Hey guys,
I’ve been reading the fastaiv2 paper https://arxiv.org/pdf/2002.04688v2.pdf and putting code snippets into google colab notebook. When I ran code in 2.1. Vision
“”"
dls=ImageDataLoaders.from_name_re(path=path,bs=64,
fnames=get_image_files(path/“images”),pat=r’ /([^/]+)_\d+.jpg$’,
item_tfms=RandomResizedCrop(450,min_scale=0.75),
batch_tfms=[*aug_transforms(size=224,max_warp=0.),Normalize.from_stats(*imagenet_stats)])
“”"
I kept getting re.compile errors. I’ve only managed to resolve them by replacing r’ /([^/]+)\d+.jpg$’ with
r’ ([^/]+)\d+.jpg$’
I’m not sure whether there is something I’ve missed and the code should be working as intended, but I’m just putting it out here just in case.
Apologies in advance if I’ve posted in wrong section, as I’m not super familiar with the forum layout.
Cheers 
P.S.
fastaiv2 and fastcore installed from clones.
torch 1.4
python 3.6.9