前期准备
TensorFlow 相关 API 可以到在实验 TensorFlow - 相关 API 中学习。
模型构建
示例代码：
现在您可以在 /home/ubuntu 目录下创建源文件 linear_regression_model.py，内容可参考：
示例代码：/home/ubuntu/linear_regression_model.py
linear_regression_model.py
#!/usr/bin/python
import numpy as np
class linearRegressionModel:
def __init__(self,x_dimen):
self.x_dimen = x_dimen
self._index_in_epoch = 0
self.constructModel()#呵呵
self.sess = tf.Session()#这个
self.sess.run(tf.global_variables_initializer())#和这个
#权重初始化
def weight_variable(self,shape):
initial = tf.truncated_normal(shape,stddev = 0.1)#难道mean默认了
return tf.Variable(initial)#把初始化好的值装到变量里面
#偏置项初始化
def bias_variable(self,shape):
initial = tf.constant(0.1,shape=shape)#这丫的shape是用shape=传递的
return tf.Variable(initial)
#每次选取100个样本，如果选完，重新打乱
def next_batch(self,batch_size):#目前定义的都是函数，还没有开始使用
start = self._index_in_epoch
self._index_in_epoch +=batch_size#i +=1 => i = i + 1,假设i=1，那就是2了
if self._index_in_epoch > self._num_datas:#判断是否已经训练完了，即训练的次数大于训练集总的数据量
perm = np.arange(self._num_datas)#self._num_datas指训练集的样本数量
np.random.shuffle(perm)#打乱arange的值，接下来就是类似通过index/id这些标记来获取训练样本，那么获取的训练样本也是被打乱的了
self._datas = self._datas[perm]#所有训练集输入数据被重置
self._labels = self._labels[perm]#所有训练集输出数据被重置
start = 0
self._index_in_epoch = batch_size#batch_size是固定值吧，每次把它搞成固定值，不应该呀
assert batch_size <= self._num_datas#这个asset函数啥意思？大概是如果batch_size <= self._num_datas为假就发出警报吧
end = self._index_in_epoch
return self._datas[start:end],self._labels[start:end]
def constructModel(self):
self.x = tf.placeholder(tf.float32, [None,self.x_dimen])
self.y = tf.placeholder(tf.float32,[None,1])
self.w = self.weight_variable([self.x_dimen,1])
self.b = self.bias_variable([1])
self.y_prec = tf.nn.bias_add(tf.matmul(self.x, self.w), self.b)
mse = tf.reduce_mean(tf.squared_difference(self.y_prec, self.y))
l2 = tf.reduce_mean(tf.square(self.w))
self.loss = mse + 0.15*l2
self.train_step = tf.train.AdamOptimizer(0.1).minimize(self.loss)
def train(self,x_train,y_train,x_test,y_test):
self._datas = x_train
self._labels = y_train
self._num_datas = x_train.shape[0]
for i in range(5000):
batch = self.next_batch(100)
self.sess.run(self.train_step,feed_dict={self.x:batch[0],self.y:batch[1]})
if i%10 == 0:
train_loss = self.sess.run(self.loss,feed_dict={self.x:batch[0],self.y:batch[1]})
print('step %d,test_loss %f' % (i,train_loss))
def predict_batch(self,arr,batch_size):
for i in range(0,len(arr),batch_size):
yield arr[i:i + batch_size]
def predict(self, x_predict):
pred_list = []
for x_test_batch in self.predict_batch(x_predict,100):
pred = self.sess.run(self.y_prec, {self.x:x_test_batch})
pred_list.append(pred)
return np.vstack(pred_list)
训练模型并和 sklearn 库线性回归模型对比
示例代码：
现在您可以在 /home/ubuntu 目录下创建源文件 run.py，内容可参考：
示例代码：/home/ubuntu/run.py
run.py
#!/usr/bin/python
# -*- coding: utf-8 -*
from sklearn.model_selection import train_test_split
from sklearn.metrics import r2_score
from sklearn.datasets import make_regression
from sklearn.linear_model import LinearRegression
from linear_regression_model import linearRegressionModel as lrm
if __name__ == '__main__':
x, y = make_regression(7000)
x_train,x_test,y_train, y_test = train_test_split(x, y, test_size=0.5)
y_lrm_train = y_train.reshape(-1, 1)
y_lrm_test = y_test.reshape(-1, 1)
linear = lrm(x.shape[1])
linear.train(x_train, y_lrm_train,x_test,y_lrm_test)
y_predict = linear.predict(x_test)
print("Tensorflow R2: ", r2_score(y_predict.ravel(), y_lrm_test.ravel()))
lr = LinearRegression()
y_predict = lr.fit(x_train, y_train).predict(x_test)
print("Sklearn R2: ", r2_score(y_predict, y_test)) #采用r2_score评分函数