IanTaehoonYoo semantic-segmentation-pytorch issues

IanTaehoonYoo / semantic-segmentation-pytorch

Pytorch implementation of FCN, UNet, PSPNet, and various encoder models.

MIT License

86 stars 19 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Only getting black image predictions when n_classes = 2

#23 bc-bytes opened 2 years ago
0
Assertion Error

#22 joydipb01 closed 2 years ago
1
performance issue

#21 soans1994 opened 3 years ago
0
how to make the network work well with num_class=2

#20 zhou-rui1 opened 3 years ago
0
Skip layer connections from the second layer

#19 zhou-rui1 opened 3 years ago
3
how to output the gray image without random.randint?

#18 zhou-rui1 opened 3 years ago
5
pspnet's encoder (resnet50) need dilation, however, I didn't find it.

#17 LeeNannaner opened 3 years ago
0
FCN8 object has no attribute 'img_width'

#16 rkuo2000 closed 3 years ago
1
Using on high channel image dataset

#15 torjusn opened 3 years ago
1
what is train_epoch_batch_.png represents?

#14 zhou-rui1 closed 3 years ago
3
Dear sir, I find one thing interesting,when I use finetune(pretrained =true; fixedfeature=ture), the result is worse than (pretrained =true; fixedfeature=false)

#13 LeeNannaner closed 3 years ago
3
sir,I find it(fcn8-vgg16) accuracy only 0.2657, how the result so terrible ?

#12 LeeNannaner closed 3 years ago
5
hi ,sir! I can't find 'dataclasses' in the project. Is it a package, or need me to create?

#11 LeeNannaner closed 3 years ago
1
How to add our own pretrained resnet?

#10 leocd91 closed 4 years ago
0
How can I get the location coordinate of detected object at inference time ?

#9 bemoregt opened 4 years ago
0
Scaling Error when training using different ratio size of image in batch_size > 1

#8 leocd91 opened 4 years ago
3
Error loading model then predict after updating to 0.1.7

#7 leocd91 closed 4 years ago
4
I met this error when run.

#4 bemoregt closed 4 years ago
0
: Given transposed=1, weight of size 1024 2048 4 4, expected input[8, 2048, 10, 10] to have 1024 channels, but got 2048 channels instead

#3 jiafengshen closed 4 years ago
2
How to make checkpoint and save the model?

#2 leocd91 closed 4 years ago
7
How can I make a "result image" from trained model ?

#1 bemoregt closed 4 years ago
1