Tried running Resnet34 model, but got the below error.
Databunch was created as follows:
data = (
ImageItemList
.from_folder(path/‘train’)
.random_split_by_pct(seed=SEED)
.label_from_func(lambda path: fn2label[path.name])
.add_test(ImageItemList.from_folder(path/‘test’))
.transform(get_transforms(do_flip=False, max_zoom=1, max_warp=0, max_rotate=25), size=SZ, resize_method=ResizeMethod.SQUISH)
.databunch(bs=BS, num_workers=NUM_WORKERS, path=path)
)
This is a multi-class problem. Some of the images are only 1, is it due to image class present in train but not in validate dataset
Can someone please help!!