Giters

walsvid / CoordConv

Pytorch implementation of "An intriguing failing of convolutional neural networks and the CoordConv solution" - https://arxiv.org/abs/1807.03247

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Training time increases linearly for each epoch.

v1093 opened this issue · comments

commented

Currently, I have been using this implementation for one of my object detection models.
As the model is training, batch processing time increases linearly. I still haven't figured out why.

Does anyone see a similar behavior?