pytorch记录：seq2seq例子看看这torch怎么玩的

先看看简单例子：

import torch
import torch.autograd as autograd
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

torch.manual_seed(1)

• 用torch.tensor让list成为tensor：

# Create a 3D tensor of size 2x2x2.
T_data = [[[1., 2.], [3., 4.]],
          [[5., 6.], [7., 8.]]]
T = torch.tensor(T_data)
print(T)

• 自动求导设requires_grad=True:

# Computation Graphs and Automatic Differentiation
x = torch.tensor([1., 2., 3], requires_grad=True)
y = torch.tensor([4., 5., 6], requires_grad=True)
z = x + y
print(z)
print(z.grad_fn)

tensor([ 5.,  7.,  9.])
<AddBackward1 object at 0x00000247781E0BE0>

• detach()方法获取z的值，但是不能对获取后的值求导了。

new_z = z.detach()
print(new_z.grad_fn)

None

• 好了，重点来了

Translation with a Sequence to Sequence Network and Attention

from __future__ import unicode_literals, print_function, division
from io import open
import unicodedata
import string
import re
import random

import torch
import torch.nn as nn
from torch import optim
import torch.nn.functional as F

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

准备数据：

SOS_token = 0
EOS_token = 1

class lang:
    def __init__(self, name):
        self.name = name
        self.word2index = {}
        self.word2count = {}
        self.index2word = {0:'SOS', 1:'EOS'}
        self.n_words = 2 # Count SOS and EOS

    def addSentence(self, sentence):
        for word in sentence.split():
            self.addWord(word)

    def addWord(self, word):
        if word not in self.word2index:
            self.word2index[word] = self.n_words
            self.word2count[word] = 1
            self.index2word[self.n_words] = word
            self.n_words += 1
        else:
            self.word2count[word] += 1

• Unicode字符转为ASCII,用小写字母表示一切，去掉标点符号：

# Turn a Unicode string to plain ASCII, thanks to
# http://stackoverflow.com/a/518232/2809427
def unicodeToAscii(s):
    return ''.join(
        c for c in unicodedata.normalize('NFD', s)
        if unicodedata.category(c) != 'Mn'
    )

# Lowercase,trim,remove non-letter characters
#re.sub(pattern, repl, string, count=0, flags=0)
def normalizeString(s):
    s = unicodeToAscii(s.lower().strip())
    # (re)  匹配括号内的表达式，也表示一个组
    # [...] 用来表示一组字符,单独列出：[amk] 匹配 'a'，'m'或'k'
    # \1...\9   匹配第n个分组的内容。
    s = re.sub(r"([.!?])", r"\1", s)
    # [^...]    不在[]中的字符：[^abc] 匹配除了a,b,c之外的字符。
    s = re.sub(r"[^a-zA-Z.!?]+",r" ", s)
    return s

继续：

# 文件用的英语到其他语言，用reverse标志置换一对这样的数据。
def readlangs(lang1, lang2, reverse= False):
    print("Reading lines...")

    #Read the file and split into lines
    lines = open('data/%s-%s.txt' % (lang1, lang2), encoding='utf-8').\
    read().strip().split('\n')
    # Split every line into pairs and normalize
    pairs = [[normalizeString(s) for s in l.split('\t')] for l in lines]

    # Reverse pairs, make lang instances
    if reverse:
        pairs = [list(reversed(p)) for p in pairs]
        input_lang = lang(lang2)
        output_lang = lang(lang1)
    else:
        input_lang = lang(lang1)
        output_lang = lang(lang2)

    return input_lang, output_lang, pairs

过滤出部分样本：

MAX_LENGTH = 10

eng_prefixes = (
    "i am ", "i m ",
    "he is", "he s ",
    "she is", "she s",
    "you are", "you re ",
    "we are", "we re ",
    "they are", "they re "
)

def filterPair(p):
    return len(p[0].split(' ')) < MAX_LENGTH and \
    len(p[1].split(' ')) < MAX_LENGTH and \
    p[1].startswith(eng_prefixes)

def filterPairs(pairs):
    return [ pair for pair in pairs if filterPair(pair)]

• The full process for preparing the data is:

  • Read text file and split into lines, split lines into pairs
  • Normalize text, filter by length and content
  • Make word lists from sentences in pairs

def prepareData(lang1, lang2, reverse= False):
    input_lang, output_lang, pairs = readlangs(lang1,lang2,reverse)
    print("Read %s sentence pairs " % len(pairs))
    pairs = filterPairs(pairs)
    print("Trimmed to %s sentence pairs " % len(pairs))
    print("Counting words...")
    for pair in pairs:
        input_lang.addSentence(pair[0])
        output_lang.addSentence(pair[1])
    print("Counted word:")
    print(input_lang.name,input_lang.n_words)
    print(output_lang.name, output_lang.n_words)
    return input_lang, output_lang, pairs

input_lang, output_lang, pairs = prepareData('eng','fra',True)
print(random.choice(pairs))


Reading lines...
Read 135842 sentence pairs 
Trimmed to 11739 sentence pairs 
Counting words...
Counted word:
fra 5911
eng 3965
['elle chante les dernieres chansons populaires.', 'she is singing the latest popular songs.']

The Seq2Seq Model

• 允许句子到句子有不同长度和顺序。

The Encoder ：

#编码器
class  EncoderRNN(nn.Module):
    def __init__(self, input_size, hidden_size):
        super(EncoderRNN, self).__init__()
        self.hidden_size = hidden_size

        # 指定embedding矩阵W的大小维度
        self.embedding = nn.Embedding(input_size, hidden_size)
        # 指定gru单元的大小
        self.gru = nn.GRU(hidden_size, hidden_size)

    def forward(self, input, hidden):
        # 扁平化嵌入矩阵
        embedded = self.embedding(input).view(1, 1, -1)
        print("embedded shape:",embedded.shape)
        output = embedded

        output, hidden = self.gru(output, hidden)
        return output, hidden

    #全0初始化隐层
    def initHidden(self):
        # 这个初始化维度可以
        return torch.zeros(1, 1, self.hidden_size, device=device)

这里的self.gru = nn.GRU(hidden_size, hidden_size)中，hidden_size在后面设置为256

print("embedded shape:",embedded.shape)的结果是：
embedded shape: torch.Size([1, 1, 256])

所以self.gru(output, hidden)中传递的第一个维度是[1,1,256],被压缩为这样的。

---

nn.GRU源码：

---

The Decoder:

• seq2seq解码器的简化版：指利用encoder的最后输出,称为context vector，
• context vector 作为decoder的初始化隐层状态值
  

class DecoderRNN(nn.Module):
    def self__init__(self, hidden_size, output_size):
        super(DecoderRNN, self).__init__()
        self.hidden_size = hidden_size

        self.embedding = nn.Embedding(output_size,hidden_size)
        self.gru = nn.GRU(hidden_size, hidden_size)
        self.out = nn.Linear(hidden_size, output_size)
        self.softmax = nn.LogSoftmax(dim=1)

    def forward(self, input, hidden):
        output = self.embedding(input).view(1, 1, -1)
         # 1行X列的shape做relu
        output = F.relu(output)
        output, hidden = self.gru(output, hidden)
        #output[0]应该是shape为（*，*）的矩阵
        output = self.softmax(self.out(output[0]))
        return output, hidden
    def initHidden(self):
        return torch.zeros(1, 1, self.hidden_size, device=device)

Attention Decoder：

• 简单的解码器的缺点：把整个句子做编码成一个向量，信息容易丢失，翻译一个词的时候需要追溯之前很长的距离，一般翻译的对应性也没有利用，如翻译第一个词，对应大概率在原句子的第一个位置的信息。
• encoder的输出向量 会乘以一个attention weights,这个权值用NN来计算完成attn，使用解码器的输入和隐藏状态作为输入。。
• 因为在训练数据中有各种大小的句子，为了实际创建和训练这一层，我们必须选择一个最大的句子长度（输入长度，对于编码器输出）因为在训练数据中有各种大小的句子，为了实际创建和训练这一层，我们必须选择一个最大的句子长度（输入长度，对于编码器输出）
  

class AttnDecoderRNN(nn.Module):
    def __init__(self, hidden_size, output_size, 
                dropout_p = 0.1, max_length=MAX_LENGTH):
        super(AttnDecoderRNN,self).__init__()
        self.hidden_size = hidden_size
        self.output_size = output_size
        self.dropout_p = dropout_p
        self.max_length = max_length

        self.embedding = nn.Embedding(self.output_size, self.hidden_size)
        self.attn = nn.Linear(self.hidden_size * 2, self.max_length)
        self.attn_combine = nn.Linear(self.hidden_size * 2, self.hidden_size)
        self.dropout = nn.Dropout(self.dropout_p)
        #输入向量的维度是10,隐层的长度是10,默认是一层GRU
        self.gru = nn.GRU(self.hidden_size, self.hidden_size)
        self.out = nn.Linear(self.hidden_size, self.output_size)

    def forward(self, input, hidden, encoder_outputs):
        embedded = self.embedding(input).view(1,1,-1)
        embedded = self.dropout(embedded)

        attn_weights = F.softmax(
            self.attn(torch.cat((embedded[0],hidden[0]),1)),dim=1)
        # unsqueeze:在指定的轴上多增加一个维度
        attn_applied = torch.bmm(attn_weights.unsqueeze(0),
                                encoder_outputs.unsqueeze(0))

        output = torch.cat((embedded[0],attn_applied[0]),1)
        output = self.attn_combine(output).unsqueeze(0)

        output = F.relu(output)
        output, hidden = self.gru(output, hidden)
        #print("output shape:",output.shape)
        #print("output[0]:",output[0])
        output = F.log_softmax(self.out(output[0]),dim=1)
        return output , hidden, attn_weights

    def initHidden(self):
        return torch.zeros(1, 1, self.hidden_size, device=device)

继续准备数据：

def indexesFromSentence(lang, sentence):
    return [lang.word2index[word] for word in sentence.split(' ')]


def tensorFromSentence(lang, sentence):
    indexes = indexesFromSentence(lang, sentence)
    indexes.append(EOS_token)
    return torch.tensor(indexes, dtype=torch.long, device=device).view(-1, 1)


def tensorsFromPair(pair):
    input_tensor = tensorFromSentence(input_lang, pair[0])
    target_tensor = tensorFromSentence(output_lang, pair[1])
    return (input_tensor, target_tensor)

训练模型

• 解码器的第一个输入是SOS符，并且把编码器最后的隐层状态作为解码器的第一隐层状态。
• “Teacher forcing”指用真实样本数据作为下一步的输入，而不是解码器猜测的数据作为下一步输入。

teacher_forcing_ratio = 0.5


def train(input_tensor, output_tensor, encoder, decoder, encoder_optimizer,
          decoder_optimizer, criterion, max_length=MAX_LENGTH):
    # 这的隐层大小封装在encoder中，然后拿过来在train的时候初始化隐层的大小
    encoder_hidden = encoder.initHidden()
    encoder_optimizer.zero_grad()
    decoder_optimizer.zero_grad()
    # 第一维度的大小即输入长度
    input_length = input_tensor.size(0)
    output_length = output_tensor.size(0)

    encoder_outputs = torch.zeros(max_length, encoder.hidden_size,device=device)

    loss = 0

    for ei in range(input_length):
        encoder_output, encoder_hidden = encoder(input_tensor[ei],encoder_hidden)
        # [0,0]选取最大数组的第一个元素组里的第一个
        encoder_outputs[ei] = encoder_output[0 , 0]
        if ei == 0 :
            print("encoder_output[0, 0] shape: ",encoder_outputs[ei].shape)

        decoder_input = torch.tensor([[SOS_token]], device=device)
        decoder_hidden = encoder_output
        # niubi 
        use_teacher_forcing = True if random.random() < teacher_forcing_ratio else False

        if use_teacher_forcing:
            # Teacher forcing: Feed the target as the next input
            for di in range(output_length):
                decoder_ouput,decoder_hidden,decoder_attention = decoder( decoder_input, decoder_hidden, encoder_outputs)
                loss = loss + criterion(decoder_ouput, output_tensor[di])
                decoder_input = output_tensor[di] # Teacher forcing
        else:
            for di in range(output_length):
                decoder_output,decoder_hidden,decoder_attention=decoder(decoder_input, decoder_hidden, encoder_outputs)
                topv ,topi = decoder_output.topk(1)
                decoder_input=  topi.squeeze().detach() # # detach from history as input


                loss = loss + criterion(decoder_output, output_tensor[di])
                if decoder_input.item() == EOS_token:
                    break  
    loss.backward()
    encoder_optimizer.step()
    decoder_optimizer.step()
    return loss.item() / target_length

好了，模型准备结束：

import time
import math

def asMinutes(s):
    m = math.floors(s / 60)
    s -= m * 60
    return "%s(- %s)" % (asMinutes(s), asMinutes(rs))


def timeSince(since, percent):
    now = time.time()
    s = now - since
    es = s / (percent)
    rs = es - s
    return '%s (- %s)' % (asMinutes(s), asMinutes(rs))

训练过程：

def trainIters(encoder, decoder, n_iters, print_every=1000, plot_every=100, learning_rate=0.01):
    start = time.time()
    plot_losses = []
    print_loss_total = 0  # Reset every print_every
    plot_loss_total = 0  # Reset every plot_every

    encoder_optimizer = optim.SGD(encoder.parameters(), lr=learning_rate)
    decoder_optimizer = optim.SGD(decoder.parameters(), lr=learning_rate)
    training_pairs = [tensorsFromPair(random.choice(pairs))
                      for i in range(n_iters)]
    criterion = nn.NLLLoss()

    for iter in range(1, n_iters + 1):
        training_pair = training_pairs[iter - 1]
        input_tensor = training_pair[0]
        target_tensor = training_pair[1]

        loss = train(input_tensor, target_tensor, encoder,
                     decoder, encoder_optimizer, decoder_optimizer, criterion)
        print_loss_total = loss + print_loss_total
        plot_loss_total = loss + plot_loss_total 

        if iter % print_every == 0:
            print_loss_avg = print_loss_total / print_every
            print_loss_total = 0
            print('%s (%d %d%%) %.4f' % (timeSince(start, iter / n_iters),
                                         iter, iter / n_iters * 100, print_loss_avg))

        if iter % plot_every == 0:
            plot_loss_avg = plot_loss_total / plot_every
            plot_losses.append(plot_loss_avg)
            plot_loss_total = 0

    showPlot(plot_losses)

画图的这段：

import matplotlib.pyplot as plt
plt.switch_backend('agg')
import matplotlib.ticker as ticker
import numpy as np


def showPlot(points):
    plt.figure()
    fig, ax = plt.subplots()
    # this locator puts ticks at regular intervals
    loc = ticker.MultipleLocator(base=0.2)
    ax.yaxis.set_major_locator(loc)
    plt.plot(points)

验证的代码：

def evaluate(encoder, decoder, sentence, max_length=MAX_LENGTH):
    with torch.no_grad():
        input_tensor = tensorFromSentence(input_lang, sentence)
        input_length = input_tensor.size()[0]
        encoder_hidden = encoder.initHidden()

        encoder_outputs = torch.zeros(max_length, encoder.hidden_size, device=device)

        for ei in range(input_length):
            encoder_output, encoder_hidden = encoder(input_tensor[ei],
                                                     encoder_hidden)
            encoder_outputs[ei] += encoder_output[0, 0]

        decoder_input = torch.tensor([[SOS_token]], device=device)  # SOS

        decoder_hidden = encoder_hidden

        decoded_words = []
        decoder_attentions = torch.zeros(max_length, max_length)

        for di in range(max_length):
            decoder_output, decoder_hidden, decoder_attention = decoder(
                decoder_input, decoder_hidden, encoder_outputs)
            decoder_attentions[di] = decoder_attention.data
            topv, topi = decoder_output.data.topk(1)
            if topi.item() == EOS_token:
                decoded_words.append('<EOS>')
                break
            else:
                decoded_words.append(output_lang.index2word[topi.item()])

            decoder_input = topi.squeeze().detach()

        return decoded_words, decoder_attentions[:di + 1]

def evaluateRandomly(encoder, decoder, n=10):
    for i in range(n):
        pair = random.choice(pairs)
        print('>', pair[0])
        print('=', pair[1])
        output_words, attentions = evaluate(encoder, decoder, pair[0])
        output_sentence = ' '.join(output_words)
        print('<', output_sentence)
        print('')

最后一步：

hidden_size = 256
encoder1 = EncoderRNN(input_lang.n_words, hidden_size).to(device)
attn_decoder1 = AttnDecoderRNN(hidden_size, output_lang.n_words, dropout_p=0.1).to(device)

trainIters(encoder1, attn_decoder1, 75000, print_every=5000)