about eval

[igodrr]

Hi authers:
Thank you very much for your work that I have learned a lot！I would like to ask if there is a code for testing the evaluation metrics of the dataset，I can't find it in the code repository. I'm looking forward to your code for evaling.
Thank you very much！

[zubair-irshad]

Hello @igodrr Could you please follow this (#10) issue for getting evaluation for your trained model. We mainly follow the evaluation scripts from object-deformnet with minor changes to account for the category agnostic predictions from our network i.e. we use class ids from mask rcnn, as mentioned in our paper, for a fair evaluation with other category-specific approaches.