Some Question

[LemonGrate]

Hi! I met some difficulty when I training the model：
First，in the dataset_val: one is the root about masks and images; the other is whether should we resize the input into the input_size or should we have other processes?
Second, when I training the model, the eval results always 0.0, did I make some mistakes during the training?
Thanks for your patience!

[Wi-sc]

Hi, there. Thanks for your interest. For your first question, the parameters of the dataset.py are image and mask paths, the input size, and the mean and standard deviation of images. You can directly use default parameters about size, mean, and std. For your second question, I think you made some mistakes. You can check if the training loss decreases as long as training and visualize the evaluation outputs to debug.