-
已经将
ROOT= './dataset/CamVid'
BINARY_CLASS = 5
改为测试需要的值,并且写好读该数据的函数,测试用的loss为loss.py的MulticlassDiceLoss,因维度问题出现多出报错
` N, H, W = target.size(0), target.size(2), target.size(3)
IndexError: Dimens…
-
For the code in the "multiclass segmentation (camvid).ipynb" in "examples", the Unet uses batch normalization but does not use Dropout layers, right?
-
Hi, friend , I'm new to `semantic segmentation` , so I have to understand and test every step in your example code ` cars segmentation (camvid).ipynb`,
Can you tell me …
-
## 🐛 Bug
## To Reproduce
Steps to reproduce the behavior:
when i use the notebook from https://github.com/qubvel/segmentation_models/blob/master/examples/multiclass%20segmentation%20(camv…
-
can you please give an sample of class_dict.csv ?
-
In the process of building a tool to facilitate distributed training + `fastai v1` in Jupyter notebook, I realized `lr_find()` doesn't work in distributed mode, and it doesn't need a distributed learn…
-
@meetshah1995 @adam9500370 The pretrained models of ICNet come from https://github.com/meetshah1995/pytorch-semseg/pull/81. I do not how to use.
-
First off, i would like to say i really like rectlabel. It's a great tool. Im finding the codumentation slightly confusing - it would be great if there where some more demos / guides on youtube as som…
-
Hi @qubvel
I am looking for a trained model to detect objects on traffic scene(car,pedestrain,etc.)and the example code you provide is a good implementation of this scenario. But downloading the big…
-
Hi @ycszen
The FPS about BiSeNet in paper abstract is tested on a 2048x1024 input image is 105.
But, I just get 2 FPS about BiSeNet(Xception) and 9.5 FPS about BiSeNet(ResNet-18) on TiTan Xp.