In section Sentiment, datamodel md2 is built from splits as:
md2 = TextData.from_splits(PATH, splits, bs)
If I chose not to do this and continue with datamodel built earlier as LanguageModelData from text_files,
(I am basically using learner i/o m3 in rest of the code), I get this error:
TypeError: eq received an invalid combination of arguments - got (Variable), but expected one of:
- (int value)
didn’t match because some of the arguments have invalid types: (Variable) - (torch.cuda.LongTensor other)
didn’t match because some of the arguments have invalid types: (Variable)
This error is from site-packages\torch\tensor.py (line 360 return self.eq(other))
Has anyone else seen such an error?