无脑bert使用，你会了吗？

在这里使用的是keras_bert库，可以直接pip安装

1.参数设置

#! -*- coding:utf-8 -*-
import re, os, json, codecs, gc
import numpy as np
import pandas as pd
from random import choice
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import KFold
from keras_bert import load_trained_model_from_checkpoint, Tokenizer
from keras.utils import to_categorical
from keras.layers import *
from keras.callbacks import *
from keras.models import Model
import keras.backend as K
from keras.metrics import top_k_categorical_accuracy
from keras.optimizers import Adam

maxlen = 512
batch_size = 2
drop_out_rate = 0.2
learning_rate = 1e-5
epochs = 5

train_data_path = 'C:/Users/jkx/Desktop/一些比赛/网民情绪分析/nCoV_100k_train.labled.csv'
test_data_path = 'C:/Users/jkx/Desktop/一些比赛/网民情绪分析/nCov_10k_test.csv'
config_path = 'D:/bert/chinese_wwm_ext_L-12_H-768_A-12/bert_config.json'
checkpoint_path = 'D:/bert/chinese_wwm_ext_L-12_H-768_A-12/bert_model.ckpt'
dict_path = 'D:/bert/chinese_wwm_ext_L-12_H-768_A-12/vocab.txt'

2.读取数据

train_data = pd.read_csv(train_data_path)
test_data = pd.read_csv(test_data_path)

3.处理数据

train_x = train_data['微博中文内容']
train_y = train_data['情感倾向'].apply(lambda x:x if x!=-1 else 2)

Train = []
Test = []
for i,j in zip(train_x,train_y):
    Train.append((i, to_categorical(j, 3)))
    
for i in test_data['微博中文内容']:
    Test.append((i, to_categorical(0, 3)))

4.定义tokenizer

token_dict = {}
with codecs.open(dict_path, 'r', 'utf8') as reader:
    for line in reader:
        token = line.strip()
        token_dict[token] = len(token_dict)

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]') # space类用未经训练的[unused1]表示
            else:
                R.append('[UNK]') # 剩余的字符是[UNK]
        return R

tokenizer = OurTokenizer(token_dict)

5.实现填充函数与获取batch函数

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
    ])


class data_generator:
    def __init__(self, data, batch_size=batch_size, shuffle=True):
        self.data = data
        self.batch_size = batch_size
        self.shuffle = shuffle
        self.steps = len(self.data) // self.batch_size
        if len(self.data) % self.batch_size != 0:
            self.steps += 1
    def __len__(self):
        return self.steps
    def __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] = [], [], []

6.bert模型

def acc_top2(y_true, y_pred):
    return top_k_categorical_accuracy(y_true, y_pred, k=2)

def build_bert(nclass):
    bert_model = load_trained_model_from_checkpoint(config_path, checkpoint_path)

    for l in bert_model.layers:
        l.trainable = True

    x1_in = Input(shape=(None,))
    x2_in = Input(shape=(None,))

    x = bert_model([x1_in, x2_in])
    x = Lambda(lambda x: x[:, 0])(x)
    p = Dense(nclass, activation='softmax')(x)

    model = Model([x1_in, x2_in], p)
    model.compile(loss='categorical_crossentropy', 
                  optimizer=Adam(1e-5),
                  metrics=['accuracy', acc_top2])
    print(model.summary())
    return model

7.模型训练

def run_cv(nfold, data, data_label, 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)
        plateau = ReduceLROnPlateau(monitor="val_acc", verbose=1, mode='max', factor=0.5, patience=2)
        checkpoint = ModelCheckpoint('./bert_dump/' + str(i) + '.hdf5', monitor='val_acc', 
                                         verbose=2, save_best_only=True, mode='max',save_weights_only=True)
        
        train_D = data_generator(X_train, shuffle=True)
        valid_D = data_generator(X_valid, shuffle=True)
        test_D = data_generator(data_test, shuffle=False)
        
        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 model
        train_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 model; gc.collect()
        K.clear_session()
        
        # break
        
    return train_model_pred, test_model_pred

train_model_pred, test_model_pred = run_cv(10, np.array(Train), None, np.array(Test))

load_trained_model_from_checkpoint() 函数：
用于加载官方训练好的模型模型，参数如下

• config_file：JSON 配置文件路径
• checkpoint_file：checkpoint 文件路径
• training：True 的话，会返回整个模型，否则会忽略 MLM 和 NSP 部分。默认 False
• trainable：模型是否可训练，默认和 training 设置一样
• output_layer_num：多少个FeedForward-Norm层的输出被连接为单个输出。仅在 training 为 False 时可用。默认1
• seq_len：如果这个数值比配置文件中的长度小，position embeddings 会被切成适用于这个长度。默认1e9

使用bert模型的话后面可以加上其他网络用来实现相应的功能。