Hello fellow fastai users,
a friend of mine and I want to do some bounding box prediction on the new OpenImageV5 dataset. We are however struggling to create the databunch to train our model.
This minimum working example:
from fastai.vision import *
from pathlib import Path
PATH = Path('path')
paths = [PATH/'to/train/file1.jpg', PATH/'to/train/file2.jpg', PATH/'to/train/file3.jpg']
bboxes = [
[[[32, 1, 4, 5], [89, 5, 37, 5]], ['Footwear', 'Person']],
[[[32, 2, 14, 35]], ['Table']],
[[[2, 132, 41, 51], [9, 25, 39, 3]], ['Cup', 'Plate']],
]
paths = [str(item) for item in paths]
img2bbox = dict(zip(paths, bboxes))
img2bbox.keys()
get_y_func = lambda o: img2bbox[str(o)]
df = pd.DataFrame({"Path": ['to/train/file1.jpg', 'to/train/file2.jpg', 'to/train/file3.jpg'],
"ID": [item.split('/')[-1].split('.')[0] for item in paths]}, columns=["Path", "ID"])
def get_data(bs, size):
src = ImageList.from_df(df, path=PATH)
src = src.split_none()
src = src.label_from_func(get_y_func)
return src.databunch(path=PATH, bs=bs, collate_fn=bb_pad_collate)
data = get_data(bs=1, size=128)
reproduces the error we encounter:
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-1-78c0fbf390e6> in <module>()
25 return src.databunch(path=PATH, bs=bs, collate_fn=bb_pad_collate)
26
---> 27 data = get_data(bs=1, size=128)
<ipython-input-1-78c0fbf390e6> in get_data(bs, size)
22 src = ImageList.from_df(df, path=PATH)
23 src = src.split_none()
---> 24 src = src.label_from_func(get_y_func)
25 return src.databunch(path=PATH, bs=bs, collate_fn=bb_pad_collate)
26
~/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/fastai/data_block.py in _inner(*args, **kwargs)
466 self.valid = fv(*args, from_item_lists=True, **kwargs)
467 self.__class__ = LabelLists
--> 468 self.process()
469 return self
470 return _inner
~/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/fastai/data_block.py in process(self)
520 "Process the inner datasets."
521 xp,yp = self.get_processors()
--> 522 for ds,n in zip(self.lists, ['train','valid','test']): ds.process(xp, yp, name=n)
523 #progress_bar clear the outputs so in some case warnings issued during processing disappear.
524 for ds in self.lists:
~/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/fastai/data_block.py in process(self, xp, yp, name)
683 def process(self, xp:PreProcessor=None, yp:PreProcessor=None, name:str=None):
684 "Launch the processing on `self.x` and `self.y` with `xp` and `yp`."
--> 685 self.y.process(yp)
686 if getattr(self.y, 'filter_missing_y', False):
687 filt = array([o is None for o in self.y.items])
~/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/fastai/data_block.py in process(self, processor)
73 if processor is not None: self.processor = processor
74 self.processor = listify(self.processor)
---> 75 for p in self.processor: p.process(self)
76 return self
77
~/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/fastai/data_block.py in process(self, ds)
334
335 def process(self, ds):
--> 336 if self.classes is None: self.create_classes(self.generate_classes(ds.items))
337 ds.classes = self.classes
338 ds.c2i = self.c2i
~/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/fastai/data_block.py in generate_classes(self, items)
389 "Generate classes from `items` by taking the sorted unique values."
390 classes = set()
--> 391 for c in items: classes = classes.union(set(c))
392 classes = list(classes)
393 classes.sort()
TypeError: unhashable type: 'list'
We tried around a lot but somehow this error keeps popping up. We have been looking at the Pascal-notebook since it seems to be doing mostly the same thing except that it works…
