机器翻译数据集
import osimport torchfrom d2l import torch as d2l
下载和预处理数据集
#@saved2l.DATA_HUB['fra-eng'] = (d2l.DATA_URL + 'fra-eng.zip','94646ad1522d915e7b0f9296181140edcf86a4f5')#@savedef read_data_nmt():"""载入“英语-法语”数据集"""data_dir = d2l.download_extract('fra-eng')with open(os.path.join(data_dir, 'fra.txt'), 'r',encoding='utf-8') as f:return f.read()raw_text = read_data_nmt()print(raw_text[:75])
Downloading D:\environment\data\data\fra-eng.zip from http://d2l-data.s3-/fra-eng.zip...Go.Va !Hi.Salut !Run!Cours !Run!Courez !Who?Qui ?Wow!Ça alors !
#@savedef preprocess_nmt(text):"""预处理“英语-法语”数据集"""def no_space(char, prev_char):return char in set(',.!?') and prev_char != ' '# 使用空格替换不间断空格# 使用小写字母替换大写字母text = text.replace('\u202f', ' ').replace('\xa0', ' ').lower()# 在单词和标点符号之间插入空格out = [' ' + char if i > 0 and no_space(char, text[i - 1]) else charfor i, char in enumerate(text)]return ''.join(out)text = preprocess_nmt(raw_text)print(text[:80])
go .va !hi .salut !run !cours !run !courez !who ?qui ?wow !ça alors !
词元化
#@savedef tokenize_nmt(text, num_examples=None):"""词元化“英语-法语”数据数据集"""source, target = [], []for i, line in enumerate(text.split('\n')):if num_examples and i > num_examples:breakparts = line.split('\t')if len(parts) == 2:source.append(parts[0].split(' '))target.append(parts[1].split(' '))return source, targetsource, target = tokenize_nmt(text)source[:6], target[:6]
([['go', '.'],['hi', '.'],['run', '!'],['run', '!'],['who', '?'],['wow', '!']],[['va', '!'],['salut', '!'],['cours', '!'],['courez', '!'],['qui', '?'],['ça', 'alors', '!']])
看看效果
def show_list_len_pair_hist(legend, xlabel, ylabel, xlist, ylist):"""绘制列表长度对的直方图"""d2l.set_figsize()_, _, patches = d2l.plt.hist([[len(l) for l in xlist], [len(l) for l in ylist]])d2l.plt.xlabel(xlabel)d2l.plt.ylabel(ylabel)for patch in patches[1].patches:patch.set_hatch('/')d2l.plt.legend(legend)show_list_len_pair_hist(['source', 'target'], '# tokens per sequence','count', source, target);
在这个简单的“英-法”数据集中,大多数文本序列的词元数量少于20个。
词表
src_vocab = d2l.Vocab(source, min_freq=2,reserved_tokens=['<pad>', '<bos>', '<eos>'])len(src_vocab)
‘’:填充符
‘’:开始符
‘’:结束符
10012
加载数据集
截断或填充
#@savedef truncate_pad(line, num_steps, padding_token):"""截断或填充文本序列"""if len(line) > num_steps:return line[:num_steps] # 截断return line + [padding_token] * (num_steps - len(line)) # 填充truncate_pad(src_vocab[source[0]], 10, src_vocab['<pad>'])
[47, 4, 1, 1, 1, 1, 1, 1, 1, 1]
#@savedef build_array_nmt(lines, vocab, num_steps):"""将机器翻译的文本序列转换成小批量"""lines = [vocab[l] for l in lines]lines = [l + [vocab['<eos>']] for l in lines]array = torch.tensor([truncate_pad(l, num_steps, vocab['<pad>']) for l in lines])valid_len = (array != vocab['<pad>']).type(torch.int32).sum(1)return array, valid_len
整合一下
#@savedef load_data_nmt(batch_size, num_steps, num_examples=600):"""返回翻译数据集的迭代器和词表"""text = preprocess_nmt(read_data_nmt())source, target = tokenize_nmt(text, num_examples)src_vocab = d2l.Vocab(source, min_freq=2,reserved_tokens=['<pad>', '<bos>', '<eos>'])tgt_vocab = d2l.Vocab(target, min_freq=2,reserved_tokens=['<pad>', '<bos>', '<eos>'])src_array, src_valid_len = build_array_nmt(source, src_vocab, num_steps)tgt_array, tgt_valid_len = build_array_nmt(target, tgt_vocab, num_steps)data_arrays = (src_array, src_valid_len, tgt_array, tgt_valid_len)data_iter = d2l.load_array(data_arrays, batch_size)return data_iter, src_vocab, tgt_vocab
测试一下加载数据
train_iter, src_vocab, tgt_vocab = load_data_nmt(batch_size=2, num_steps=8)for X, X_valid_len, Y, Y_valid_len in train_iter:print('X:', X.type(torch.int32))print('X的有效长度:', X_valid_len)print('Y:', Y.type(torch.int32))print('Y的有效长度:', Y_valid_len)break
输出
X: tensor([[71, 5, 3, 1, 1, 1, 1, 1],[ 0, 4, 3, 1, 1, 1, 1, 1]], dtype=torch.int32)X的有效长度: tensor([3, 3])Y: tensor([[98, 5, 3, 1, 1, 1, 1, 1],[ 0, 5, 3, 1, 1, 1, 1, 1]], dtype=torch.int32)Y的有效长度: tensor([3, 3])
看看各中间变量格式:
read_data_nmt()
:加载原始数据集
Go. Va !
Hi. Salut !
Run! Cours !
Run! Courez !
Who? Qui ?
Wow! Ça alors !
text = preprocess_nmt(read_data_nmt())
:大写变小写,在单词和标点符号之间插入空格
go . va !
hi . salut !
run ! cours !
run ! courez !
who ? qui ?
wow ! ça alors !
source, target = tokenize_nmt(text, 600)
:词元化
[[‘go’, ‘.’],
[‘hi’, ‘.’],
[‘run’, ‘!’],
[‘run’, ‘!’],
[‘who’, ‘?’],
[‘wow’, ‘!’],…]
src_array, src_valid_len = build_array_nmt(source, src_vocab, 10)
:
将词转换成词表中的对应下标,裁成长度为10的,加上填充和特舒符号。
tensor([[ 9, 4, 3, …, 1, 1, 1],
[113, 4, 3, …, 1, 1, 1],
[ 54, 5, 3, …, 1, 1, 1],
…,
[ 6, 112, 4, …, 1, 1, 1],
[ 6, 112, 4, …, 1, 1, 1],
[ 6, 0, 4, …, 1, 1, 1]])
src_valid_len
:除了填充每个样本有用的长度