well, the code in training file is:
train_dataset = gluoncv.data.LstDetection(lst_file_path)
train_loader = gluon.data.DataLoader(train_dataset.transform(YOLO3DefaultTrainTransform(width, height, net, mixup=args.mixup)), batch_size, True, batchify_fn=batchify_fn, last_batch='rollover', num_workers=num_workers)
when I print the data shape, I find in class LstDetection(Dataset)
, the label shape is (1,6) which means (xmin, ymin, xmax, ymax, ids, difficult), but in class YOLO3DefaultTrainTransform(object)
, the label shape become (2, 7).
I was confused.