3.6波士顿房价预测：回归问题

from tensorflow.keras.datasets import boston_housing

(train_data,train_targets),(test_data,test_targets) = boston_housing.load_data()

train_data.shape

(404, 13)

test_data.shape

(102, 13)

train_targets[0]

15.2

#准备数据时对每个特征数据进行标准化预处理，方法是对输入数据的每个特征（输入数据矩阵中的列）减去

#准备数据
mean = train_data.mean(axis=0)
train_data -= mean
std = train_data.std(axis=0)
train_data /= std

test_data -= mean
test_data /= std

#模型定义
from tensorflow.keras import models
from tensorflow.keras import layers

def build_model():#因为要将同一个模型多次实例化，所以用一个函数来构建模型
    model = models.Sequential()
    model.add(layers.Dense(64,activation='relu',input_shape=(train_data.shape[1],)))
    model.add(layers.Dense(64,activation='relu'))
    model.add(layers.Dense(1))#线性层
    model.compile(optimizer='rmsprop',loss='mse',metrics=['mae'])#使用均方误差为损失函数，训练过程监控平均绝对误差mae
    return model

#使用K折验证来验证方法（k通常取4或者5）
#k折验证用于数据量小导致验证集非常小的情况，可以很好的处理由于验证集划分造成验证分数的方差很大的问题，从而更可靠的评估模型
import numpy as np

k=4
num_val_samples = len(train_data)//k
num_epochs = 100
all_scores = []

for i in range(k):
print(‘processing fold #’,i)
val_data = train_data[i*num_val_samples:(i+1)num_val_samples]
val_targets = train_targets[inum_val_samples:(i+1)*num_val_samples]

partial_train_data = np.concatenate(
[train_data[:i*num_val_samples],
train_data[(i+1)*num_val_samples:]],
axis=0)
partial_train_targets = np.concatenate(
[train_targets[:i*num_val_samples],
train_targets[(i+1)*num_val_samples:]],
axis=0)

model = build_model()
model.fit(partial_train_data,partial_train_targets,epochs=num_epochs,batch_size=1,verbose=0)#训练模式（静默模式，verbose=0）
val_mse,val_mae = model.evaluate(val_data,val_targets,verbose=0)#在验证数据上评估模型
all_scores.append(val_mae)

print(all_scores)
print(np.mean(all_scores))

#保存每折的k折验证
import numpy as np

k=4
num_val_samples = len(train_data)//k
num_epochs = 500
all_mae_histories = []

for i in range(k):
    print('processing fold #',i)
    val_data = train_data[i*num_val_samples:(i+1)*num_val_samples]
    val_targets = train_targets[i*num_val_samples:(i+1)*num_val_samples]
    
    partial_train_data = np.concatenate(
    [train_data[:i*num_val_samples],
    train_data[(i+1)*num_val_samples:]],
    axis=0)
    partial_train_targets = np.concatenate(
    [train_targets[:i*num_val_samples],
    train_targets[(i+1)*num_val_samples:]],
    axis=0)
    
    model = build_model()
    history = model.fit(partial_train_data,partial_train_targets,
                        validation_data=(val_data,val_targets),
                        epochs=num_epochs,batch_size=1,verbose=0)#训练模式（静默模式，verbose=0）
   

processing fold # 0
processing fold # 1
processing fold # 2
processing fold # 3

history_dict=history.history
history_dict.keys()

dict_keys(['loss', 'mae', 'val_loss', 'val_mae'])

想要取出history.history字典中某一组数据 首先应该看history.history中有哪些键值对，然后赋值给另一个变量，在将其扩展

mae_history = history.history['val_mae']
all_mae_histories.append(mae_history)
    

#计算所有伦茨中的K折验证对分数平均值
average_mae_history =[
    np.mean([x[i] for x in all_mae_histories]) for i in range(num_epochs)
]

print(len(average_mae_history))

500

#绘制验证分数
import matplotlib.pyplot as plt

plt.plot(range(1,len(average_mae_history)+1),average_mae_history)#range(1,len(average_mae_history)+1)是横坐标，average_mae_history是纵坐标
plt.xlabel('Epochs')
plt.ylabel('Validation MAE')
plt.show()

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-BCmq5Oml-1615636827919)(output_13_0.png)]

#绘制验证分数（删除前10个数据点）
def smooth_curve(points,factor=0.9):
    smoothed_points = []
    for point in points:
        if smoothed_points:
            previous = smoothed_points[-1]
            smoothed_points.append(previous*factor+point*(1-factor))
        else:
            smoothed_points.append(point)
    return smoothed_points

smooth_mae_history = smooth_curve(average_mae_history[10:])

plt.plot(range(1,len(smooth_mae_history)+1),smooth_mae_history)
plt.xlabel('Epochs')
plt.ylabel('Validation MAE')
plt.show()

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-I2w5FShP-1615636827920)(output_14_0.png)]

#训练最终模型
model = build_model()
model.fit(train_data,train_targets,epochs=80,batch_size=16,verbose=0)

test_mse_score,test_mae_score = model.evaluate(test_data,test_targets)
test_mae_score

102/102 [==============================] - 0s 910us/sample - loss: 17.7348 - mae: 2.6016





2.6015923

本文参考转载《Pyhton 深度学习》keras之父弗朗索瓦.肖莱这本书