Open Hupo777 opened 9 months ago
Hello, have you encountered a situation during training where the loss doesn't converge, and both semantic loss and sb loss remain consistently at zero? If not, could you provide details on how you set the parameters during training?
We use our own data to train .
File "D:\A\segmentation\PIDNet-main\tools..\utils\utils.py", line 42, in forward ph, pw = outputs[0].size(2), outputs[0].size(3) IndexError: Dimension out of range (expected to be in range of [-3, 2], but got 3)
base_size and crop_size :512*512 IMAGE_SIZE:
Not using pre-trained models, the model in train.py: model = models.pidnet.get_pred_model(config,6)
Is this error caused by the size of the image?