dhecloud / hand-pose-for-rheumatoid-arthritis

Hi 👋, I'm Andrew

hand-pose-for-rheumatoid-arthritis's Issues

About model training

Hi, thanks for you sharing the code. I trained the model as the instruction in readme, but the model I got have a poor performance on train dataset when I try to visualize the joint. Is there anything I should do besides modifying the MSRA data path before training the model?

train code error

when I run python main.py --name experiment_1 --save_dir experiments/ --batchSize 128

File "main.py", line 323, in
main(args)
File "main.py", line 144, in main
loss_train = train(train_loader, model, criterion, optimizer, epoch, args)
File "main.py", line 182, in train
for i, (input, target) in enumerate(train_loader):
File "/usr/local/lib/python2.7/dist-packages/torch/utils/data/dataloader.py", line 264, in next
batch = self.collate_fn([self.dataset[i] for i in indices])
File "/home/lxd/NET/Hand-Pose-for-Rheumatoid-Arthritis-master/pytorch/MSRADataset.py", line 51, in getitem
depth,joint = data_translate_chance(depth,joint, p = self.p)
File "/home/lxd/NET/Hand-Pose-for-Rheumatoid-Arthritis-master/pytorch/MSRADataset.py", line 185, in data_translate_chance
depth = cv2.warpAffine(depth,M,(cols,rows), cv2.BORDER_REPLICATE)
TypeError: Expected cv::UMat for argument 'dst'