AdeDZY / DeepCT

DeepCT and HDCT uses BERT to generate novel, context-aware bag-of-words term weights for documents and queries.
BSD 3-Clause "New" or "Revised" License
312 stars 46 forks source link

Question QTR & TR #16

Open ming19956 opened 3 years ago

ming19956 commented 3 years ago

In your training data, I found that they are all 1-to-1 query-passage pairs, so that the only possible values for QTR or TR are 0 or 1. And in this case, the values of QTR and TR are the same. Does this make QTR and TR meaningless? Is it possible to change the input data to a query-passages pair, so that the values of QTR or TR have more possibilities and higher values?