from this link : https://github.com/bentrevett/pytorch-sentiment-analysis/blob/master/4%20-%20Convolutional%20Sentiment%20Analysis.ipynb
You will see they use this code for generate iterator.
import torch
from torchtext import data
from torchtext import datasets
import random
SEED = 1234
torch.manual_seed(SEED)
torch.cuda.manual_seed(SEED)
TEXT = data.Field(tokenize='spacy')
LABEL = data.LabelField(tensor_type=torch.FloatTensor)
train, test = datasets.IMDB.splits(TEXT, LABEL)
train, valid = train.split(random_state=random.seed(SEED))
BATCH_SIZE = 64
train_iterator, valid_iterator, test_iterator = data.BucketIterator.splits(
(train, valid, test),
batch_size=BATCH_SIZE,
sort_key=lambda x: len(x.text),
repeat=False)
Is it possible to create my iterator with the torchtext function