pytorch-semseg
pytorch-semseg copied to clipboard
meetps
Semantic Segmentation Architectures Implemented in PyTorch
My platform: win10+segnet+pascal Excuse me, did you perform a minus mean operation on the input data during training? Seen from the code, there is no such operation. `self.tf = transforms.Compose(...
The code is great, and I use the code to get coarse result, but I find it difficult to get the best performance on CamVid or other dataset, like segnet,...
In the original FRRN paper, the output channel of the last FRRU block was 96 and you chose 48. Are there any specific reasons?
I see you guys set img_norm to False in train and val. However, we can usually see others' code normalize input image to [0,1]. So what's the difference? In what...
I want to train pspnet on my own dataset, but I don't know how to calculate the loss because there are two outputs x and x_aux when training, the x_aux...
 like this
In issue #122 , the OP point out that the scale weights are not on the same device as the losses. I am guessing a fix was made such that...
File "train.py", line 172, in train loss = multi_scale_cross_entropy2d(input=outputs, target=labels, device = device) File "/home/victor/catkin_ws/src/semantic_slam/semantic_cloud/include/ptsemseg/loss.py", line 75, in multi_scale_cross_entropy2d loss = loss + scale_weight[i] * cross_entropy2d(input=inp, target=target, weight=weight, size_average=size_average) File...
Hi! Is there any method to evaluate your pspnet using voc2012 val data and pretrained caffe model? Thanks alot!
