Hello.
I’m trying to reproduce: https://github.com/keitakurita/Practical_NLP_in_PyTorch/blob/master/fastai/bert_with_fastai.ipynb
I’m trying to create a TextDataBunch from dataframe, and I don’t understand what is wrong.
That’s my code:
databunch = TextDataBunch.from_df(".", train_df, valid_df, test_df=test_df,
tokenizer=fastai_tokenizer,
vocab=fastai_bert_vocab,
include_bos=False,
include_eos=False,
text_cols=“question”,
label_cols=label_cols,
bs=config.bs,
collate_fn=partial(pad_collate, pad_first=False, pad_idx=0),
)
I receive the following error:
TypeError: int() argument must be a string, a bytes-like object or a number, not 'NoneType'
Those are the shapes:
train_df.shape, valid_df.shape, test_df.shape
Out[3]: ((3200, 2), (500, 2), (125, 2))
Each dataframe contains 2 columns: ‘question’ and ‘answer’, when the ‘answer’ is the label.
Why am I getting this error? How can I create this TextDataBunch? Thanks