Hi,
I am trying to use this project to predict the structural labels of sentences in the abstracts of my dataset. But I found that the dataset for model input is of batch size. But when I use the batches as the input, the batch_creator would shuffle the order of the sentences, then I could not identify which predicted label is matched with the corresponding sentence. I wonder if there is anyway I could close the shuffle function in the bucketing.py document?
I would be very appreciate if anyone could help to solve my problem.
Hi,
I am trying to use this project to predict the structural labels of sentences in the abstracts of my dataset. But I found that the dataset for model input is of batch size. But when I use the batches as the input, the batch_creator would shuffle the order of the sentences, then I could not identify which predicted label is matched with the corresponding sentence. I wonder if there is anyway I could close the shuffle function in the bucketing.py document?
I would be very appreciate if anyone could help to solve my problem.