TorchSeg
TorchSeg copied to clipboard
ycszen
Fast, modular reference implementation and easy training of Semantic Segmentation algorithms in PyTorch.
@ycszen 作者你好，我在训练的过程成一直报页面文件太小，无法操作，显存爆了的问题，但是我是6g显存，单卡，而且是训练的res18网络，我把batch_size都设置成1了，还是会报这样的错，请问是问题出在哪了呢？感谢！ OSError: [WinError 1455] 页面文件太小，无法完成操作。
``` 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' ```
Trying to run bash script.sh I got the following error Traceback (most recent call last): File "train.py", line 17, in from dataloader import get_train_loader File "/media/D/users/Idan/Suha_Maryam/TorchSemiSeg-main/exp.voc/voc8.res50v3+.CPS/dataloader.py", line 8, in from...
Hi, thanks for your great work. I want to know if the model 'resnet50_v1c.pth' is pretrained on the imagenet? By the way, what does 'v1c' mean?
Hi @ycszen , thank you for providing this great repo. I have a question about the difference between realtime res18 and non-realtime res18 model. As shown in the readme, realtime...
can you give me a example about the train.txt, val.txt, test.txt
作者你好，你在cityscapes.bisenet.R18.speed训练过程中的数据处理部分，把label下采样的1/8大小，后面用1/8 feature map与1/8 label做loss, 而且在evaluate过程中，也是用1/8 feature map与1/8 label算mIoU，但是我看一般的做法是把1/8 feature map上采样到原图大小，然后与label做loss,或者求mIoU,请问你的这种做法能在语义分割领域得到同行的认可嘛，如果可以，那你的这种做法与一般做法的优劣又在哪呐，希望得到你的回复,我想在你提出的构架的基础上，做自己的东西，我的微信号为：18434391857，希望能加你微信。
I have one question, I trained model for one class(0 & 1) with 2000 images of dimension 512x512. The accuracy of the output is somewhat good. After that I have...
I want to evaluate the model on my own images (not from those datasets). How can I do that ? Thanks
