search

yu-changqian / TorchSeg

Fast, modular reference implementation and easy training of Semantic Segmentation algorithms in PyTorch.
MIT License
1.41k stars 253 forks source link

How to prepare the training data? How to specify the folder? #4

Open jiapei100 opened 5 years ago

jiapei100 commented 5 years ago 
FileNotFoundError: [Errno 2] No such file or directory: '....../Cityscapes/config_new/train.txt'
FileNotFoundError: [Errno 2] No such file or directory: '....../ADE20K_2016_07_26/images/training.txt'
yu-changqian commented 5 years ago

This needs you to construct the dataset. I will release an example how to prepare the data later.

zhixuanli commented 5 years ago

Hoping the example too :smile: @ycszen

kazenokizi commented 5 years ago

@zhixuanli Thanks for your repo; it is very useful. I wonder whether the ground truths should be encoded as RGB images (i.e. aeroplane: [0,0,128]) or greyscale images (background: [1]), to make BiSeNet work? Thanks.

Dorispaopao commented 5 years ago

when will you release the example?

shivSD commented 5 years ago

@ycszen Did you release the examples ? please can you share the link ?

miscedence12 commented 2 years ago

@ycszen Did you release the examples ? please can you share the link ?

miscedence12 commented 2 years ago

@zhixuanli Thanks for your repo; it is very useful. I wonder whether the ground truths should be encoded as RGB images (i.e. aeroplane: [0,0,128]) or greyscale images (background: [1]), to make BiSeNet work? Thanks.

hello@kazenokizi, Have you ever solved the problem?