format error in my yaml file has been solved.
now there is ERROR in my yaml file, maybe it isrelated to the launcher which I dont use very well.
thanks for helping me and your excellent reply.
the command is accuracy_check -c /opt/intel/openvino_2019.3.334/deployment_tools/open_model_zoo/models/public/mobilenet-yolo-syg/accuracy-check.yml -m omz-test -s omz-test/VOCdevkit_accuracycheck -a omz-test
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ming,
It seems from the error message "cannot parse future versions" means that the IR model file you are using is created using newer version of OpenVINO.
You would need to use models created from the same OpenVINO version or older.
Alternatively, I would recommend you to get either the Long Term Support version or the latest version.
Regards,
Rizal
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
What are the differences in the overall usage of the model transformed with the newer version compared to the older version of OPENVINO? Assuming that the IR model comes from someone else, how do I know which version is converted? and how can I use it better?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I rebuilt the IR model and tested for accuracy checker with the same version of Openvino, and found the following error.What should I do???
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have solved some probems,but I faced new one. Thank you for your careful and earnest reply.
I run the yolo-v3-tf models,which is in the open model zoo,but I cant the right map value. The yolo3-omz dictionary includes all the files the command needs which are copied in the open model zoo.
accuracy_check -c /opt/intel/openvino_2020.3.341/deployment_tools/open_model_zoo/tools/accuracy_checker/configs/yolo-v3-tf.yml -m yolo3-omz -s yolo3-omz -a yolo3-omz -td cpu
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ming,
I have replicated your case on my device and found no issue. I suggest you check back your model and program. Please note to follow the steps of 'example of usage' on this page.
Regards,
Adli
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ming,
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,
Adli

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page