python机器学习 第六章
日记目录
一. 单词级one-hot编码
将每个单词与一个唯一的整数索引相关联,然后将这个整数索引 i 转换为长度为 N 的二进制向量(N 是词表大小),这个向量只有第 i 个元素是 1,其余元素都为 0。
import numpy as np
samples = ['The cat sat on the mat.', 'The dog ate my homework.']
##构建索引
token_index = {}
for sample in samples:
##对样本分词,实际引用中需去除其他符号,如:.split("str",num),str默认所有空字符,num分几次,默认为-1,分割所有
for word in sample.split():
if word not in token_index:
## 为每个唯一单词指定一个唯一索引,至少为1
token_index[word] = len(token_index) + 1
max_length = 10
## 借助numpy库建立0矩阵,3维,
results = np.zeros(shape=(len(samples),
max_length,
max(token_index.values()) + 1))
##enumerate() 函数用于将一个可遍历的数据对象(如列表、元组或字符串)组合为一个索引序列,同时列出数据和数据下标
for i, sample in enumerate(samples):
##(j索引,word字符串),[:max_length]前10个
for j, word in list(enumerate(sample.split()))[:max_length]:
index = token_index.get(word)
results[i, j, index] = 1.
二. 字符级的one-hot编码
import string
samples = ['The cat sat on the mat.', 'The dog ate my homework.']
##所有所有可打印的 ASCII 字符
characters = string.printable
##建立字典
##zip(),将参数打包为元组,后返回其组成的列表
token_index = dict(zip(range(1, len(characters) + 1), characters))
max_length = 50
results = np.zeros((len(samples), max_length, max(token_index.keys()) + 1))
for i, sample in enumerate(samples):
for j, character in enumerate(sample):
index = token_index.get(character)
results[i, j, index] = 1.
三. 用keras实现单词级的one-hot编码
from keras.preprocessing.text import Tokenizer
samples = ['The cat sat on the mat.', 'The dog ate my homework.']
##分词器,只考虑前1000个常用单词
tokenizer = Tokenizer(num_words=1000)
## 构建索引
tokenizer.fit_on_texts(samples)
## 序列列表,返回为字符串转化为整数索引组成的列表
sequences = tokenizer.texts_to_sequences(samples)
## 将字符串转化为二进制向量
one_hot_results = tokenizer.texts_to_matrix(samples, mode='binary')
##单词索引
word_index = tokenizer.word_index
print('Found %s unique tokens.' % len(word_index))
四. 使用散列技巧的单词级的 one-hot 编码(简单示例)
samples = ['The cat sat on the mat.', 'The dog ate my homework.']
dimensionality = 1000
max_length = 10
results = np.zeros((len(samples), max_length, dimensionality))
for i, sample in enumerate(samples):
for j, word in list(enumerate(sample.split()))[:max_length]:
index = abs(hash(word)) % dimensionality
results[i, j, index] = 1.
五. 循环神经网路
循环神经网络(RNN):它处理序列的方式是,遍历所有序列元素,并保存一个状态(state),其中包含与已查看内容相关的信息
简单RNN的Numpy实现
import numpy as np
##输入序列的时间步数
timesteps = 100
##输入特征空间的维度
input_features = 32
##输出特征空间的维度
output_features = 64
##输入数据,随机随机噪声【100*32】
inputs = np.random.random((timesteps, input_features))
##初始全零向量
state_t = np.zeros((output_features,))
##随机权重矩阵
W = np.random.random((output_features, input_features))
U = np.random.random((output_features, output_features))
b = np.random.random((output_features,))
successive_outputs = []
##input_t为(32, )向量
for input_t in inputs:
##由输入和当前状态(前一个输出)计算得到当前输出
output_t = np.tanh(np.dot(W, input_t) + np.dot(U, state_t) + b)
##将输出值存储于列表中
successive_outputs.append(output_t)
##更新状态
state_t = output_t
## 最终输出为形状为(100,64)的二维向量
final_output_sequence = np.stack(successive_outputs, axis=0)
六. IMDB电影评论与RNN
数据处理
from keras.datasets import imdb
from keras.preprocessing import sequence
max_features = 10000
maxlen = 500
batch_size = 32
print('Loading data...')
(input_train, y_train), (input_test, y_test) = imdb.load_data(
num_words=max_features)
print(len(input_train), 'train sequences')
print(len(input_test), 'test sequences')
print('Pad sequences (samples x time)')
input_train = sequence.pad_sequences(input_train, maxlen=maxlen)
input_test = sequence.pad_sequences(input_test, maxlen=maxlen)
print('input_train shape:', input_train.shape)
print('input_test shape:', input_test.shape)
训练模型
from keras.layers import Dense
model = Sequential()
model.add(Embedding(max_features, 32))
model.add(SimpleRNN(32))
model.add(Dense(1, activation='sigmoid'))
model.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['acc'])
history = model.fit(input_train, y_train,
epochs=10,
batch_size=128,
validation_split=0.2)
绘制结果
import matplotlib.pyplot as plt
acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs = range(1, len(acc) + 1)
plt.plot(epochs, acc, 'bo', label='Training acc')
plt.plot(epochs, val_acc, 'b', label='Validation acc')
plt.title('Training and validation accuracy')
plt.legend()
plt.figure()
plt.plot(epochs, loss, 'bo', label='Training loss')
plt.plot(epochs, val_loss, 'b', label='Validation loss')
plt.title('Training and validation loss')
plt.legend()
plt.show()