How to use torchtext?

"""
    http://www.nlpuser.com/pytorch/2018/10/30/useTorchText/
    http://anie.me/On-Torchtext/
"""

import pandas as pd
from torchtext import data

def get_dataset(data_, text_field, label_field, test=False):
	fields = [('id',None),('comment',text_field),('label', label_field)]
	examples = []
	if test:
		for text in data_['comment']:
			examples.append(data.Example.fromlist([None, text, None],fields))
	else:
		for text, label in (zip(data_['comment'], data_['label'])):
			examples.append(data.Example.fromlist([None, text, label],fields))
	return examples, fields

def get_data_iter(data):
    """
        simulation function with yield
    """
    for i in range(len(data)):
        yield data[i]

if __name__ == '__main__':
    
   # data = [1,3,4,5,5,6,6]
   # data_iter = get_data_iter(data)
   # for idx, batch in enumerate(data_iter):
   #     sample = batch
   #     print(sample)

    train = pd.read_csv('data/train.csv', sep='\t')
    test = pd.read_csv('data/test.csv', sep='\t')

    tokenizer = lambda x: x.split()
    TEXT = data.Field(sequential=True, tokenize=tokenizer, lower=True)
    LABEL = data.Field(sequential=False, use_vocab=False)
    train_examples, train_fields = get_dataset(train,TEXT, LABEL)
    test_examples, test_fields = get_dataset(test,TEXT, None, True)

    train_ = data.Dataset(train_examples, train_fields)
    test_ = data.Dataset(test_examples, test_fields)

    TEXT.build_vocab(train_)

    train_batches = data.BucketIterator(train_, batch_size=3, device=-1, sort_key=lambda x: len(x.comment), sort_within_batch=True, repeat=False)

    test_batches = data.Iterator(test_, batch_size=4, device=-1, sort=False, repeat=False)


    for idx, batch in enumerate(train_batches):
        comment, label = batch.comment, batch.label
        print(comment.shape, label.shape)