getting 'ValueError-TextInputSequence must be str' in 'train_dataset = train_dataset.map(convert_to_features)'

[hemantwani]

Hello friends,
I want to fine tune longformer QnA for different data. I found this code very relevant to my requirement, so was trying to understand the same thoroughly. but getting an error.
i am trying to run the same code that Suraj has on github.

**

'ValueError-TextInputSequence must be str' in 'train_dataset = train_dataset.map(convert_to_features)'.

**

`train_dataset = nlp.load_dataset('squad', split=nlp.Split.TRAIN)
valid_dataset = nlp.load_dataset('squad', split=nlp.Split.VALIDATION)

train_dataset = train_dataset.map(convert_to_features)
valid_dataset = valid_dataset.map(convert_to_features, load_from_cache_file=False)

#patil_suraj

[hemantwani]

@patil-suraj
Can you help me with this?