• 🍨 本文为🔗365天深度学习训练营 中的学习记录博客
• 🍖 原作者：K同学啊# 前言

前言

前面学习了相关自然语言编码，这周进行相关实战

导入依赖库和设置设备

import torch
import torch.nn as nn
import torchvision
from torchvision import transforms, datasets
import os, PIL, pathlib, warningswarnings.filterwarnings("ignore")  # 忽略警告
# win10
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

这段代码导入了必要的库并设置了设备（GPU或CPU）。

数据预处理和词汇表构建

from torchtext.data.utils import get_tokenizer
from torchtext.vocab import build_vocab_from_iterator
from torchtext.datasets import AG_NEWStrain_iter = AG_NEWS(split='train')
tokenizer = get_tokenizer('basic_english')  # 返回分词器函数def yield_tokens(data_iter):for _, text in data_iter:yield tokenizer(text)vocab = build_vocab_from_iterator(yield_tokens(train_iter), specials=["<unk>"])
vocab.set_default_index(vocab["<unk>"])  # 设置默认索引，如果找不到单词，则会选择默认索引

这里使用torchtext库加载AG_NEWS数据集，定义了一个分词器并构建了词汇表。

数据处理管道

text_pipeline = lambda x: vocab(tokenizer(x))
label_pipeline = lambda x: int(x) - 1
text_pipeline('here is the an example')

定义了两个数据处理管道：text_pipeline用于将文本转化为词汇表中的索引序列，label_pipeline用于将标签转化为整数索引。

定义数据加载器

from torch.utils.data import DataLoaderdef collate_batch(batch):label_list, text_list, offsets = [], [], [0]for (_label, _text) in batch:label_list.append(label_pipeline(_label))processed_text = torch.tensor(text_pipeline(_text), dtype=torch.int64)text_list.append(processed_text)offsets.append(processed_text.size(0))label_list = torch.tensor(label_list, dtype=torch.int64)text_list = torch.cat(text_list)offsets = torch.tensor(offsets[:-1]).cumsum(dim=0)  # 返回维度dim中输入元素的累计和return label_list.to(device), text_list.to(device), offsets.to(device)dataloader = DataLoader(train_iter, batch_size=8, shuffle=False, collate_fn=collate_batch)

定义了一个collate_batch函数用于将一个批次的数据整合在一起，并创建了一个数据加载器。

定义模型

from torch import nnclass TextClassificationModel(nn.Module):def __init__(self, vocab_size, embed_dim, num_class):super(TextClassificationModel, self).__init__()self.embedding = nn.EmbeddingBag(vocab_size, embed_dim, sparse=False)self.fc = nn.Linear(embed_dim, num_class)self.init_weights()def init_weights(self):initrange = 0.5self.embedding.weight.data.uniform_(-initrange, initrange)self.fc.weight.data.uniform_(-initrange, initrange)self.fc.bias.data.zero_()def forward(self, text, offsets):embedded = self.embedding(text, offsets)return self.fc(embedded)num_class = len(set([label for (label, text) in train_iter]))
vocab_size = len(vocab)
em_size = 64
model = TextClassificationModel(vocab_size, em_size, num_class).to(device)

定义了一个文本分类模型TextClassificationModel，包括初始化函数、权重初始化和前向传播函数。模型由一个嵌入层和一个线性层组成。

训练和评估函数

import timedef train(dataloader):model.train()  # 切换为训练模式total_acc, train_loss, total_count = 0, 0, 0log_interval = 500start_time = time.time()for idx, (label, text, offsets) in enumerate(dataloader):predicted_label = model(text, offsets)optimizer.zero_grad()  # grad属性归零loss = criterion(predicted_label, label)  # 计算网络输出和真实值之间的差距，label为真实值loss.backward()  # 反向传播optimizer.step()  # 每一步自动更新total_acc += (predicted_label.argmax(1) == label).sum().item()train_loss += loss.item()total_count += label.size(0)if idx % log_interval == 0 and idx > 0:elapsed = time.time() - start_timeprint('| epoch {:1d} | {:4d}/{:4d} batches ''| train_acc {:4.3f} train_loss {:4.5f}'.format(epoch, idx, len(dataloader),total_acc / total_count, train_loss / total_count))total_acc, train_loss, total_count = 0, 0, 0start_time = time.time()def evaluate(dataloader):model.eval()  # 切换为测试模式total_acc, train_loss, total_count = 0, 0, 0with torch.no_grad():for idx, (label, text, offsets) in enumerate(dataloader):predicted_label = model(text, offsets)loss = criterion(predicted_label, label)  # 计算loss值total_acc += (predicted_label.argmax(1) == label).sum().item()train_loss += loss.item()total_count += label.size(0)return total_acc / total_count, train_loss / total_count

定义了训练和评估函数，用于训练模型和评估模型性能。

数据集分割和数据加载器创建

from torch.utils.data.dataset import random_split
from torchtext.data.functional import to_map_style_datasetEPOCHS = 10  # epoch
LR = 5  # 学习率
BATCH_SIZE = 64  # batch size for trainingcriterion = torch.nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.parameters(), lr=LR)
scheduler = torch.optim.lr_scheduler.StepLR(optimizer, 1.0, gamma=0.1)
total_accu = Nonetrain_iter, test_iter = AG_NEWS()  # 加载数据
train_dataset = to_map_style_dataset(train_iter)
test_dataset = to_map_style_dataset(test_iter)
num_train = int(len(train_dataset) * 0.95)split_train_, split_valid_ = random_split(train_dataset,[num_train, len(train_dataset) - num_train])train_dataloader = DataLoader(split_train_, batch_size=BATCH_SIZE,shuffle=True, collate_fn=collate_batch)
valid_dataloader = DataLoader(split_valid_, batch_size=BATCH_SIZE,shuffle=True, collate_fn=collate_batch)
test_dataloader = DataLoader(test_dataset, batch_size=BATCH_SIZE,shuffle=True, collate_fn=collate_batch)

加载数据集并将其转换为适用于随机访问的数据集，分割训练集和验证集，并创建相应的数据加载器。

训练和验证模型

for epoch in range(1, EPOCHS + 1):epoch_start_time = time.time()train(train_dataloader)val_acc, val_loss = evaluate(valid_dataloader)if total_accu is not None and total_accu > val_acc:scheduler.step()else:total_accu = val_accprint('-' * 69)print('| epoch {:1d} | time: {:4.2f}s | ''valid_acc {:4.3f} valid_loss {:4.3f}'.format(epoch,time.time() - epoch_start_time,val_acc, val_loss))print('-' * 69)

进行训练和验证，在每个epoch结束时打印验证准确率和损失，并根据验证结果调整学习率。

测试模型

print('Checking the results of test dataset.')
test_acc, test_loss = evaluate(test_dataloader)
print('test accuracy {:8.3f}'.format(test_acc))

在测试集上评估模型性能并打印测试准确率。

结果

总结

这个案例实现了一个完整的文本分类流程，从数据预处理、模型定义到训练和评估。使用torchtext加载数据，并利用PyTorch构建和训练深度学习模型，实现了对AG_NEWS数据集的文本分类任务，达到了90.1%的精度。