frcnn-from-scratch-with-keras
frcnn-from-scratch-with-keras copied to clipboard
Training is slow
The training is slow because NMS operation is done in python. should enable NMS be done in cython for faster training and inference.
Training is indeed a bit on the slow side.. Do you know any parameter settings that may influence this? e.g.: --num_rois has a significant impact, has it not?
kindly tell me , in where i have to change the code ??