Comments by Linlin Zeng (Top 5 by date)

Sorry for misunderstanding your reply; I am using following packages;

'''import spacy
import pandas as pd
from torchtext.legacy.data import Field, TabularDataset, BucketIterator, LabelField
from sklearn.model_selection import train_test_split
from torchtext.vocab import GloVe'''

for batch in dataset_iter:
batch.text[0]
batch.term_index[0]
it suppose to print:
torchtext.data.batch.Batch of size 10
.text torch.cuda.LongTensor of size 104*10
.term_index torch.cuda.LongTensor of size 104*10;
just tell the batch can be run successfully

TEXT.vocab.vectors.size() # torch.Size([4, 300])
I suppose when building vocab, there is something wrong
'''
import spacy
import pandas as pd
from torchtext.legacy.data import Field, TabularDataset, BucketIterator, LabelField
from sklearn.model_selection import train_test_split
from torchtext.vocab import GloVe
'''
This is all packages I used; not import bird

oh; actually, it is more helpful if you can tell me how to modify it

dataset_iter = Iterator(
train, batch_size=10,
train=True, shuffle= True)
here dataset_iter is an iteration of train after using batch size; train is a tabular dataset after using Field to transfer word into tensors.