Lavender105 / DFF

Code for Dynamic Feature Fusion for Semantic Edge Detection https://arxiv.org/abs/1902.09104

Lavender105/DFF Issues

In train.py, DataParallelModel(self.model) is not cuda() enabled.
Updated a year ago
RuntimeError: NCCL Error 13: invalid data type
Updated a year ago1
ModuleNotFoundError: No module named 'encoding.datasets.cityscapes_orig'
Updated a year ago1
处理NYUv2数据集
Updated a year ago
Released Pretrained model only get 0.714 Mean MF-ODS in Cityscapes Val Set
Closed 4 years ago1
ModuleNotFoundError: No module named 'encoding.utils.lr_scheduler_orig'
Updated a year ago9
What is the difference between training with instance-sensitive(inst)/non-instance-sensitive(cls) edge label?
Closed 4 years ago1
For Create Edge Maps
Updated 2 years ago3
Question about your loss function
Updated 3 years ago3
Upsampling differences between CaseNet Paper and Code
Updated 3 years ago
about the pretrained model?
Updated 3 years ago2
Has anynone succeed in getting the performance in the paper?
Updated 3 years ago3
how can I reach or close to the performance in the papper by training with single GPU?
Updated 3 years ago2
cannot get the perfomance of the paper
Updated 3 years ago
No module named 'enclib_cpu', how to solve this problem.
Updated 3 years ago1
ground truth 怎么生成的啊？求指导
Updated 3 years ago1
RuntimeError: Error(s) in loading state_dict for DFF
Updated 3 years ago
About the pretrained model downloaded from the BaiduYun and Google
Updated 4 years ago2
关于test的问题？
Closed 5 years ago2
Is pytorch-encoding required?
Updated 4 years ago
Thanks for you code. However, could you please tell me which file defines the network architecture in the PyTorch version?
Updated 4 years ago1
Build pytorch from the source on Windows
Updated 4 years ago
Inconsistent Code and Paper description
Closed 4 years ago
Def _sync_transform(self, img, mask):
Updated 5 years ago
RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn
Updated 5 years ago
RuntimeError: Error(s) in loading state_dict for ResNet:
Updated 5 years ago3
为什么_sync_transform img/255 但是_val_sync_transform img不归一化
Closed 5 years ago
Backward issues.
Closed 5 years ago1
Already-trained model
Updated 5 years ago1
ModuleNotFoundError: No module named 'torchvision'
Closed 5 years ago