本文将通过训练一个文本分类模型来实现情感分析任务。其中包括torchtext的基本用法——BucketIterator和torch.nn的一些基本模型——Conv2d。

在情感分类任务中，我们的数据包括文本字符和两种情感，“pos”和“neg”。Field的参数决定了数据会被怎么处理，我们使用TEXT field来定义如何处理电影评论，用LABEL field来处理两个情感类别。
其中，TEXT field参数有tokenize=‘spacy’，这表示我们会用spaCy.tokenizer来tokenize英文句子。默认分词方法是空格。
步骤总结：

1. 构建TEXT、LABEL两个Field，设置数据相关参数；
2. 读取我们的train、validation、test数据集；
3. 使用build_voca构建词汇表，处理unk，pad问题，也可以使用vectors预处理词向量；
4. 分别构建指定batch_size大小的iterator。

import torch
import random
from torchtext import data

SEED = 1234

torch.manual_seed(SEED)
torch.cuda.manual_seed(SEED)
torch.backends.cudnn.deterministic = True
#两个Field初始化，定义vocab和tokenization类型等数据预处理工作
TEXT = data.Field(tokenize='spacy')
LABEL = data.LabelField(dtype=torch.float)
#下载IMDB数据集，包括5W条标注电影评论数据。
from torchtext import datasets
train_data, test_data = datasets.IMDB.splits(TEXT, LABEL)
#【查看每个数据split有多少条数据】
print(f'Number of training examples: {len(train_data)}')
print(f'Number of testing examples: {len(test_data)}')
#【查看一个example】
print(vars(train_data.examples[0]))
#split()数据分割，默认训练:测试-7:3，使用split_ratio参数调整比例
train_data, valid_data = train_data.split(random_state=random.seed(SEED))
#【查看每部分数据有多少条】
print(f'Number of training examples: {len(train_data)}')
print(f'Number of validation examples: {len(valid_data)}')
print(f'Number of testing examples: {len(test_data)}')
#使用glove预训练的词向量，创建vocabulary。（将每个单词映射到一个数字）
# TEXT.build_vocab(train_data, max_size=25000)
# LABEL.build_vocab(train_data)
TEXT.build_vocab(train_data, max_size=25000, vectors="glove.6B.100d", unk_init=torch.Tensor.normal_)
LABEL.build_vocab(train_data)
#【查看vocabulary的token数、最常见词、itos（int to string）、stoi（string to int）、LABEL信息】
print(f"Unique tokens in TEXT vocabulary: {len(TEXT.vocab)}")
print(f"Unique tokens in LABEL vocabulary: {len(LABEL.vocab)}")
print(TEXT.vocab.freqs.most_common(20))
print(TEXT.vocab.itos[:10])
print(LABEL.vocab.stoi)
#使用BucketIterator构建iterator来生成batch（note:<pad>应该在输入数据中消除。
BATCH_SIZE = 64
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
train_iterator, valid_iterator, test_iterator = data.BucketIterator.splits(
    (train_data, valid_data, test_data), 
    batch_size=BATCH_SIZE,
    device=device)

参考：

从简单到复杂模型，依次构建Word Averaging模型、RNN/LSTM模型、CNN模型。

• 说下都有啥层，上个图就算介绍完了。这个模型比较简单，但是效果很好，鲁棒性很高。
• 输入的单词->Embedding层->average层->Linear层进行分类。
  
• 关于实现：avg_pool2d做average pooling。（sent_len, embedding_dim) -> (1,embedding_dim)
• note:avg_pool2d的kernel_size是（embedded.shape[1],1），所以句子长度这个维度会被压扁。

import torch.nn as nn
import torch.nn.functional as F
#模型构建
class WordAVGModel(nn.Module):
    def __init__(self, vocab_size, embedding_dim, output_dim, pad_idx):
        super().__init__()
        self.embedding = nn.Embedding(vocab_size, embedding_dim, padding_idx=pad_idx)
        self.fc = nn.Linear(embedding_dim, output_dim)
        
    def forward(self, text):
        embedded = self.embedding(text) # [sent len, batch size, emb dim]
        embedded = embedded.permute(1, 0, 2) # [batch size, sent len, emb dim]
        pooled = F.avg_pool2d(embedded, (embedded.shape[1], 1)).squeeze(1) # [batch size, embedding_dim]
        return self.fc(pooled)

RNN的作用就是一个encoder，相当于去带了avg_pool2d层。
p a s s pass pass
使用最后一个hidden stateh_T来表示整个句子
最后加一个线性层，预测句子情感。

class RNN(nn.Module):
    def __init__(self, vocab_size, embedding_dim, hidden_dim, output_dim, 
                 n_layers, bidirectional, dropout, pad_idx):
        super().__init__()
        self.embedding = nn.Embedding(vocab_size, embedding_dim, padding_idx=pad_idx)
        self.rnn = nn.LSTM(embedding_dim, hidden_dim, num_layers=n_layers, 
                           bidirectional=bidirectional, dropout=dropout)
        self.fc = nn.Linear(hidden_dim*2, output_dim)
        self.dropout = nn.Dropout(dropout)
        
    def forward(self, text):
        embedded = self.dropout(self.embedding(text)) #[sent len, batch size, emb dim]
        output, (hidden, cell) = self.rnn(embedded)
        #output = [sent len, batch size, hid dim * num directions]
        #hidden = [num layers * num directions, batch size, hid dim]
        #cell = [num layers * num directions, batch size, hid dim]
        
        #concat the final forward (hidden[-2,:,:]) and backward (hidden[-1,:,:]) hidden layers
        #and apply dropout
        hidden = self.dropout(torch.cat((hidden[-2,:,:], hidden[-1,:,:]), dim=1)) # [batch size, hid dim * num directions]
        return self.fc(hidden.squeeze(0))

class CNN(nn.Module):
    def __init__(self, vocab_size, embedding_dim, n_filters, 
                 filter_sizes, output_dim, dropout, pad_idx):
        super().__init__()
        
        self.embedding = nn.Embedding(vocab_size, embedding_dim, padding_idx=pad_idx)
        self.convs = nn.ModuleList([
                                    nn.Conv2d(in_channels = 1, out_channels = n_filters, 
                                              kernel_size = (fs, embedding_dim)) 
                                    for fs in filter_sizes
                                    ])
        self.fc = nn.Linear(len(filter_sizes) * n_filters, output_dim)
        self.dropout = nn.Dropout(dropout)
        
    def forward(self, text):
        text = text.permute(1, 0) # [batch size, sent len]
        embedded = self.embedding(text) # [batch size, sent len, emb dim]
        embedded = embedded.unsqueeze(1) # [batch size, 1, sent len, emb dim]
        conved = [F.relu(conv(embedded)).squeeze(3) for conv in self.convs]
            
        #conv_n = [batch size, n_filters, sent len - filter_sizes[n]]
        
        pooled = [F.max_pool1d(conv, conv.shape[2]).squeeze(2) for conv in conved]
        
        #pooled_n = [batch size, n_filters]
        
        cat = self.dropout(torch.cat(pooled, dim=1))

        #cat = [batch size, n_filters * len(filter_sizes)]
            
        return self.fc(cat)

RNN、CNN不同的地方也就是参数，训练方式等，所以不再赘述。

#返回所有requires_grad的参数
def count_parameters(model):
    return sum(p.numel() for p in model.parameters() if p.requires_grad)
#模型需要使用的参数
INPUT_DIM = len(TEXT.vocab)
EMBEDDING_DIM = 100 #与glove词向量维度相同
OUTPUT_DIM = 1
PAD_IDX = TEXT.vocab.stoi[TEXT.pad_token]
#实例化一个模型对象
model = WordAVGModel(INPUT_DIM, EMBEDDING_DIM, OUTPUT_DIM, PAD_IDX)
#给模型设置预训练词向量。
pretrained_embeddings = TEXT.vocab.vectors
model.embedding.weight.data.copy_(pretrained_embeddings)
#初始化UNK、PAD的token
UNK_IDX = TEXT.vocab.stoi[TEXT.unk_token]
model.embedding.weight.data[UNK_IDX] = torch.zeros(EMBEDDING_DIM)
model.embedding.weight.data[PAD_IDX] = torch.zeros(EMBEDDING_DIM)
#【查看参数数量】
print(f'The model has {count_parameters(model):,} trainable parameters')
#训练模型，还是那几步，optimizer，计算损失，反向传播，zero_grad()
import torch.optim as optim
#定义optimizer和loss计算方式
optimizer = optim.Adam(model.parameters())
criterion = nn.BCEWithLogitsLoss()
model = model.to(device)
criterion = criterion.to(device)
#计算准确率
def binary_accuracy(preds, y):
    """
    Returns accuracy per batch, i.e. if you get 8/10 right, this returns 0.8, NOT 8
    """
    #round predictions to the closest integer
    rounded_preds = torch.round(torch.sigmoid(preds))
    correct = (rounded_preds == y).float() #convert into float for division 
    acc = correct.sum()/len(correct)
    return acc
#训练模型
def train(model, iterator, optimizer, criterion):
    
    epoch_loss = 0
    epoch_acc = 0
    model.train()
    
    for batch in iterator:
        optimizer.zero_grad()
        predictions = model(batch.text).squeeze(1)
        loss = criterion(predictions, batch.label)
        acc = binary_accuracy(predictions, batch.label)
        loss.backward()
        optimizer.step()
        
        epoch_loss += loss.item()
        epoch_acc += acc.item()
        
    return epoch_loss / len(iterator), epoch_acc / len(iterator)
#评价模型
def evaluate(model, iterator, criterion):
    
    epoch_loss = 0
    epoch_acc = 0
    model.eval()
    
    with torch.no_grad():
        for batch in iterator:
            predictions = model(batch.text).squeeze(1)
            loss = criterion(predictions, batch.label)
            acc = binary_accuracy(predictions, batch.label)
            epoch_loss += loss.item()
            epoch_acc += acc.item()
        
    return epoch_loss / len(iterator), epoch_acc / len(iterator)
#计算epoch_time
import time

def epoch_time(start_time, end_time):
    elapsed_time = end_time - start_time
    elapsed_mins = int(elapsed_time / 60)
    elapsed_secs = int(elapsed_time - (elapsed_mins * 60))
    return elapsed_mins, elapsed_secs
N_EPOCHS = 10
best_valid_loss = float('inf')
#正式开始训练
for epoch in range(N_EPOCHS):
    start_time = time.time()
    train_loss, train_acc = train(model, train_iterator, optimizer, criterion)
    valid_loss, valid_acc = evaluate(model, valid_iterator, criterion)
    end_time = time.time()
    epoch_mins, epoch_secs = epoch_time(start_time, end_time)
    if valid_loss < best_valid_loss:
        best_valid_loss = valid_loss
        torch.save(model.state_dict(), 'wordavg-model.pt')
    print(f'Epoch: {epoch+1:02} | Epoch Time: {epoch_mins}m {epoch_secs}s')
    print(f'tTrain Loss: {train_loss:.3f} | Train Acc: {train_acc*100:.2f}%')
    print(f't Val. Loss: {valid_loss:.3f} |  Val. Acc: {valid_acc*100:.2f}%')
#生成预测-输入句子判断情感正负
import spacy
nlp = spacy.load('en')

def predict_sentiment(sentence):
    tokenized = [tok.text for tok in nlp.tokenizer(sentence)]
    indexed = [TEXT.vocab.stoi[t] for t in tokenized]
    tensor = torch.LongTensor(indexed).to(device)
    tensor = tensor.unsqueeze(1)
    prediction = torch.sigmoid(model(tensor))
    return prediction.item()
#进行预测，我尼玛还挺准
predict_sentiment("This film is terrible")
predict_sentiment("This film is great")