I’m getting an error and wondering if my regex isn’t working properly. Is there a way to debug ImageDataBunch so I can find out what it thinks are the labels?
You can deactivate this warning by passing `no_check=True`.
/opt/conda/envs/fastai/lib/python3.6/site-packages/fastai/basic_data.py:226: UserWarning: There seems to be something wrong with your dataset, can't access these elements in self.train_ds: 263,258
warn(warn_msg)You can access elements of training set via data.train_ds, and then the inputs/targets via data.train_ds.x and data.train_ds.y.
So that data looks good when I poke around at it. It has an ImageList and I can open the images. It has categories and they are what I expected. Would be nice if the error message was more descriptive. But then when I run the show_batch it fails.
Here’s a screen shot. One of the images is even of Elon Musk at kite beach in Maui. 
Any ideas why I would get the warning and why would show_batch fail? I checked the file permissions and they all readable. (Plus I can render them in the notebook, as Mr. Musk demonstrates.)
Doh, I see that later in the error stack that it has a problem with a specific image. I must have a broken jpeg.
Problem solved. Remove the broken image…
OSError: cannot identify image file
'/storage/adam/uninteresting/01f5ee43bea31694c5396a9481b88c3f4324c8f4.jpg'One more thing, you can run the images through verify_images() and it works like a charm. This patched up my images and I’m off to the races. So cool!
verify_images(path_img/'favorites')
verify_images(path_img/'uninteresting')Note that in your first warning, you had the specific indexes of bad images (at least the one it tried to access and fail, not necessarily all the bad ones).
data=ImageDataBunch.from_name_re(p_image,fnames,pat=pat,ds_tfms=get_transforms(),size=224)
data.normalize(imagenet_stats)
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
in
----> 1 data=ImageDataBunch.from_name_re(p_image,fnames,pat,ds_tfms=get_transforms(),size=224)
2 data.normalize(imagenet_stats)
~\Miniconda3\lib\site-packages\fastai\vision\data.py in from_name_re(cls, path, fnames, pat, valid_pct, **kwargs)
153 pat = re.compile(pat)
154 def _get_label(fn): return pat.search(str(fn)).group(1)
–> 155 return cls.from_name_func(path, fnames, _get_label, valid_pct=valid_pct, **kwargs)
156
157 @staticmethod
~\Miniconda3\lib\site-packages\fastai\vision\data.py in from_name_func(cls, path, fnames, label_func, valid_pct, **kwargs)
146 “Create from list of fnames in path with label_func.”
147 src = ImageItemList(fnames, path=path).random_split_by_pct(valid_pct)
–> 148 return cls.create_from_ll(src.label_from_func(label_func), **kwargs)
149
150 @classmethod
~\Miniconda3\lib\site-packages\fastai\data_block.py in _inner(*args, **kwargs)
369 assert isinstance(fv, Callable)
370 def _inner(*args, **kwargs):
–> 371 self.train = ft(*args, **kwargs)
372 assert isinstance(self.train, LabelList)
373 self.valid = fv(*args, **kwargs)
~\Miniconda3\lib\site-packages\fastai\data_block.py in label_from_func(self, func, **kwargs)
229 def label_from_func(self, func:Callable, **kwargs)->‘LabelList’:
230 “Apply func to every input to get its label.”
–> 231 return self.label_from_list([func(o) for o in self.items], **kwargs)
232
233 def label_from_folder(self, **kwargs)->‘LabelList’:
~\Miniconda3\lib\site-packages\fastai\data_block.py in (.0)
229 def label_from_func(self, func:Callable, **kwargs)->‘LabelList’:
230 “Apply func to every input to get its label.”
–> 231 return self.label_from_list([func(o) for o in self.items], **kwargs)
232
233 def label_from_folder(self, **kwargs)->‘LabelList’:
~\Miniconda3\lib\site-packages\fastai\vision\data.py in _get_label(fn)
152 “Create from list of fnames in path with re expression pat.”
153 pat = re.compile(pat)
–> 154 def _get_label(fn): return pat.search(str(fn)).group(1)
155 return cls.from_name_func(path, fnames, _get_label, valid_pct=valid_pct, **kwargs)
156
AttributeError: ‘NoneType’ object has no attribute ‘group’
