I am using Tensorflow object detection API and faster rcnn for detecting characters and digits in an image. How can i measure the accuracy of the model? Like is there any evaluation metrics for the same?
As I know that there are two options.
1) you can train your model with train.py and then you can run evaluate.py
2) you can train your model with model_main.py then you won't need to run evaluate file
I am using Tensorflow object detection API and faster rcnn for detecting characters and digits in an image. How can i measure the accuracy of the model? Like is there any evaluation metrics for the same?