-
-
## 🚀 Feature
Support ADE20k for semantic segmentation with deeplabv3+
## Motivation & Examples
This will be helpful when people need to benchmark different self-supervised learning method, sinc…
-
Maybe I'm not understanding something but with the following code summary throws an error:
```
import torch
from torchvision import models
from torchsummary import summary
device = torch.devi…
-
Hello, I am planning to buy a NVIDIA Jetson AGX and run a trained deelplabv3+ model on it.
However, when I search on google I found this [post](https://forums.developer.nvidia.com/t/dla-bugs-using-de…
fclof updated
3 years ago
-
ValueError: Shape must be rank 3 but is rank 4 for '{{node deeplabv3plus/concatenate_1/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32](deeplabv3plus/lambda_3/resize/Squeeze, deeplabv3plus/activati…
-
epoch: 2 iter: 659/662 lr: 0.006748 loss: 0.5660 (0.4378)
epoch: 2 iter: 660/662 lr: 0.006748 loss: 0.3179 (0.4366)
epoch: 2 iter: 661/662 lr: 0.006748 loss: 0.46…
Lmoer updated
5 years ago
-
Features from #134 are still present in GDL. Have they been used lately? Maybe time for deprecation.
-
up主你好，你的Deeplabv3+用的是voc数据集，
由于没有找到cityscape转voc的脚本
所以直接将cityscape转成coco数据集放进去训练，但结果moiu只有40左右
问题①：想问一下需不需要先cityscape转coco,然后再coco转voc进行训练？
问题②：用你提供的voc_mobilenetv2预训练权重会不会对cityscape效果造成影响？
在改进的deeplabv3+网络上进行的训练，标签也按照您的代码进行处理，满足【0，1】.但是训练的miou一直卡在49.14.mpa…
-
deeplabv3+ without xception base file ， very slow
can you share the base model file ?
or how to get it?
thx