Is there a way to use either of these text/nlp libraries to create a model with more than one text field? For example, I can create torchtext splits with two text fields using the modified arxiv custom torchtext class:
fields = [(‘text1’, text_field), (‘text2’, text_field), (‘label’, label_field)]
But when I try to create a model data object it will only accept one text and one label field like so:
md2 = TextData.from_splits(PATH, splits, bs)
Is there any way to use more than one text field? The use case here would be taking two separate blobs of text and trying to find those that “match” based on some criteria.
thanks!