TensorFlow2实战-基于神经网络的油耗预测

TensorFlow2-基于神经网络的耗油预测问题

采用的是Auto MPG数据集，包含5个字段，具体信息见前5行数据和相应的字段解释信息：

image image

# 1. 数据集
import pandas as pd
import tensorflow as tf
from tensorflow import keras
dataset_path = keras.utils.get_file("auto-mpg.data", "http://archive.ics.uci.edu/ml/machine-learning-databases/auto-mpg/auto-mpg.data")
column_names = ['MPG','Cylinders','Displacement','Horsepower','Weight',
'Acceleration', 'Model Year', 'Origin']  # 6个属性字段
raw_dataset = pd.read_csv(dataset_path, names=column_names,
                         na_values="?",comment='\t',sep="  ",skipinitial=True)
dataset = raw_dataset.copy()

dataset.head()

dataset.isna().sum() # 统计空白数据
dataset = dataset.dropna()  #删除空白数据
dataset.isna().sum()   # 再次统计空白数据

# 将Origin的3个产地分别用1，2，3代替

# 弹出并且返回Origin这列数据
origin = dataset.pop('Origin')
dataset['USA'] = (origin == 1) * 1.0
dataset['Europe'] = (origin == 2) * 1.0
dataset['Japan'] = (origin == 3) * 1.0

dataset.tail()

# 8:2划分数据集
train_dataset = dataset.sample(frac=0.8, random_state=0)
test_dataset = dataset.drop(train_dataset.index)

# 移动MPG油耗效能这列数据为真实标签y
train_lables = train_dataset.pop('MPG')
test_lables = test_dataset.pop('MPG')

# 查看训练集X的数据
train_stats = train_dataset.describe()
train_status.pop("MPG")
train_status = train_stats.transpose()

# 标准化数据
def norm(x):
  return (x -  train.stats['mean']) / (train_stats['std'])

normed_train_data = norm(train_dataset)
normed_test_data = norm(test_dataset)

# 打印训练集和测试机的大小
print(normed_train_data.shape, train_labels.shape)
print(normed_test_data.shape, test_labels.shape)

# 利用切分的训练集数据构建数据集对象
train_db = tf.data.Dataset.from_tensor_slices((normed_train_data.values, train_labels.values))    # 构建Dataset对象
train_db = train_db_shuffle(100).batch(32)  # 随机散打，批量化过程


# 2.创建网络
class Network(keras.Model):
  # 回归网络
  def __init__(self):
    super(Network, self).__init__()
    # 创建3个全连接层
    self.fc1 = layers.Dense(64, activation="relu")
    self.fc2 = layers.Dense(64, activation="relu")
    self.fc3 = layers.Dense(1)
    
  def call(self, input, training=None, mask=None):
    # 依次通过3个连接层
    x = self.fc1(inputs)
    x = self.fc2(x)
    x = self.fc3(x)
    
    return x
  
# 3. 训练与测试：实例化网络对象和创建优化器
model = Network()
model.build(input_shape=(4,9))
model.summary()
optimizer = tf.keras.optimizers.RMSprop(0.001)

for epoch in range(200):
  for step, (x,y) in enumerate(train_db):
    with tf.GradientTape() as tape: # 梯度记录器
      out = model(x)  # 通过网络获得输出
      loss = tf.reduce_mean(losses.MSE(y, out))  # 计算MSE
      mae_loss = tf.reduce_mean(losses.MAE(y, out))  # 计算MAE
      
    if step % 10 == 0:
      print(epoch, step, float(loss))
    #  更新梯度信息
    grads = tape.gradient(loss, model.train_varciables)
    optimizer.apply_gradients(zip(grads, model.train_variables))

image