The logging works fine. But I am trying to train several videos and it is a long list of information. All I need is the average accuracy of all the videos combined.
For a direct feed video, it is not supported in Accuracy Checker. Only image datasets are validated in OpenVINO Accuracy Checker. If you want to use a video, you can segregate it into frames or still image and use it as a dataset.
Meanwhile, you can check out the Dataset Preparation Guide.
This thread will no longer be monitored since we have provided information. If you need any additional information from Intel, please submit a new question.