14.9. The Dataset for Pretraining BERT¶ Open the notebook in SageMaker Studio Lab
To pretrain the BERT model as implemented in Section 14.8, we need to generate the dataset in the ideal format to facilitate the two pretraining tasks: masked language modeling and next sentence prediction. On one hand, the original BERT model is pretrained on the concatenation of two huge corpora BookCorpus and English Wikipedia (see Section 14.8.5), making it hard to run for most readers of this book. On the other hand, the off-the-shelf pretrained BERT model may not fit for applications from specific domains like medicine. Thus, it is getting popular to pretrain BERT on a customized dataset. To facilitate the demonstration of BERT pretraining, we use a smaller corpus WikiText-2 ().
Comparing with the PTB dataset used for pretraining word2vec in Section 14.3, WikiText-2 (i) retains the original punctuation, making it suitable for next sentence prediction; (ii) retains the original case and numbers; (iii) is over twice larger.
import os
import random
from mxnet import gluon, np, npx
from d2l import mxnet as d2l
npx.set_np()
import os
import random
import torch
from d2l import torch as d2l
In the WikiText-2 dataset, each line represents a paragraph where space is inserted between any punctuation and its preceding token. Paragraphs with at least two sentences are retained. To split sentences, we only use the period as the delimiter for simplicity. We leave discussions of more complex sentence splitting techniques in the exercises at the end of this section.
#@save
d2l.DATA_HUB['wikitext-2'] = (
'https://s3.amazonaws.com/research.metamind.io/wikitext/'
'wikitext-2-v1.zip', '3c914d17d80b1459be871a5039ac23e752a53cbe')
#@save
def _read_wiki(data_dir):
file_name = os.path.join(data_dir, 'wiki.train.tokens')
with open(file_name, 'r') as f:
lines = f.readlines()
# Uppercase letters are converted to lowercase ones
paragraphs = [line.strip().lower().split(' . ')
for line in lines if len(line.split(' . ')) >= 2]
random.shuffle(paragraphs)
return paragraphs
#@save
d2l.DATA_HUB['wikitext-2'] = (
'https://s3.amazonaws.com/research.metamind.io/wikitext/'
'wikitext-2-v1.zip', '3c914d17d80b1459be871a5039ac23e752a53cbe')
#@save
def _read_wiki(data_dir):
file_name = os.path.join(data_dir, 'wiki.train.tokens')
with open(file_name, 'r') as f:
lines = f.readlines()
# Uppercase letters are converted to lowercase ones
paragraphs = [line.strip().lower().split(' . ')
for line in lines if len(line.split(' . ')) >= 2]
random.shuffle(paragraphs)
return paragraphs
14.9.1. Defining Helper Functions for Pretraining Tasks¶
In the following, we begin by implementing helper functions for the two BERT pretraining tasks: next sentence prediction and masked language modeling. These helper functions will be invoked later when transforming the raw text corpus into the dataset of the ideal format to pretrain BERT.
14.9.1.1. Generating the Next Sentence Prediction Task¶
According to descriptions of Section 14.8.5.2, the
_get_next_sentence function generates a training example for the
binary classification task.
#@save
def _get_next_sentence(sentence, next_sentence, paragraphs):
if random.random() < 0.5:
is_next = True
else:
# `paragraphs` is a list of lists of lists
next_sentence = random.choice(random.choice(paragraphs))
is_next = False
return sentence, next_sentence, is_next
#@save
def _get_next_sentence(sentence, next_sentence, paragraphs):
if random.random() < 0.5:
is_next = True
else:
# `paragraphs` is a list of lists of lists
next_sentence = random.choice(random.choice(paragraphs))
is_next = False
return sentence, next_sentence, is_next
The following function generates training examples for next sentence
prediction from the input paragraph by invoking the
_get_next_sentence function. Here paragraph is a list of
sentences, where each sentence is a list of tokens. The argument
max_len specifies the maximum length of a BERT input sequence during
pretraining.
#@save
def _get_nsp_data_from_paragraph(paragraph, paragraphs, vocab, max_len):
nsp_data_from_paragraph = []
for i in range(len(paragraph) - 1):
tokens_a, tokens_b, is_next = _get_next_sentence(
paragraph[i], paragraph[i + 1], paragraphs)
# Consider 1 '<cls>' token and 2 '<sep>' tokens
if len(tokens_a) + len(tokens_b) + 3 > max_len:
continue
tokens, segments = d2l.get_tokens_and_segments(tokens_a, tokens_b)
nsp_data_from_paragraph.append((tokens, segments, is_next))
return nsp_data_from_paragraph
#@save
def _get_nsp_data_from_paragraph(paragraph, paragraphs, vocab, max_len):
nsp_data_from_paragraph = []
for i in range(len(paragraph) - 1):
tokens_a, tokens_b, is_next = _get_next_sentence(
paragraph[i], paragraph[i + 1], paragraphs)
# Consider 1 '<cls>' token and 2 '<sep>' tokens
if len(tokens_a) + len(tokens_b) + 3 > max_len:
continue
tokens, segments = d2l.get_tokens_and_segments(tokens_a, tokens_b)
nsp_data_from_paragraph.append((tokens, segments, is_next))
return nsp_data_from_paragraph
14.9.1.2. Generating the Masked Language Modeling Task¶
In order to generate training examples for the masked language modeling
task from a BERT input sequence, we define the following
_replace_mlm_tokens function. In its inputs, tokens is a list of
tokens representing a BERT input sequence, candidate_pred_positions
is a list of token indices of the BERT input sequence excluding those of
special tokens (special tokens are not predicted in the masked language
modeling task), and num_mlm_preds indicates the number of
predictions (recall 15% random tokens to predict). Following the
definition of the masked language modeling task in
Section 14.8.5.1, at each prediction position, the input may be
replaced by a special “<mask>” token or a random token, or remain
unchanged. In the end, the function returns the input tokens after
possible replacement, the token indices where predictions take place and
labels for these predictions.
#@save
def _replace_mlm_tokens(tokens, candidate_pred_positions, num_mlm_preds,
vocab):
# Make a new copy of tokens for the input of a masked language model,
# where the input may contain replaced '<mask>' or random tokens
mlm_input_tokens = [token for token in tokens]
pred_positions_and_labels = []
# Shuffle for getting 15% random tokens for prediction in the masked
# language modeling task
random.shuffle(candidate_pred_positions)
for mlm_pred_position in candidate_pred_positions:
if len(pred_positions_and_labels) >= num_mlm_preds:
break
masked_token = None
# 80% of the time: replace the word with the '<mask>' token
if random.random() < 0.8:
masked_token = '<mask>'
else:
# 10% of the time: keep the word unchanged
if random.random() < 0.5:
masked_token = tokens[mlm_pred_position]
# 10% of the time: replace the word with a random word
else:
masked_token = random.choice(vocab.idx_to_token)
mlm_input_tokens[mlm_pred_position] = masked_token
pred_positions_and_labels.append(
(mlm_pred_position, tokens[mlm_pred_position]))
return mlm_input_tokens, pred_positions_and_labels
#@save
def _replace_mlm_tokens(tokens, candidate_pred_positions, num_mlm_preds,
vocab):
# Make a new copy of tokens for the input of a masked language model,
# where the input may contain replaced '<mask>' or random tokens
mlm_input_tokens = [token for token in tokens]
pred_positions_and_labels = []
# Shuffle for getting 15% random tokens for prediction in the masked
# language modeling task
random.shuffle(candidate_pred_positions)
for mlm_pred_position in candidate_pred_positions:
if len(pred_positions_and_labels) >= num_mlm_preds:
break
masked_token = None
# 80% of the time: replace the word with the '<mask>' token
if random.random() < 0.8:
masked_token = '<mask>'
else:
# 10% of the time: keep the word unchanged
if random.random() < 0.5:
masked_token = tokens[mlm_pred_position]
# 10% of the time: replace the word with a random word
else:
masked_token = random.choice(vocab.idx_to_token)
mlm_input_tokens[mlm_pred_position] = masked_token
pred_positions_and_labels.append(
(mlm_pred_position, tokens[mlm_pred_position]))
return mlm_input_tokens, pred_positions_and_labels
By invoking the aforementioned _replace_mlm_tokens function, the
following function takes a BERT input sequence (tokens) as an input
and returns indices of the input tokens (after possible token
replacement as described in Section 14.8.5.1), the token indices
where predictions take place, and label indices for these predictions.
#@save
def _get_mlm_data_from_tokens(tokens, vocab):
candidate_pred_positions = []
# `tokens` is a list of strings
for i, token in enumerate(tokens):
# Special tokens are not predicted in the masked language modeling
# task
if token in ['<cls>', '<sep>']:
continue
candidate_pred_positions.append(i)
# 15% of random tokens are predicted in the masked language modeling task
num_mlm_preds = max(1, round(len(tokens) * 0.15))
mlm_input_tokens, pred_positions_and_labels = _replace_mlm_tokens(
tokens, candidate_pred_positions, num_mlm_preds, vocab)
pred_positions_and_labels = sorted(pred_positions_and_labels,
key=lambda x: x[0])
pred_positions = [v[0] for v in pred_positions_and_labels]
mlm_pred_labels = [v[1] for v in pred_positions_and_labels]
return vocab[mlm_input_tokens], pred_positions, vocab[mlm_pred_labels]
#@save
def _get_mlm_data_from_tokens(tokens, vocab):
candidate_pred_positions = []
# `tokens` is a list of strings
for i, token in enumerate(tokens):
# Special tokens are not predicted in the masked language modeling
# task
if token in ['<cls>', '<sep>']:
continue
candidate_pred_positions.append(i)
# 15% of random tokens are predicted in the masked language modeling task
num_mlm_preds = max(1, round(len(tokens) * 0.15))
mlm_input_tokens, pred_positions_and_labels = _replace_mlm_tokens(
tokens, candidate_pred_positions, num_mlm_preds, vocab)
pred_positions_and_labels = sorted(pred_positions_and_labels,
key=lambda x: x[0])
pred_positions = [v[0] for v in pred_positions_and_labels]
mlm_pred_labels = [v[1] for v in pred_positions_and_labels]
return vocab[mlm_input_tokens], pred_positions, vocab[mlm_pred_labels]
14.9.2. Transforming Text into the Pretraining Dataset¶
Now we are almost ready to customize a Dataset class for pretraining
BERT. Before that, we still need to define a helper function
_pad_bert_inputs to append the special “<mask>” tokens to the
inputs. Its argument examples contain the outputs from the helper
functions _get_nsp_data_from_paragraph and
_get_mlm_data_from_tokens for the two pretraining tasks.
#@save
def _pad_bert_inputs(examples, max_len, vocab):
max_num_mlm_preds = round(max_len * 0.15)
all_token_ids, all_segments, valid_lens, = [], [], []
all_pred_positions, all_mlm_weights, all_mlm_labels = [], [], []
nsp_labels = []
for (token_ids, pred_positions, mlm_pred_label_ids, segments,
is_next) in examples:
all_token_ids.append(np.array(token_ids + [vocab['<pad>']] * (
max_len - len(token_ids)), dtype='int32'))
all_segments.append(np.array(segments + [0] * (
max_len - len(segments)), dtype='int32'))
# `valid_lens` excludes count of '<pad>' tokens
valid_lens.append(np.array(len(token_ids), dtype='float32'))
all_pred_positions.append(np.array(pred_positions + [0] * (
max_num_mlm_preds - len(pred_positions)), dtype='int32'))
# Predictions of padded tokens will be filtered out in the loss via
# multiplication of 0 weights
all_mlm_weights.append(
np.array([1.0] * len(mlm_pred_label_ids) + [0.0] * (
max_num_mlm_preds - len(pred_positions)), dtype='float32'))
all_mlm_labels.append(np.array(mlm_pred_label_ids + [0] * (
max_num_mlm_preds - len(mlm_pred_label_ids)), dtype='int32'))
nsp_labels.append(np.array(is_next))
return (all_token_ids, all_segments, valid_lens, all_pred_positions,
all_mlm_weights, all_mlm_labels, nsp_labels)
#@save
def _pad_bert_inputs(examples, max_len, vocab):
max_num_mlm_preds = round(max_len * 0.15)
all_token_ids, all_segments, valid_lens, = [], [], []
all_pred_positions, all_mlm_weights, all_mlm_labels = [], [], []
nsp_labels = []
for (token_ids, pred_positions, mlm_pred_label_ids, segments,
is_next) in examples:
all_token_ids.append(torch.tensor(token_ids + [vocab['<pad>']] * (
max_len - len(token_ids)), dtype=torch.long))
all_segments.append(torch.tensor(segments + [0] * (
max_len - len(segments)), dtype=torch.long))
# `valid_lens` excludes count of '<pad>' tokens
valid_lens.append(torch.tensor(len(token_ids), dtype=torch.float32))
all_pred_positions.append(torch.tensor(pred_positions + [0] * (
max_num_mlm_preds - len(pred_positions)), dtype=torch.long))
# Predictions of padded tokens will be filtered out in the loss via
# multiplication of 0 weights
all_mlm_weights.append(
torch.tensor([1.0] * len(mlm_pred_label_ids) + [0.0] * (
max_num_mlm_preds - len(pred_positions)),
dtype=torch.float32))
all_mlm_labels.append(torch.tensor(mlm_pred_label_ids + [0] * (
max_num_mlm_preds - len(mlm_pred_label_ids)), dtype=torch.long))
nsp_labels.append(torch.tensor(is_next, dtype=torch.long))
return (all_token_ids, all_segments, valid_lens, all_pred_positions,
all_mlm_weights, all_mlm_labels, nsp_labels)
Putting the helper functions for generating training examples of the two
pretraining tasks, and the helper function for padding inputs together,
we customize the following _WikiTextDataset class as the WikiText-2
dataset for pretraining BERT. By implementing the
__getitem__function, we can arbitrarily access the pretraining
(masked language modeling and next sentence prediction) examples
generated from a pair of sentences from the WikiText-2 corpus.
The original BERT model uses WordPiece embeddings whose vocabulary size
is 30000 (). The tokenization method of
WordPiece is a slight modification of the original byte pair encoding
algorithm in Section 14.6.2. For simplicity, we
use the d2l.tokenize function for tokenization. Infrequent tokens
that appear less than five times are filtered out.
#@save
class _WikiTextDataset(gluon.data.Dataset):
def __init__(self, paragraphs, max_len):
# Input `paragraphs[i]` is a list of sentence strings representing a
# paragraph; while output `paragraphs[i]` is a list of sentences
# representing a paragraph, where each sentence is a list of tokens
paragraphs = [d2l.tokenize(
paragraph, token='word') for paragraph in paragraphs]
sentences = [sentence for paragraph in paragraphs
for sentence in paragraph]
self.vocab = d2l.Vocab(sentences, min_freq=5, reserved_tokens=[
'<pad>', '<mask>', '<cls>', '<sep>'])
# Get data for the next sentence prediction task
examples = []
for paragraph in paragraphs:
examples.extend(_get_nsp_data_from_paragraph(
paragraph, paragraphs, self.vocab, max_len))
# Get data for the masked language model task
examples = [(_get_mlm_data_from_tokens(tokens, self.vocab)
+ (segments, is_next))
for tokens, segments, is_next in examples]
# Pad inputs
(self.all_token_ids, self.all_segments, self.valid_lens,
self.all_pred_positions, self.all_mlm_weights,
self.all_mlm_labels, self.nsp_labels) = _pad_bert_inputs(
examples, max_len, self.vocab)
def __getitem__(self, idx):
return (self.all_token_ids[idx], self.all_segments[idx],
self.valid_lens[idx], self.all_pred_positions[idx],
self.all_mlm_weights[idx], self.all_mlm_labels[idx],
self.nsp_labels[idx])
def __len__(self):
return len(self.all_token_ids)
#@save
class _WikiTextDataset(torch.utils.data.Dataset):
def __init__(self, paragraphs, max_len):
# Input `paragraphs[i]` is a list of sentence strings representing a
# paragraph; while output `paragraphs[i]` is a list of sentences
# representing a paragraph, where each sentence is a list of tokens
paragraphs = [d2l.tokenize(
paragraph, token='word') for paragraph in paragraphs]
sentences = [sentence for paragraph in paragraphs
for sentence in paragraph]
self.vocab = d2l.Vocab(sentences, min_freq=5, reserved_tokens=[
'<pad>', '<mask>', '<cls>', '<sep>'])
# Get data for the next sentence prediction task
examples = []
for paragraph in paragraphs:
examples.extend(_get_nsp_data_from_paragraph(
paragraph, paragraphs, self.vocab, max_len))
# Get data for the masked language model task
examples = [(_get_mlm_data_from_tokens(tokens, self.vocab)
+ (segments, is_next))
for tokens, segments, is_next in examples]
# Pad inputs
(self.all_token_ids, self.all_segments, self.valid_lens,
self.all_pred_positions, self.all_mlm_weights,
self.all_mlm_labels, self.nsp_labels) = _pad_bert_inputs(
examples, max_len, self.vocab)
def __getitem__(self, idx):
return (self.all_token_ids[idx], self.all_segments[idx],
self.valid_lens[idx], self.all_pred_positions[idx],
self.all_mlm_weights[idx], self.all_mlm_labels[idx],
self.nsp_labels[idx])
def __len__(self):
return len(self.all_token_ids)
By using the _read_wiki function and the _WikiTextDataset class,
we define the following load_data_wiki to download and WikiText-2
dataset and generate pretraining examples from it.
#@save
def load_data_wiki(batch_size, max_len):
"""Load the WikiText-2 dataset."""
num_workers = d2l.get_dataloader_workers()
data_dir = d2l.download_extract('wikitext-2', 'wikitext-2')
paragraphs = _read_wiki(data_dir)
train_set = _WikiTextDataset(paragraphs, max_len)
train_iter = gluon.data.DataLoader(train_set, batch_size, shuffle=True,
num_workers=num_workers)
return train_iter, train_set.vocab
#@save
def load_data_wiki(batch_size, max_len):
"""Load the WikiText-2 dataset."""
num_workers = d2l.get_dataloader_workers()
data_dir = d2l.download_extract('wikitext-2', 'wikitext-2')
paragraphs = _read_wiki(data_dir)
train_set = _WikiTextDataset(paragraphs, max_len)
train_iter = torch.utils.data.DataLoader(train_set, batch_size,
shuffle=True, num_workers=num_workers)
return train_iter, train_set.vocab
Setting the batch size to 512 and the maximum length of a BERT input sequence to be 64, we print out the shapes of a minibatch of BERT pretraining examples. Note that in each BERT input sequence, \(10\) (\(64 \times 0.15\)) positions are predicted for the masked language modeling task.
batch_size, max_len = 512, 64
train_iter, vocab = load_data_wiki(batch_size, max_len)
for (tokens_X, segments_X, valid_lens_x, pred_positions_X, mlm_weights_X,
mlm_Y, nsp_y) in train_iter:
print(tokens_X.shape, segments_X.shape, valid_lens_x.shape,
pred_positions_X.shape, mlm_weights_X.shape, mlm_Y.shape,
nsp_y.shape)
break
Downloading ../data/wikitext-2-v1.zip from https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-v1.zip...
(512, 64) (512, 64) (512,) (512, 10) (512, 10) (512, 10) (512,)
batch_size, max_len = 512, 64
train_iter, vocab = load_data_wiki(batch_size, max_len)
for (tokens_X, segments_X, valid_lens_x, pred_positions_X, mlm_weights_X,
mlm_Y, nsp_y) in train_iter:
print(tokens_X.shape, segments_X.shape, valid_lens_x.shape,
pred_positions_X.shape, mlm_weights_X.shape, mlm_Y.shape,
nsp_y.shape)
break
Downloading ../data/wikitext-2-v1.zip from https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-v1.zip...
torch.Size([512, 64]) torch.Size([512, 64]) torch.Size([512]) torch.Size([512, 10]) torch.Size([512, 10]) torch.Size([512, 10]) torch.Size([512])
In the end, let us take a look at the vocabulary size. Even after filtering out infrequent tokens, it is still over twice larger than that of the PTB dataset.
len(vocab)
20256
len(vocab)
20256
14.9.3. Summary¶
Comparing with the PTB dataset, the WikiText-2 dataset retains the original punctuation, case and numbers, and is over twice larger.
We can arbitrarily access the pretraining (masked language modeling and next sentence prediction) examples generated from a pair of sentences from the WikiText-2 corpus.
14.9.4. Exercises¶
For simplicity, the period is used as the only delimiter for splitting sentences. Try other sentence splitting techniques, such as the spaCy and NLTK. Take NLTK as an example. You need to install NLTK first:
pip install nltk. In the code, firstimport nltk. Then, download the Punkt sentence tokenizer:nltk.download('punkt'). To split sentences such assentences = 'This is great ! Why not ?', invokingnltk.tokenize.sent_tokenize(sentences)will return a list of two sentence strings:['This is great !', 'Why not ?'].What is the vocabulary size if we do not filter out any infrequent token?