pytorch-deeplab-xception
pytorch-deeplab-xception copied to clipboard
DeepLab v3+ model in PyTorch. Support different backbones.
I've tried both xception and resnet backbone.The resent backbone is working good but when using xception backbone, results are very bad. And even, training for longer making the resulting image...
I noticed a code in the file "deeplab.py". That is, if backbone == 'drn': output_stride = 8 Why do we need to force to set the output stride to 8?...
Hello thank you for sharing the code :) I noticed that "xception pretrain model url " in https://github.com/jfzhang95/pytorch-deeplab-xception/blob/master/modeling/backbone/xception.py#L248 doesn't exist. Did anyone have the same problem? If you have solved...
Have you ever test the params of the net(dp3+,mobilenetv2)?I got 7.5M. But when I use the code in mmsegmentaion, I got 15M? I think the params is only related with...
Hello, first thank you for providing the code ! Could anybody tell me how the augmentation used in deeplabv3+ paper** is done? I have encountered some problem with mIoU issue....
In the project of pytorch-deeplab-xception , I can't find the inference.py file. so I can only make a trainning of my dataset
![Uploading thai-nickname-396184.png…]() _Originally posted by @unnln in https://github.com/jfzhang95/pytorch-deeplab-xception/issues/238#issuecomment-2289439845_
![Uploading 9da3c33a4903eb23970199c363aa29e9_0.jpeg…]() _Originally posted by @unnln in https://github.com/jfzhang95/pytorch-deeplab-xception/issues/236#issuecomment-2289434787_
เพือ่ความปลอดภัยและความถูกต้องในการใช้งาน