<数据分析> 中文文本分类FastText-pytorch

摘要：中文文本分类FastText-pytorch
代码：https://github.com/649453932/Chinese-Text-Classification-Pytorch

引言

因为学习需要用到文本分析相关模型，就根据一个github库中的代码尝试实现文本分类的过程，在注释中添加学习笔记，若有错误请指正。
代码地址：https://github.com/649453932/Chinese-Text-Classification-Pytorch

项目结构

其中runFastText为主函数，train_eval是训练函数，utils_fastTextTest是数据处理函数，名字与原github中有差别，代码大部分都相同。

utils_fastTextTest.py

# coding: UTF-8
import os
import torch
import numpy as np
import pickle as pkl
from tqdm import tqdm
import time
from datetime import timedelta

MAX_VOCAB_SIZE = 10000 # 词表长度限制
UNK, PAD = '<UNK>', '<PAD>' # 未知字，padding符号

def build_vocab(file_path, tokenizer, max_size, min_freq):
    """
      构建一个词表：
      首先对数据集中的每一行句子按字/空格进行分割，然后统计所有元素的出现频率
      接下来按照频率从高到低的顺序对所有频率大于min_freq的元素进行排序，取前max_size个元素
      最后按照频率降序构建字典vocab_dic：{元素:序号}，vocab_dic的最后两个元素是'<UNK>'和'<PAD>'
    """
    vocab_dic = {}
    with open(file_path, 'r', encoding='UTF-8') as f:
        # tqdm是一个python的进度条工具包，在终端运行时显示程序循环的进度
        for line in tqdm(f):# 处理每一行
            lin = line.strip() # 移除头尾空格或换行符
            if not lin: # 跳过空行
                continue
            content = lin.split('\t')[0] # 句子和标签通过tab分割，前面的是句子内容，后面的是标签(lin.split('\t')[1])
            for word in tokenizer(content): # 按空格分割或者按字分割,tokenizer为可选择参数
                vocab_dic[word] = vocab_dic.get(word, 0) + 1 # 统计词频或字频，即为每个词或字在训练集中出现的次数
        # 遍历词典，筛选出词频大于min_freq=1的词，然后按照词频从高到低排序，取前max_size=10000个词，组成新的列表vocab_list，vocab_list中的元素为元组(word, freq)
        """
        sorted函数用法：sorted(iterable, cmp=None, key=None, reverse=False)
        iterable -- 可迭代对象。
        cmp -- 比较的函数，这个具有两个参数，参数的值都是从可迭代对象中取出，此函数必须遵守的规则为，大于则返回1，小于则返回-1，等于则返回0。
        key -- 主要是用来进行比较的元素，只有一个参数，具体的函数的参数就是取自于可迭代对象中，指定可迭代对象中的一个元素来进行排序。
        reverse -- 排序规则，reverse = True 降序 ， reverse = False 升序（默认）。
        """
        vocab_list = sorted([_ for _ in vocab_dic.items() if _[1] >= min_freq], key=lambda x: x[1], reverse=True)[:max_size]
        vocab_dic = {word_count[0]: idx for idx, word_count in enumerate(vocab_list)}
        # 在vocab_dic的最后增加两个元素：{'<UNK>':len(vocab_dic)}和{'<PAD>':len(vocab_dic)+1}
        vocab_dic.update({UNK: len(vocab_dic), PAD: len(vocab_dic) + 1})
    return vocab_dic

def build_dataset(config, ues_word):
    """
    加载数据集：
    对数据集中的每一行，先分离内容和标签
    然后对句子内容，按指定的方式进行分割（依照空格或字符），接着根据pad_size进行补足或截断
    接着把分割后的元素，通过词表转化成一串序号words_line
    最后把所有句子处理后的结果组成一个大列表，列表中的元素为：[(words_line, int(label), seq_len),...]
    """
    if ues_word:
        tokenizer = lambda x: x.split(' ')  # 以空格隔开，word-level
    else:
        tokenizer = lambda x: [y for y in x]  # char-level
    # if os.path.exists(config.vocab_path):
    #     vocab = pkl.load(open(config.vocab_path, 'rb'))
    # else:
    # 此处使用训练集自己构建词表
    vocab = build_vocab(config.train_path, tokenizer=tokenizer, max_size=MAX_VOCAB_SIZE, min_freq=1)
    # 构建完了之后保存为pickle
    pkl.dump(vocab, open(config.vocab_path, 'wb'))
    print(f"Vocab size: {len(vocab)}")

    def biGramHash(sequence, t, buckets):
        t1 = sequence[t - 1] if t - 1 >= 0 else 0
        return (t1 * 14918087) % buckets

    def triGramHash(sequence, t, buckets):
        t1 = sequence[t - 1] if t - 1 >= 0 else 0
        t2 = sequence[t - 2] if t - 2 >= 0 else 0
        return (t2 * 14918087 * 18408749 + t1 * 14918087) % buckets

    def load_dataset(path, pad_size=32):
        contents = []
        with open(path, 'r', encoding='UTF-8') as f:
            for line in tqdm(f): # 打开数据文件，按行读取
                lin = line.strip() # 移除头尾空格或换行符
                if not lin: # 跳过空行
                    continue
                content, label = lin.split('\t') # 句子和标签通过tab分割，前面的是句子内容，后面的是标签
                words_line = []  # words_line是句子通过词表转化后得到的数字表示
                token = tokenizer(content) # 按空格或字符来分割句子
                seq_len = len(token)  # 得到分割后的元素数量
                if pad_size: # 如果有指定填充长度
                    if len(token) < pad_size:
                        token.extend([PAD] * (pad_size - len(token))) #padding填充
                    else:
                        token = token[:pad_size] # 直接截断至填充长度
                        seq_len = pad_size  # 更新元素数量
                # word to id
                for word in token:
                    # 拿到该元素在词表中的序号，然后将这个序号添加到words_line中。如果该元素不在词表中，则填入'<UNK>'（未知字）的序号
                    words_line.append(vocab.get(word, vocab.get(UNK)))

                # fasttext ngram
                buckets = config.n_gram_vocab
                bigram = []
                trigram = []
                # ------ngram------
                for i in range(pad_size):
                    bigram.append(biGramHash(words_line, i, buckets))
                    trigram.append(triGramHash(words_line, i, buckets))
                # -----------------
                # 在contents中存入一个元组，元组的内容为（words_line，数字标签，元素数量，bigram，trigram）
                contents.append((words_line, int(label), seq_len, bigram, trigram))
        return contents  # [([...], 0), ([...], 1), ...]
    train = load_dataset(config.train_path, config.pad_size)
    dev = load_dataset(config.dev_path, config.pad_size)
    test = load_dataset(config.test_path, config.pad_size)
    return vocab, train, dev, test

class DatasetIterater(object):
    """
      根据数据集产生batch
      这里需要注意的是，在_to_tensor()中，代码把batch中的数据处理成了`(x, seq_len), y`的形式
      其中x是words_line，seq_len是pad前的长度(超过pad_size的设为pad_size)，y是数据标签
    """
    # 这里的batches就是经过build_dataset()中的load_dataset()处理后得到的contents：(words_line, int(label), seq_len, bigram, trigram)
    def __init__(self, batches, batch_size, device):
        self.batch_size = batch_size # batch的容量（一次进多少个句子）
        self.batches = batches  # 数据集
        self.n_batches = len(batches) // batch_size # 数据集大小整除batch容量
        self.residue = False  # 记录batch数量是否为整数，false代表可以，true代表不可以，residuere是‘剩余物，残渣'的意思
        if len(batches) % self.n_batches != 0:
            self.residue = True
        self.index = 0 # 迭代用的索引
        self.device = device

    def _to_tensor(self, datas):
        # xx = [xxx[2] for xxx in datas]
        # indexx = np.argsort(xx)[::-1]
        # datas = np.array(datas)[indexx]
        x = torch.LongTensor([_[0] for _ in datas]).to(self.device) # 句子words_line
        y = torch.LongTensor([_[1] for _ in datas]).to(self.device) # 标签
        bigram = torch.LongTensor([_[3] for _ in datas]).to(self.device)
        trigram = torch.LongTensor([_[4] for _ in datas]).to(self.device)

        # pad前的长度(超过pad_size的设为pad_size，未超过的为原seq_size不变)
        seq_len = torch.LongTensor([_[2] for _ in datas]).to(self.device)
        return (x, seq_len, bigram, trigram), y

    def __next__(self):
        if self.residue and self.index == self.n_batches: # 如果batch外还剩下一点句子，并且迭代到了最后一个batch
            batches = self.batches[self.index * self.batch_size: len(self.batches)] # 直接拿出剩下的所有数据
            self.index += 1
            batches = self._to_tensor(batches)
            return batches

        elif self.index >= self.n_batches:
            self.index = 0
            raise StopIteration
        else: # 迭代器的入口，刚开始self.index是0，肯定小于self.n_batches
            batches = self.batches[self.index * self.batch_size: (self.index + 1) * self.batch_size] # 正常取一个batch的数据
            self.index += 1
            batches = self._to_tensor(batches) # 转化为tensor
            return batches

    def __iter__(self):
        return self

    def __len__(self):
        if self.residue:
            return self.n_batches + 1
        else:
            return self.n_batches

def build_iterator(dataset, config):  # 这里的dataset是经过build_dataset()处理后得到的数据（vocab, train, dev, test）
    iter = DatasetIterater(dataset, config.batch_size, config.device)
    return iter

def get_time_dif(start_time):
    """获取已使用时间"""
    end_time = time.time()
    time_dif = end_time - start_time
    return timedelta(seconds=int(round(time_dif)))

if __name__ == "__main__":
    '''提取预训练词向量'''
    vocab_dir = "./THUCNews/data/vocab.pkl"
    pretrain_dir = "./THUCNews/data/sgns.sogou.char"
    emb_dim = 300
    filename_trimmed_dir = "./THUCNews/data/vocab.embedding.sougou"
    word_to_id = pkl.load(open(vocab_dir, 'rb'))
    embeddings = np.random.rand(len(word_to_id), emb_dim)
    f = open(pretrain_dir, "r", encoding='UTF-8')
    for i, line in enumerate(f.readlines()):
        # if i == 0:  # 若第一行是标题，则跳过
        #     continue
        lin = line.strip().split(" ")
        if lin[0] in word_to_id:
            idx = word_to_id[lin[0]]
            emb = [float(x) for x in lin[1:301]]
            embeddings[idx] = np.asarray(emb, dtype='float32')
    f.close()
    np.savez_compressed(filename_trimmed_dir, embeddings=embeddings)

train_eval.py

# coding: UTF-8
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from sklearn import metrics
import time
from utils import get_time_dif
from tensorboardX import SummaryWriter


# 权重初始化，默认xavier（如果不初始化，则默认的随机权重会特别大，对模型训练造成影响）
def init_network(model, method='xavier', exclude='embedding', seed=123):
    for name, w in model.named_parameters(): # 迭代网络中所有可训练的参数
        if exclude not in name:  # 排除名字中包含指定关键词的参数（默认为'embedding'）
            if 'weight' in name: # 对权重进行初始化
                if method == 'xavier':
                    nn.init.xavier_normal_(w) # 调用不同的初始化方法
                elif method == 'kaiming':
                    nn.init.kaiming_normal_(w)
                else:
                    nn.init.normal_(w)
            elif 'bias' in name: # 对偏置进行初始化
                nn.init.constant_(w, 0)
            else: # 跳过除权重和偏置外的其他参数
                pass


def train(config, model, train_iter, dev_iter, test_iter):
    start_time = time.time()
    model.train() # model.train()将启用BatchNormalization和Dropout，相应的，model.eval()则不启用BatchNormalization和Dropout
    optimizer = torch.optim.Adam(model.parameters(), lr=config.learning_rate) # 指定优化方法

    # 学习率指数衰减，每次epoch：学习率 = gamma * 学习率
    # scheduler = torch.optim.lr_scheduler.ExponentialLR(optimizer, gamma=0.9)
    total_batch = 0  # 记录进行到多少batch
    dev_best_loss = float('inf')
    last_improve = 0  # 记录上次验证集loss下降的batch数
    flag = False  # 记录是否很久没有效果提升
    writer = SummaryWriter(log_dir=config.log_path + '/' + time.strftime('%m-%d_%H.%M', time.localtime()))
    for epoch in range(config.num_epochs):
        print('Epoch [{}/{}]'.format(epoch + 1, config.num_epochs))
        # scheduler.step() # 学习率衰减
        for i, (trains, labels) in enumerate(train_iter):
            outputs = model(trains)
            model.zero_grad()
            loss = F.cross_entropy(outputs, labels)
            loss.backward()
            optimizer.step()
            if total_batch % 100 == 0:
                # 每多少轮输出在训练集和验证集上的效果
                true = labels.data.cpu() # 从cpu tensor中取出标签数据
                predic = torch.max(outputs.data, 1)[1].cpu() # 返回每一行中最大值的列索引
                train_acc = metrics.accuracy_score(true, predic) # 计算这个batch的分类准确率
                dev_acc, dev_loss = evaluate(config, model, dev_iter) # 计算开发集上的准确率和训练误差
                if dev_loss < dev_best_loss:  # 使用开发集判断模型性能是否提升
                    dev_best_loss = dev_loss
                    torch.save(model.state_dict(), config.save_path)
                    improve = '*'
                    last_improve = total_batch
                else:
                    improve = ''
                time_dif = get_time_dif(start_time)
                msg = 'Iter: {0:>6},  Train Loss: {1:>5.2},  Train Acc: {2:>6.2%},  Val Loss: {3:>5.2},  Val Acc: {4:>6.2%},  Time: {5} {6}'
                print(msg.format(total_batch, loss.item(), train_acc, dev_loss, dev_acc, time_dif, improve))
                writer.add_scalar("loss/train", loss.item(), total_batch)
                writer.add_scalar("loss/dev", dev_loss, total_batch)
                writer.add_scalar("acc/train", train_acc, total_batch)
                writer.add_scalar("acc/dev", dev_acc, total_batch)
                model.train()
            total_batch += 1
            if total_batch - last_improve > config.require_improvement:
                # 验证集loss超过1000batch没下降，结束训练
                # 开发集loss超过一定数量的batch没下降，则结束训练
                print("No optimization for a long time, auto-stopping...")
                flag = True
                break
        if flag:
            break
    writer.close()
    test(config, model, test_iter)


def test(config, model, test_iter):
    # test
    model.load_state_dict(torch.load(config.save_path))
    model.eval()
    start_time = time.time()
    test_acc, test_loss, test_report, test_confusion = evaluate(config, model, test_iter, test=True)
    msg = 'Test Loss: {0:>5.2},  Test Acc: {1:>6.2%}'
    print(msg.format(test_loss, test_acc))
    print("Precision, Recall and F1-Score...")
    print(test_report)
    print("Confusion Matrix...")
    print(test_confusion)
    time_dif = get_time_dif(start_time)
    print("Time usage:", time_dif)


def evaluate(config, model, data_iter, test=False):
    model.eval() # 不启用BatchNormalization和Dropout
    loss_total = 0
    predict_all = np.array([], dtype=int)
    labels_all = np.array([], dtype=int)
    with torch.no_grad():  # 不追踪梯度
        for texts, labels in data_iter:  # 对数据集中的每一组数据
            outputs = model(texts)  # 使用模型进行预测
            loss = F.cross_entropy(outputs, labels) # 计算模型损失
            loss_total += loss # 累加模型损失
            labels = labels.data.cpu().numpy()
            predic = torch.max(outputs.data, 1)[1].cpu().numpy()
            labels_all = np.append(labels_all, labels) # 记录标签
            predict_all = np.append(predict_all, predic) # 记录预测结果

    acc = metrics.accuracy_score(labels_all, predict_all) # 计算分类误差
    if test:
        report = metrics.classification_report(labels_all, predict_all, target_names=config.class_list, digits=4)
        confusion = metrics.confusion_matrix(labels_all, predict_all)
        return acc, loss_total / len(data_iter), report, confusion
    return acc, loss_total / len(data_iter) # 返回分类误差和平均模型损失

FastText模型

# coding: UTF-8
import torch
import torch.nn as nn
import torch.nn.functional as F
import numpy as np

class Config(object):

    """配置参数"""
    def __init__(self, dataset, embedding):
        self.model_name = 'FastText'
        self.train_path = dataset + '/data/train.txt'                                # 训练集
        self.dev_path = dataset + '/data/dev.txt'                                    # 验证集
        self.test_path = dataset + '/data/test.txt'                                  # 测试集
        self.class_list = [x.strip() for x in open(
            dataset + '/data/class.txt', encoding='utf-8').readlines()]              # 类别名单
        self.vocab_path = dataset + '/data/vocab.pkl'                                # 词表
        self.save_path = dataset + '/saved_dict/' + self.model_name + '.ckpt'        # 模型训练结果
        self.log_path = dataset + '/log/' + self.model_name
        self.embedding_pretrained = torch.tensor(
            np.load(dataset + '/data/' + embedding)["embeddings"].astype('float32'))\
            if embedding != 'random' else None                                       # 预训练词向量
        self.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')   # 设备

        self.dropout = 0.5                                              # 随机失活
        self.require_improvement = 1000                                 # 若超过1000batch效果还没提升，则提前结束训练
        self.num_classes = len(self.class_list)                         # 类别数
        self.n_vocab = 0                                                # 词表大小，在运行时赋值
        self.num_epochs = 20                                            # epoch数
        self.batch_size = 128                                           # mini-batch大小
        self.pad_size = 32                                              # 每句话处理成的长度(短填长切)
        self.learning_rate = 1e-3                                       # 学习率
        self.embed = self.embedding_pretrained.size(1)\
            if self.embedding_pretrained is not None else 300           # 字向量维度
        self.hidden_size = 256                                          # 隐藏层大小
        self.n_gram_vocab = 250499                                      # ngram 词表大小


'''Bag of Tricks for Efficient Text Classification'''

class Model(nn.Module):
    def __init__(self, config):
        super(Model, self).__init__()
        if config.embedding_pretrained is not None:
            self.embedding = nn.Embedding.from_pretrained(config.embedding_pretrained, freeze=False)
        else:
            self.embedding = nn.Embedding(config.n_vocab, config.embed, padding_idx=config.n_vocab - 1)
        self.embedding_ngram2 = nn.Embedding(config.n_gram_vocab, config.embed)
        self.embedding_ngram3 = nn.Embedding(config.n_gram_vocab, config.embed)
        self.dropout = nn.Dropout(config.dropout)
        self.fc1 = nn.Linear(config.embed * 3, config.hidden_size)
        # self.dropout2 = nn.Dropout(config.dropout)
        self.fc2 = nn.Linear(config.hidden_size, config.num_classes)

    def forward(self, x):

        out_word = self.embedding(x[0])
        out_bigram = self.embedding_ngram2(x[2])
        out_trigram = self.embedding_ngram3(x[3])
        out = torch.cat((out_word, out_bigram, out_trigram), -1)

        out = out.mean(dim=1)
        out = self.dropout(out)
        out = self.fc1(out)
        out = F.relu(out)
        out = self.fc2(out)
        return out

runFastText.py主函数

# coding: UTF-8
import time
import torch
import numpy as np
from train_eval import train, init_network
from importlib import import_module
from utils_fastTextTest import build_dataset, build_iterator, get_time_dif
# import argparse

# parser = argparse.ArgumentParser(description='Chinese Text Classification')
# parser.add_argument('--model', type=str, required=True, help='choose a model: TextCNN, TextRNN, FastText, TextRCNN, TextRNN_Att, DPCNN, Transformer')
# parser.add_argument('--embedding', default='pre_trained', type=str, help='random or pre_trained')
# parser.add_argument('--word', default=False, type=bool, help='True for word, False for char')
# args = parser.parse_args()


if __name__ == '__main__':
    dataset = 'THUCNews'  # 数据集

    # 搜狗新闻:embedding_SougouNews.npz, 腾讯:embedding_Tencent.npz, 随机初始化:random
    embedding = 'embedding_SougouNews.npz'
    model_name = 'FastText'  # 'TextRCNN'  # TextCNN, TextRNN, FastText, TextRCNN, TextRNN_Att, DPCNN, Transformer

    x = import_module('models.' + model_name)
    config = x.Config(dataset, embedding) # 加载神经网络模型的参数
    np.random.seed(1)
    torch.manual_seed(1)
    torch.cuda.manual_seed_all(1)
    torch.backends.cudnn.deterministic = True  # 固定随机因子，保证每次结果一样

    start_time = time.time()
    print("Loading data...")
    vocab, train_data, dev_data, test_data = build_dataset(config, False)
    train_iter = build_iterator(train_data, config)
    dev_iter = build_iterator(dev_data, config)
    test_iter = build_iterator(test_data, config)
    time_dif = get_time_dif(start_time)
    print("Time usage:", time_dif)

    # train
    config.n_vocab = len(vocab)
    model = x.Model(config).to(config.device)
    if model_name != 'Transformer':
        init_network(model)
    print(model.parameters)
    train(config, model, train_iter, dev_iter, test_iter)

训练结果

因为无GPU，代码中添加了Ngram图，训练较慢。截图仅有两个epoch。
参考结果：不加N-Gram信息，就是词袋模型，准确率89.59%，加上2-gram和3-gram后准确率92.23%。

[video(video-TQauKbQ1-1591957225364)(type-edu_course)(url-https://edu.csdn.net/course/blogPlay?goods_id=17131&blog_creator=weixin_43433969&marketing_id=141)(image-https://img-bss.csdnimg.cn/20191229134635364.jpg)(title-PyTorch从入门到实战一次学会)]

参考文献

1.中文文本分类 pytorch实现
2.中文文本分类代码分析
3.FastText中词序的处理