Models to perform neural summarization (extractive and abstractive) using machine learning transformers and a tool to convert abstractive summarization datasets to the extractive task.
Separate predict_sentences from predict which works on input text.
Return list when raw_scores=True instead of dict. This is because we might have the same text such as "Thank you." appear in multiple places in the text. However, their score might be different since they are at different positions/context. Using a dict overwrites the previous position's score.
Changes:
predict_sentences
frompredict
which works on input text.list
when raw_scores=True instead ofdict
. This is because we might have the same text such as "Thank you." appear in multiple places in the text. However, their score might be different since they are at different positions/context. Using adict
overwrites the previous position's score.