本文共 15271 字，大约阅读时间需要 50 分钟。

模块调用

2021/3/8 周一：基于模块调用部分（如下）bug，重装Anaconda与Tensorflow，解决bug。

import numpy as npfrom bert4keras.backend import keras, Kfrom bert4keras.models import build_transformer_modelfrom bert4keras.tokenizers import Tokenizerfrom bert4keras.optimizers import Adamfrom bert4keras.snippets import sequence_padding, DataGeneratorfrom bert4keras.snippets import open, ViterbiDecoderfrom bert4keras.layers import ConditionalRandomFieldfrom keras.layers import Densefrom keras.models import Modelfrom tqdm import tqdmfrom tensorflow import ConfigProtofrom tensorflow import InteractiveSession#上述两句中的“tensorflow”原为 tensorflow.compat.v1

报错原因与解决方案：1.Anaconda内置的Python版本与Tensorflow版本不一致——未注意到Tensorflow不太适合Python3.7以上的版本。重装内置3.6Python版本的Anaconda，再重装相应版本的Tensorflow，即可解决。2.安装Tensorflow-GPU多次，均以失败告终。原来，我的电脑的显卡配置并不支持GPU版本。

小结：有些时候，bug“缠身”，不妨直接卸载重装，可能更节省时间。

2021/3/9 周二：开组会，接受批评，反思自己。

保存、加载并使用已训练的Bert/Albert-CRF模型

2021/3/10：使用训练好的Bert/Albert-CRF模型，同时，在此基础上，加一层BiLSTM网络，得修改后的Albert-BiLSTM-CRF模型（见下一篇文章），开始训练。

'''if __name__ == '__main__':    evaluator = Evaluate()    train_generator = data_generator(train_data, batch_size)    model.fit_generator(        train_generator.forfit(),        steps_per_epoch=len(train_generator),        epochs=epochs,        callbacks=[evaluator]    )else:    model.load_weights('best_model.weights')'''model.load_weights('best_model.weights')NER = NamedEntityRecognizer(trans=K.eval(CRF.trans), starts=[0], ends=[0])ner=NER.recognize("我在厦门")print(ner)

流程：先训练模型，得训练好的权重文件，此时，也可同时得到训练好的模型文件。加载上述权重文件，再修改最后的函数调用部分即可。

注意：类的实例化。

Bert/Albert-CRF模型完整代码

仅训练与评估模型：

import numpy as npfrom bert4keras.backend import keras, Kfrom bert4keras.models import build_transformer_modelfrom bert4keras.tokenizers import Tokenizerfrom bert4keras.optimizers import Adamfrom bert4keras.snippets import sequence_padding, DataGeneratorfrom bert4keras.snippets import open, ViterbiDecoderfrom bert4keras.layers import ConditionalRandomFieldfrom keras.layers import Densefrom keras.models import Modelfrom tqdm import tqdmfrom tensorflow import ConfigProtofrom tensorflow import InteractiveSession#上述两句中的“tensorflow”原为 tensorflow.compat.v1config = ConfigProto()# config.gpu_options.per_process_gpu_memory_fraction = 0.2config.gpu_options.allow_growth = Truesession = InteractiveSession(config=config)maxlen = 256epochs = 1#10batch_size = 16bert_layers = 12learing_rate = 1e-5  # bert_layers越小，学习率应该要越大crf_lr_multiplier = 10  # 必要时扩大CRF层的学习率#1000# # bert配置# config_path = './bert_model/chinese_L-12_H-768_A-12/bert_config.json'# checkpoint_path = './bert_model/chinese_L-12_H-768_A-12/bert_model.ckpt'# dict_path = './bert_model/chinese_L-12_H-768_A-12/vocab.txt'#albert配置config_path = './bert_model/albert_large/albert_config.json'checkpoint_path = './bert_model/albert_large/model.ckpt-best'dict_path = './bert_model/albert_large/vocab_chinese.txt'def load_data(filename):    D = []    with open(filename, encoding='utf-8') as f:        f = f.read()        for l in f.split('\n\n'):            if not l:                continue            d, last_flag = [], ''            for c in l.split('\n'):                char, this_flag = c.split(' ')                if this_flag == 'O' and last_flag == 'O':                    d[-1][0] += char                elif this_flag == 'O' and last_flag != 'O':                    d.append([char, 'O'])                elif this_flag[:1] == 'B':                    d.append([char, this_flag[2:]])                else:                    d[-1][0] += char                last_flag = this_flag            D.append(d)    return D# 标注数据train_data = load_data('./data/example.train')valid_data = load_data('./data/example.dev')test_data = load_data('./data/example.test')# 建立分词器tokenizer = Tokenizer(dict_path, do_lower_case=True)# 类别映射labels = ['PER', 'LOC', 'ORG']id2label = dict(enumerate(labels))label2id = {j: i for i, j in id2label.items()}num_labels = len(labels) * 2 + 1class data_generator(DataGenerator):    """数据生成器    """    def __iter__(self, random=False):        batch_token_ids, batch_segment_ids, batch_labels = [], [], []        for is_end, item in self.sample(random):            token_ids, labels = [tokenizer._token_start_id], [0]            for w, l in item:                w_token_ids = tokenizer.encode(w)[0][1:-1]                if len(token_ids) + len(w_token_ids) < maxlen:                    token_ids += w_token_ids                    if l == 'O':                        labels += [0] * len(w_token_ids)                    else:                        B = label2id[l] * 2 + 1                        I = label2id[l] * 2 + 2                        labels += ([B] + [I] * (len(w_token_ids) - 1))                else:                    break            token_ids += [tokenizer._token_end_id]            labels += [0]            segment_ids = [0] * len(token_ids)            batch_token_ids.append(token_ids)            batch_segment_ids.append(segment_ids)            batch_labels.append(labels)            if len(batch_token_ids) == self.batch_size or is_end:                batch_token_ids = sequence_padding(batch_token_ids)                batch_segment_ids = sequence_padding(batch_segment_ids)                batch_labels = sequence_padding(batch_labels)                yield [batch_token_ids, batch_segment_ids], batch_labels                batch_token_ids, batch_segment_ids, batch_labels = [], [], []"""后面的代码使用的是bert类型的模型，如果你用的是albert，那么前几行请改为："""model = build_transformer_model(    config_path,    checkpoint_path,    model='albert',)output_layer = 'Transformer-FeedForward-Norm'output = model.get_layer(output_layer).get_output_at(bert_layers - 1)# model = build_transformer_model(#     config_path,#     checkpoint_path,# )## output_layer = 'Transformer-%s-FeedForward-Norm' % (bert_layers - 1)# output = model.get_layer(output_layer).outputoutput = Dense(num_labels)(output)CRF = ConditionalRandomField(lr_multiplier=crf_lr_multiplier)output = CRF(output)model = Model(model.input, output)model.summary()model.compile(    loss=CRF.sparse_loss,    optimizer=Adam(learing_rate),    metrics=[CRF.sparse_accuracy])class NamedEntityRecognizer(ViterbiDecoder):    """命名实体识别器    """    def recognize(self, text):        tokens = tokenizer.tokenize(text)        while len(tokens) > 512:            tokens.pop(-2)        mapping = tokenizer.rematch(text, tokens)        token_ids = tokenizer.tokens_to_ids(tokens)        segment_ids = [0] * len(token_ids)        nodes = model.predict([[token_ids], [segment_ids]])[0]        labels = self.decode(nodes)        entities, starting = [], False        for i, label in enumerate(labels):            if label > 0:                if label % 2 == 1:                    starting = True                    entities.append([[i], id2label[(label - 1) // 2]])                elif starting:                    entities[-1][0].append(i)                else:                    starting = False            else:                starting = False        return [(text[mapping[w[0]][0]:mapping[w[-1]][-1] + 1], l)                for w, l in entities]NER = NamedEntityRecognizer(trans=K.eval(CRF.trans), starts=[0], ends=[0])def evaluate(data):    """评测函数    """    X, Y, Z = 1e-10, 1e-10, 1e-10    for d in tqdm(data):        text = ''.join([i[0] for i in d])        R = set(NER.recognize(text))        T = set([tuple(i) for i in d if i[1] != 'O'])        X += len(R & T)        Y += len(R)        Z += len(T)    f1, precision, recall = 2 * X / (Y + Z), X / Y, X / Z    return f1, precision, recallclass Evaluate(keras.callbacks.Callback):    def __init__(self):        self.best_val_f1 = 0    def on_epoch_end(self, epoch, logs=None):        trans = K.eval(CRF.trans)        NER.trans = trans        print(NER.trans)        f1, precision, recall = evaluate(valid_data)        # 保存最优        if f1 >= self.best_val_f1:            self.best_val_f1 = f1            model.save_weights('best_model.weights')        print(            'valid:  f1: %.5f, precision: %.5f, recall: %.5f, best f1: %.5f\n' %            (f1, precision, recall, self.best_val_f1)        )        f1, precision, recall = evaluate(test_data)        print(            'test:  f1: %.5f, precision: %.5f, recall: %.5f\n' %            (f1, precision, recall)        )if __name__ == '__main__':    evaluator = Evaluate()    train_generator = data_generator(train_data, batch_size)    model.fit_generator(        train_generator.forfit(),        steps_per_epoch=len(train_generator),        epochs=epochs,        callbacks=[evaluator]    )else:    model.load_weights('best_model.weights')

使用训练好的模型，简例：

import numpy as npfrom bert4keras.backend import keras, Kfrom bert4keras.models import build_transformer_modelfrom bert4keras.tokenizers import Tokenizerfrom bert4keras.optimizers import Adamfrom bert4keras.snippets import sequence_padding, DataGeneratorfrom bert4keras.snippets import open, ViterbiDecoderfrom bert4keras.layers import ConditionalRandomFieldfrom keras.layers import Densefrom keras.models import Modelfrom tqdm import tqdmfrom tensorflow import ConfigProtofrom tensorflow import InteractiveSession#上述两句中的“tensorflow”原为 tensorflow.compat.v1config = ConfigProto()# config.gpu_options.per_process_gpu_memory_fraction = 0.2config.gpu_options.allow_growth = Truesession = InteractiveSession(config=config)maxlen = 256epochs = 1#10batch_size = 16bert_layers = 12learing_rate = 1e-5  # bert_layers越小，学习率应该要越大crf_lr_multiplier = 10  # 必要时扩大CRF层的学习率#1000# # bert配置# config_path = './bert_model/chinese_L-12_H-768_A-12/bert_config.json'# checkpoint_path = './bert_model/chinese_L-12_H-768_A-12/bert_model.ckpt'# dict_path = './bert_model/chinese_L-12_H-768_A-12/vocab.txt'#albert配置config_path = './bert_model/albert_large/albert_config.json'checkpoint_path = './bert_model/albert_large/model.ckpt-best'dict_path = './bert_model/albert_large/vocab_chinese.txt'def load_data(filename):    D = []    with open(filename, encoding='utf-8') as f:        f = f.read()        for l in f.split('\n\n'):            if not l:                continue            d, last_flag = [], ''            for c in l.split('\n'):                char, this_flag = c.split(' ')                if this_flag == 'O' and last_flag == 'O':                    d[-1][0] += char                elif this_flag == 'O' and last_flag != 'O':                    d.append([char, 'O'])                elif this_flag[:1] == 'B':                    d.append([char, this_flag[2:]])                else:                    d[-1][0] += char                last_flag = this_flag            D.append(d)    return D# 标注数据train_data = load_data('./data/example.train')valid_data = load_data('./data/example.dev')test_data = load_data('./data/example.test')# 建立分词器tokenizer = Tokenizer(dict_path, do_lower_case=True)# 类别映射labels = ['PER', 'LOC', 'ORG']id2label = dict(enumerate(labels))label2id = {j: i for i, j in id2label.items()}num_labels = len(labels) * 2 + 1class data_generator(DataGenerator):    """数据生成器    """    def __iter__(self, random=False):        batch_token_ids, batch_segment_ids, batch_labels = [], [], []        for is_end, item in self.sample(random):            token_ids, labels = [tokenizer._token_start_id], [0]            for w, l in item:                w_token_ids = tokenizer.encode(w)[0][1:-1]                if len(token_ids) + len(w_token_ids) < maxlen:                    token_ids += w_token_ids                    if l == 'O':                        labels += [0] * len(w_token_ids)                    else:                        B = label2id[l] * 2 + 1                        I = label2id[l] * 2 + 2                        labels += ([B] + [I] * (len(w_token_ids) - 1))                else:                    break            token_ids += [tokenizer._token_end_id]            labels += [0]            segment_ids = [0] * len(token_ids)            batch_token_ids.append(token_ids)            batch_segment_ids.append(segment_ids)            batch_labels.append(labels)            if len(batch_token_ids) == self.batch_size or is_end:                batch_token_ids = sequence_padding(batch_token_ids)                batch_segment_ids = sequence_padding(batch_segment_ids)                batch_labels = sequence_padding(batch_labels)                yield [batch_token_ids, batch_segment_ids], batch_labels                batch_token_ids, batch_segment_ids, batch_labels = [], [], []"""后面的代码使用的是bert类型的模型，如果你用的是albert，那么前几行请改为："""model = build_transformer_model(    config_path,    checkpoint_path,    model='albert',)output_layer = 'Transformer-FeedForward-Norm'output = model.get_layer(output_layer).get_output_at(bert_layers - 1)# model = build_transformer_model(#     config_path,#     checkpoint_path,# )## output_layer = 'Transformer-%s-FeedForward-Norm' % (bert_layers - 1)# output = model.get_layer(output_layer).outputoutput = Dense(num_labels)(output)CRF = ConditionalRandomField(lr_multiplier=crf_lr_multiplier)output = CRF(output)model = Model(model.input, output)model.summary()model.compile(    loss=CRF.sparse_loss,    optimizer=Adam(learing_rate),    metrics=[CRF.sparse_accuracy])class NamedEntityRecognizer(ViterbiDecoder):    """命名实体识别器    """    def recognize(self,text):        tokens = tokenizer.tokenize(text)        while len(tokens) > 512:            tokens.pop(-2)        mapping = tokenizer.rematch(text, tokens)        token_ids = tokenizer.tokens_to_ids(tokens)        segment_ids = [0] * len(token_ids)        nodes = model.predict([[token_ids], [segment_ids]])[0]        labels = self.decode(nodes)        entities, starting = [], False        for i, label in enumerate(labels):            if label > 0:                if label % 2 == 1:                    starting = True                    entities.append([[i], id2label[(label - 1) // 2]])                elif starting:                    entities[-1][0].append(i)                else:                    starting = False            else:                starting = False        return [(text[mapping[w[0]][0]:mapping[w[-1]][-1] + 1], l)                for w, l in entities]def evaluate(data):    """评测函数    """    X, Y, Z = 1e-10, 1e-10, 1e-10    for d in tqdm(data):        text = ''.join([i[0] for i in d])        R = set(NER.recognize(text))        T = set([tuple(i) for i in d if i[1] != 'O'])        X += len(R & T)        Y += len(R)        Z += len(T)    f1, precision, recall = 2 * X / (Y + Z), X / Y, X / Z    return f1, precision, recallclass Evaluate(keras.callbacks.Callback):    def __init__(self):        self.best_val_f1 = 0    def on_epoch_end(self, epoch, logs=None):        trans = K.eval(CRF.trans)        NER.trans = trans        print(NER.trans)        f1, precision, recall = evaluate(valid_data)        # 保存最优        if f1 >= self.best_val_f1:            self.best_val_f1 = f1            model.save_weights('best_model.weights')        print(            'valid:  f1: %.5f, precision: %.5f, recall: %.5f, best f1: %.5f\n' %            (f1, precision, recall, self.best_val_f1)        )        f1, precision, recall = evaluate(test_data)        print(            'test:  f1: %.5f, precision: %.5f, recall: %.5f\n' %            (f1, precision, recall)        )model.load_weights('best_model.weights')NER = NamedEntityRecognizer(trans=K.eval(CRF.trans), starts=[0], ends=[0])ner=NER.recognize("我在厦门")print(ner)

小总结：多尝试，改改改 + 基础编程知识（如：类的使用）要扎实。

转载地址：https://blog.csdn.net/YWP_2016/article/details/114543741 如侵犯您的版权，请留言回复原文章的地址，我们会给您删除此文章，给您带来不便请您谅解！