aleksispi / drl-rpn-tf

Official Tensorflow implementation of drl-RPN: Deep Reinforcement Learning of Region Proposal Networks (CVPR 2018 paper)
MIT License
78 stars 21 forks source link

I'd like to train from the first without checkpoint, but it is impossible. #7

Closed leesing9 closed 3 years ago

leesing9 commented 3 years ago

How can i train my model without use of ckpt file? It cause error by just remove ckpt files. <- (TypeError: argument of type 'NoneType' is not iterable)