pytorch-deeplab-xception
pytorch-deeplab-xception copied to clipboard
jfzhang95
DeepLab v3+ model in PyTorch. Support different backbones.
RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED cuda9.0 pytorch1.1.0 python3.6.2
In your code, i see in the loss.py the final loss value is the value after excluding twice the batchsize . And in your train.py，final lr multiply the batchsize. so...
File "train.py", line 303, in main() File "train.py", line 292, in main trainer = Trainer(args) File "train.py", line 37, in __init__ freeze_bn=args.freeze_bn) File "/home/ubuntu/user_space/pytorch-deeplab-xception-master/modeling/deeplab.py", line 22, in __init__ self.aspp =...
Hi It seems that the function `get_1x_lr_params` can only Iterate `nn.Conv2d` in the top node of the backbone. such as the `conv1` , while fails to Iterate 'nn.Conv2d' in subnode,...
Hi, thanks for sharing your code. I trained the model on the cityscapes dataset (btw, you missed a `self` [here](https://github.com/jfzhang95/pytorch-deeplab-xception/blob/0a5814f77fcbf3205eb77aa52c3e8793096e9f5d/dataloaders/datasets/cityscapes.py#L72)) without code edits and I can get only 68% mIoU....
when i use CrossEntropyLoss(),it has aproblem:RuntimeError: 1only batches of spatial targets supported (non-empty 3D tensors) but got targets of size: : [16, 513, 513, 4]
Hi, I am wondering after training, where will the images are saved to? As well as when I testing my own image, where will the output image will be set...
Hi all: Thanks jaemin93 provide inference.py. I used it,but i met some problem that was i can not get the crop of my object.This is the training parameter: Acc:0.9520538766597383,Acc_class:0.9498111995392555, mIoU:0.8039776748270511,...
