search

datvuthanh / HybridNets

HybridNets: End-to-End Perception Network
MIT License
594 stars 121 forks source link

TypeError: 'NoneType' object is not callable #26

Open Yunboxun opened 2 years ago

Yunboxun commented 2 years ago

image Excuse me, how can I solve this error？

Yunboxun commented 2 years ago

image image

Thank you, but I still don't know how to solve the problem

Yunboxun commented 2 years ago

说实话，我不知道为什么你的对象是Nonytype。所以我认为你应该检查它的基本对象，如 或 .progress_bar``training_generator``DataLoaderX

In fact, I'm testing these data types, but since I'm a rookie, can I have a look at the internal structure of these parameters? Compare them with mine to find out the problems. If I can, I'd like to see 'training generator 'and' train dataset‘，thank you very much.

xoiga123 commented 2 years ago

@Yunboxun are you using multiprocessing or DataParallel? That Pickle error seems familiar and was fixed in ecc835c, pull/clone the repository and use train_ddp.py instead of train.py