900字范文 > Bert文本分类(基于keras-bert实现)

Bert文本分类(基于keras-bert实现)

时间：2023-08-31 05:12:49

一、Bert 预训练模型准备

二、Bert 模型文本分类

1、数据准备

2、代码实现

3、分类过程与结果

一、Bert 预训练模型准备

中文预训练模型下载当Bert遇上Keras：这可能是Bert最简单的打开姿势 keras-bert

不同模型的性能对比如下（可根据自己的数据选择合适的模型，模型越大需要训练的时间越长）

二、Bert 模型文本分类

1、数据准备

使用的仍是用户评论情感极性判别的数据

训练集：data_train.csv ，样本数为82025，情感极性标签（0：负面、1：中性、2：正面）

测试集：data_test.csv ，样本数为35157

评论数据主要包括：食品餐饮类，旅游住宿类，金融服务类，医疗服务类，物流快递类；部分数据如下：

2、代码实现

import pandas as pdimport codecs, gcimport numpy as npfrom sklearn.model_selection import KFoldfrom keras_bert import load_trained_model_from_checkpoint, Tokenizerfrom keras.metrics import top_k_categorical_accuracyfrom keras.layers import *from keras.callbacks import *from keras.models import Modelimport keras.backend as Kfrom keras.optimizers import Adamfrom keras.utils import to_categorical#读取训练集和测试集train_df=pd.read_csv('data/data_train.csv', sep='\t', names=['id', 'type', 'contents', 'labels']).astype(str)test_df=pd.read_csv('data/data_test.csv', sep='\t', names=['id', 'type', 'contents']).astype(str)maxlen = 100 #设置序列长度为120，要保证序列长度不超过512#预训练好的模型config_path = 'chinese_roberta_wwm_large_ext_L-24_H-1024_A-16/bert_config.json'checkpoint_path = 'chinese_roberta_wwm_large_ext_L-24_H-1024_A-16/bert_model.ckpt'dict_path = 'chinese_roberta_wwm_large_ext_L-24_H-1024_A-16/vocab.txt'#将词表中的词编号转换为字典token_dict = {}with codecs.open(dict_path, 'r', 'utf8') as reader:for line in reader:token = line.strip()token_dict[token] = len(token_dict)#重写tokenizer class OurTokenizer(Tokenizer):def _tokenize(self, text):R = []for c in text:if c in self._token_dict:R.append(c)elif self._is_space(c):R.append('[unused1]') # 用[unused1]来表示空格类字符else:R.append('[UNK]') # 不在列表的字符用[UNK]表示return Rtokenizer = OurTokenizer(token_dict)#让每条文本的长度相同，用0填充def seq_padding(X, padding=0):L = [len(x) for x in X]ML = max(L)return np.array([np.concatenate([x, [padding] * (ML - len(x))]) if len(x) < ML else x for x in X])#data_generator只是一种为了节约内存的数据方式class data_generator:def __init__(self, data, batch_size=32, shuffle=True):self.data = dataself.batch_size = batch_sizeself.shuffle = shuffleself.steps = len(self.data) // self.batch_sizeif len(self.data) % self.batch_size != 0:self.steps += 1def __len__(self):return self.stepsdef __iter__(self):while True:idxs = list(range(len(self.data)))if self.shuffle:np.random.shuffle(idxs)X1, X2, Y = [], [], []for i in idxs:d = self.data[i]text = d[0][:maxlen]x1, x2 = tokenizer.encode(first=text)y = d[1]X1.append(x1)X2.append(x2)Y.append([y])if len(X1) == self.batch_size or i == idxs[-1]:X1 = seq_padding(X1)X2 = seq_padding(X2)Y = seq_padding(Y)yield [X1, X2], Y[:, 0, :][X1, X2, Y] = [], [], []#计算top-k正确率,当预测值的前k个值中存在目标类别即认为预测正确 def acc_top2(y_true, y_pred):return top_k_categorical_accuracy(y_true, y_pred, k=2)#bert模型设置def build_bert(nclass):bert_model = load_trained_model_from_checkpoint(config_path, checkpoint_path, seq_len=None) #加载预训练模型for l in bert_model.layers:l.trainable = Truex1_in = Input(shape=(None,))x2_in = Input(shape=(None,))x = bert_model([x1_in, x2_in])x = Lambda(lambda x: x[:, 0])(x) # 取出[CLS]对应的向量用来做分类p = Dense(nclass, activation='softmax')(x)model = Model([x1_in, x2_in], p)pile(loss='categorical_crossentropy',optimizer=Adam(1e-5), #用足够小的学习率metrics=['accuracy', acc_top2])print(model.summary())return model#训练数据、测试数据和标签转化为模型输入格式DATA_LIST = []for data_row in train_df.iloc[:].itertuples():DATA_LIST.append((data_row.contents, to_categorical(data_row.labels, 3)))DATA_LIST = np.array(DATA_LIST)DATA_LIST_TEST = []for data_row in test_df.iloc[:].itertuples():DATA_LIST_TEST.append((data_row.contents, to_categorical(0, 3)))DATA_LIST_TEST = np.array(DATA_LIST_TEST)#交叉验证训练和测试模型def run_cv(nfold, data, data_labels, data_test):kf = KFold(n_splits=nfold, shuffle=True, random_state=520).split(data)train_model_pred = np.zeros((len(data), 3))test_model_pred = np.zeros((len(data_test), 3))for i, (train_fold, test_fold) in enumerate(kf):X_train, X_valid, = data[train_fold, :], data[test_fold, :]model = build_bert(3)early_stopping = EarlyStopping(monitor='val_acc', patience=3) #早停法，防止过拟合plateau = ReduceLROnPlateau(monitor="val_acc", verbose=1, mode='max', factor=0.5, patience=2) #当评价指标不在提升时，减少学习率checkpoint = ModelCheckpoint('./bert_dump/' + str(i) + '.hdf5', monitor='val_acc',verbose=2, save_best_only=True, mode='max', save_weights_only=True) #保存最好的模型train_D = data_generator(X_train, shuffle=True)valid_D = data_generator(X_valid, shuffle=True)test_D = data_generator(data_test, shuffle=False)#模型训练model.fit_generator(train_D.__iter__(),steps_per_epoch=len(train_D),epochs=5,validation_data=valid_D.__iter__(),validation_steps=len(valid_D),callbacks=[early_stopping, plateau, checkpoint],)# model.load_weights('./bert_dump/' + str(i) + '.hdf5')# return modeltrain_model_pred[test_fold, :] = model.predict_generator(valid_D.__iter__(), steps=len(valid_D), verbose=1)test_model_pred += model.predict_generator(test_D.__iter__(), steps=len(test_D), verbose=1)del modelgc.collect() #清理内存K.clear_session() #clear_session就是清除一个session# breakreturn train_model_pred, test_model_pred#n折交叉验证train_model_pred, test_model_pred = run_cv(2, DATA_LIST, None, DATA_LIST_TEST)test_pred = [np.argmax(x) for x in test_model_pred]#将测试集预测结果写入文件output=pd.DataFrame({'id':test_df.id,'sentiment':test_pred})output.to_csv('data/results.csv', index=None)