question:can i use my own date to train? - Githubissues

wkentaro / pytorch-fcn

PyTorch Implementation of Fully Convolutional Networks. (Training code to reproduce the original result is available.)

MIT License

1.74k stars 480 forks source link

question:can i use my own date to train? #126

Closed zhengkai17 closed 5 years ago

nankingV commented 5 years ago

i got the same question

wkentaro commented 5 years ago

If you implement a dataset class similar to this, https://github.com/wkentaro/pytorch-fcn/blob/eecae876130930d1261b1ae0113787ddf798abcd/torchfcn/datasets/voc.py#L61-L78

you can just replace the VOC dataset class with your own dataset class: https://github.com/wkentaro/pytorch-fcn/blob/eecae876130930d1261b1ae0113787ddf798abcd/examples/voc/train_fcn32s.py#L97-L103