Lesson 3 IMDB example about course-v3 HOT 7 CLOSED

me too

from course-v3.

I found I commented the line '#.filter_missing_y()', it will get right.
I think this notebook use old version fastai library.

from course-v3.

ah yeah I note that the latest notebook has that line removed as well and I have tested now and it does make the difference

from course-v3.

@Data-drone, @JerryLuYujie,
When I run the second cell in Classifier

data_clas = (TextList.from_folder(path, vocab=data_lm.vocab)
             #grab all the text files in path
             .split_by_folder(valid='test')
             #split by train and valid folder (that only keeps 'train' and 'test' so no need to filter)
             .label_from_folder(classes=['neg', 'pos'])
             #label them all with their folders
             .databunch(bs=bs))

data_clas.save('tmp_clas')

I get this error:

Exception: Your validation data contains a label that isn't present in the training set, please fix your data.

KeyError                                  Traceback (most recent call last)
~/Documents/fastai/courses/v3_dl1/fastai/data_block.py in process_one(self, item)
    281     def process_one(self,item):
--> 282         try: return self.c2i[item] if item is not None else None
    283         except:

KeyError: 'unsup'

Do you see the same?

from course-v3.

@shirishr I had get the same error as your's. Have you solved it?
Env:
ubuntu18.04
python3.7
fastai 1.0.38

from course-v3.

Just remove the unsub directory from train to some other outside location temporarily. This way train and test both have pos and neg only. Then that error goes away.

from course-v3.

@shirishr thanks.

from course-v3.