def _pred(sentence, temperature=1): if len(sentence) < max_len: print('in def _pred,length error ') return sentence = sentence[-max_len:] x_pred = np.zeros((1, max_len, len(words))) for t, char in enumerate(sentence): x_pred[0, t, word2numF(char)] = 1. preds = model.predict(x_pred, verbose=0)[0] preds = np.asarray(preds).astype('float64') exp_preds = np.power(preds, temperature) # 计算所有备选输出文字概率`preds`的`temperature`次方 preds = exp_preds / np.sum(exp_preds) # 重新统计概率分布 pro = np.random.choice(range(len(preds)), 1, p=preds) # 根据新概率随机选择候选文字 next_index = int(pro.squeeze()) next_char = num2word[next_index] return next_char
时间: 2023-11-28 09:06:40 浏览: 39
这段代码是一个用于生成文本的函数 `_pred`,它接受一个字符串 `sentence` 和一个温度参数 `temperature` 作为输入,返回一个字符作为输出。这个函数的实现基于一个训练好的模型,并利用该模型对给定的 `sentence` 进行预测,得到下一个字符的概率分布。然后,使用温度参数 `temperature` 对概率分布进行调节,加入一定的随机性,最后根据新的概率分布随机选择一个候选字符作为输出,并将其返回。
相关问题
import ast from dataclasses import dataclass from typing import List import pandas as pd import json ["text", "六十一岁还能办什么保险"] @dataclass class FAQ: title: str sim_questions: List[str] answer: str faq_id: int ori_data = pd.read_csv('baoxianzhidao_filter.csv') data = [] exist_titles = set() for index, row in enumerate(ori_data.iterrows()): row_dict = row[1] title = row_dict['title'] if title not in exist_titles: data.append(FAQ(title=title, answer=row_dict['reply'], sim_questions=[title], faq_id=index)) exist_titles.add(title) from modelscope.pipelines import pipeline from modelscope.utils.constant import Tasks pipeline_ins = pipeline(Tasks.faq_question_answering, 'damo/nlp_mgimn_faq-question-answering_chinese-base') bsz = 32 all_sentence_vecs = [] batch = [] sentence_list = [faq.title for faq in data] for i,sent in enumerate(sentence_list): batch.append(sent) if len(batch) == bsz or (i == len(sentence_list)-1 and len(batch)>0): # if i == len(sentence_list)-1 and len(batch)>0: sentence_vecs = pipeline_ins.get_sentence_embedding(batch) all_sentence_vecs.extend(sentence_vecs) batch.clear() import faiss import numpy as np hidden_size = pipeline_ins.model.network.bert.config.hidden_size # hidden_size = pipeline_ins.model.bert.config.hidden_size index = faiss.IndexFlatIP(hidden_size) vecs = np.asarray(all_sentence_vecs, dtype='float32') index.add(vecs) from modelscope.outputs import OutputKeys def ask_faq(input, history=[]): # step1: get sentence vector of query query_vec = pipeline_ins.get_sentence_embedding([input])[0] query_vec = np.asarray(query_vec, dtype='float32').reshape([1, -1]) # step2: faq dense retrieval _, indices = index.search(query_vec, k=30) # step3: build support set support_set = [] for i in indices.tolist()[0]: faq = data[i] support_set.append({"text": faq.title, "label": faq.faq_id, "index": i}) # step4: faq ranking rst = pipeline_ins(input={"query_set": input, "support_set": support_set}) rst = rst[OutputKeys.OUTPUT][0][0] pred_label = rst['label'] pred_score = rst['score'] # get answer by faq_id pred_answer = "" pred_title = "" for faq in data: if faq.faq_id == pred_label: pred_answer = faq.answer pred_title = faq.title break history.append((f'{pred_answer}|(pred_title:{pred_title},pred_score:{pred_score:.3f})')) return history优化这段代码
这段代码是一个Python脚本,用于读取CSV文件中的保险相关问题和答案,构建一个FAQ对象(包含问题、答案、相似问题和FAQ ID),并使用modelscope库中的pipeline进行常见问题解答。其中用到了ast、dataclass、List、pandas和json等Python模块。具体实现过程请参考上述代码。
解决AttributeError: module 'tensorflow._api.v2.compat.v1.nn.rnn_cell' has no attribute 'seq2seq'
在 TensorFlow 2.0 中,`tf.nn.rnn_cell` 模块已经被移除了,因此 `tf.nn.rnn_cell.seq2seq` 也不再可用。相反,您可以使用 `tf.keras.layers` 中的相应函数来构建您的模型。下面是一个使用 `tf.keras` 实现 seq2seq 模型的示例:
``` python
import tensorflow as tf
# 定义编码器
class Encoder(tf.keras.Model):
def __init__(self, vocab_size, embedding_dim, enc_units, batch_sz):
super(Encoder, self).__init__()
self.batch_sz = batch_sz
self.enc_units = enc_units
self.embedding = tf.keras.layers.Embedding(vocab_size, embedding_dim)
self.gru = tf.keras.layers.GRU(self.enc_units, return_sequences=True, return_state=True, recurrent_initializer='glorot_uniform')
def call(self, x, hidden):
x = self.embedding(x)
output, state = self.gru(x, initial_state = hidden)
return output, state
def initialize_hidden_state(self):
return tf.zeros((self.batch_sz, self.enc_units))
# 定义注意力层
class BahdanauAttention(tf.keras.layers.Layer):
def __init__(self, units):
super(BahdanauAttention, self).__init__()
self.W1 = tf.keras.layers.Dense(units)
self.W2 = tf.keras.layers.Dense(units)
self.V = tf.keras.layers.Dense(1)
def call(self, query, values):
# query: 上一时间步的隐藏状态,shape=(batch_size, hidden_size)
# values: 编码器的输出,shape=(batch_size, max_length, hidden_size)
hidden_with_time_axis = tf.expand_dims(query, 1)
score = self.V(tf.nn.tanh(
self.W1(values) + self.W2(hidden_with_time_axis)))
# attention_weights shape == (batch_size, max_length, 1)
attention_weights = tf.nn.softmax(score, axis=1)
# context_vector shape after sum == (batch_size, hidden_size)
context_vector = attention_weights * values
context_vector = tf.reduce_sum(context_vector, axis=1)
return context_vector, attention_weights
# 定义解码器
class Decoder(tf.keras.Model):
def __init__(self, vocab_size, embedding_dim, dec_units, batch_sz):
super(Decoder, self).__init__()
self.batch_sz = batch_sz
self.dec_units = dec_units
self.embedding = tf.keras.layers.Embedding(vocab_size, embedding_dim)
self.gru = tf.keras.layers.GRU(self.dec_units, return_sequences=True, return_state=True, recurrent_initializer='glorot_uniform')
self.fc = tf.keras.layers.Dense(vocab_size)
# 用于注意力
self.attention = BahdanauAttention(self.dec_units)
def call(self, x, hidden, enc_output):
# enc_output shape == (batch_size, max_length, hidden_size)
context_vector, attention_weights = self.attention(hidden, enc_output)
# x shape after passing through embedding == (batch_size, 1, embedding_dim)
x = self.embedding(x)
# 将上一时间步的隐藏状态和注意力向量拼接起来作为输入传给 GRU
x = tf.concat([tf.expand_dims(context_vector, 1), x], axis=-1)
# 将拼接后的向量传给 GRU
output, state = self.gru(x)
# output shape == (batch_size * 1, hidden_size)
output = tf.reshape(output, (-1, output.shape[2]))
# output shape == (batch_size, vocab)
x = self.fc(output)
return x, state, attention_weights
# 定义损失函数和优化器
optimizer = tf.keras.optimizers.Adam()
loss_object = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True, reduction='none')
def loss_function(real, pred):
mask = tf.math.logical_not(tf.math.equal(real, 0))
loss_ = loss_object(real, pred)
mask = tf.cast(mask, dtype=loss_.dtype)
loss_ *= mask
return tf.reduce_mean(loss_)
# 定义训练步骤
@tf.function
def train_step(inp, targ, enc_hidden):
loss = 0
with tf.GradientTape() as tape:
enc_output, enc_hidden = encoder(inp, enc_hidden)
dec_hidden = enc_hidden
dec_input = tf.expand_dims([tokenizer.word_index['<start>']] * BATCH_SIZE, 1)
# teacher forcing - 将目标词作为下一个输入传给解码器
for t in range(1, targ.shape[1]):
# 将编码器的输出和上一时间步的隐藏状态传给解码器
predictions, dec_hidden, _ = decoder(dec_input, dec_hidden, enc_output)
loss += loss_function(targ[:, t], predictions)
# 使用 teacher forcing
dec_input = tf.expand_dims(targ[:, t], 1)
batch_loss = (loss / int(targ.shape[1]))
variables = encoder.trainable_variables + decoder.trainable_variables
gradients = tape.gradient(loss, variables)
optimizer.apply_gradients(zip(gradients, variables))
return batch_loss
# 定义预测函数
def evaluate(sentence):
attention_plot = np.zeros((max_length_targ, max_length_inp))
sentence = preprocess_sentence(sentence)
inputs = [tokenizer.word_index[i] for i in sentence.split(' ')]
inputs = tf.keras.preprocessing.sequence.pad_sequences([inputs], maxlen=max_length_inp, padding='post')
inputs = tf.convert_to_tensor(inputs)
result = ''
hidden = [tf.zeros((1, units))]
enc_out, enc_hidden = encoder(inputs, hidden)
dec_hidden = enc_hidden
dec_input = tf.expand_dims([tokenizer.word_index['<start>']], 0)
for t in range(max_length_targ):
predictions, dec_hidden, attention_weights = decoder(dec_input, dec_hidden, enc_out)
# 存储注意力权重以便后面制图
attention_weights = tf.reshape(attention_weights, (-1, ))
attention_plot[t] = attention_weights.numpy()
predicted_id = tf.argmax(predictions[0]).numpy()
result += tokenizer.index_word[predicted_id] + ' '
if tokenizer.index_word[predicted_id] == '<end>':
return result, sentence, attention_plot
# 将预测的 ID 作为下一个解码器输入的 ID
dec_input = tf.expand_dims([predicted_id], 0)
return result, sentence, attention_plot
```
在上面的代码中,我们使用了 `tf.keras.layers` 中的 `Embedding`、`GRU` 和 `Dense` 层来构建编码器和解码器,使用 `tf.keras.optimizers.Adam` 作为优化器,使用 `tf.keras.losses.SparseCategoricalCrossentropy` 作为损失函数。同时,我们还定义了一个 `BahdanauAttention` 层来实现注意力机制。
相关推荐
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)