HYOJINPARK

> Setting --visualize to True doesn't appear to do anything. Do you mean that the visdom not working? when you run : python -m visdom.server ???

Oh yes I did not add the visualization code in Test_model.py You can edit by referring the main.py like this (line 487) ```shell if args.visualize: if train_config["loss"] == "Lovasz": grid_outputs...

Hello @NhatRio save_est is the output of the validation function. save_est is an estimation result from model. Thanks for your asking On Wed, 19 Aug 2020 at 16:25, NhatRio wrote:...

Could you explain more detailed thing? It is more helpful to explain which line get error

In my case : ``` Link512DATA └── Nukki └── baidu_V1 ├── input ├── target ├── train.txt └── val.txt └── baidu_V2 ├── input ├── target ├── train.txt └── val.txt └── Portrait...

If you use Lovasz loss, the final channel is 1. However, if you use cross-entropy loss, the channel size is 2

@freaad Oh..I am really sorry to late answer. yes change channel size. I guess generally it is possible. but the some small issue is making edge properly. In my code...

@avinash31d In here, I used normal CE loss and you can change it by modifying json file

@avinash31d You can use my model, I uploaded the pth file

Under my understanding, the meaning of batch normalization is normalization of feature value distribution. Therefore, batch normalization is located after each convolution layer. With same reason, I also add batch...