本系列专门用来记录我的深度学习历程,其中代码大部分均出自于李金洪老师的《深度学习之TensorFlow》,希望所有机器学习的道友都能有所知、有所得。
本章主要内容是保存/载入线性回归模型
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
plotdata = {"batchsize": [], "loss": []}
def moving_average(a, w=10):
if len(a) < w:
return a[:]
return [val if idx < w else sum(a[(idx - w):idx]) / w for idx, val in enumerate(a)]
1.准备数据
train_X = np.linspace(-1, 1, 100)
train_Y = 2 * train_X + np.random.randn(*train_X.shape) * 0.3 # 加入噪声
# 显示模拟数据点
plt.plot(train_X, train_Y, 'ro', label='Original data')
plt.legend() # 显示图例
plt.show() # 显示图
##重置图
tf.reset_default_graph()
2.创建模型
2.1正向搭建模型
X = tf.placeholder("float")# 占位符
Y = tf.placeholder("float")
# 模型参数
W = tf.Variable(tf.random_normal([1]), name="weight") # W被初始化成[-1,1]的随机数,形状为一维
b = tf.Variable(tf.zeros([1]), name="bias") # b被初始化为0,形状也为一维
# 前向结构
z = tf.multiply(X, W) + b
2.2反向搭建模型
cost = tf.reduce_mean(tf.square(Y - z))
learning_rate = 0.01
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost) # 梯度下降
3.迭代训练模型
#初始化所有变量
init = tf.global_variables_initializer()
# 定义参数
training_epochs = 20
display_step = 2
##生成saver及模型的路径
saver = tf.train.Saver()
savedir = "log/"
# 启动session
# with tf.Session() as sess:
# sess.run(init)
# ##在这里添加sess中的训练代码
# plotdata={"batchsize": [], "loss": []}
# # 向模型输入数据
# for epoch in range(training_epochs):
# for (x, y) in zip(train_X, train_Y):
# sess.run(optimizer, feed_dict={X: x, Y: y})
#
# # 显示训练中的详细信息
# if epoch % display_step == 0:
# loss = sess.run(cost, feed_dict={X: train_X, Y: train_Y})
# print("EPOCH:", epoch + 1, "cost=", loss, "W=", sess.run(W), "b=", sess.run(b))
# if not (loss == "NA"):
# plotdata["batchsize"].append(epoch)
# plotdata["loss"].append(loss)
#
# print("Finished!")
# saver.save(sess,savedir+"linermodel.cpkt")
# print("cost=", sess.run(cost, feed_dict={X: train_X, Y: train_Y}), 'W=', sess.run(W), "b=", sess.run(b))
4.训练模型可视化
# plt.plot(train_X, train_Y, 'ro', label='Original data')
# plt.plot(train_X, sess.run(W) * train_X + sess.run(b), label='Fittedline')
# plt.legend()
# plt.show()
#
# plotdata["avgloss"] = moving_average(plotdata["loss"])
# plt.figure(1)
# plt.subplot(211)
# plt.plot(plotdata["batchsize"], plotdata["avgloss"], 'b--')
# plt.xlabel('Minibatch number')
# plt.ylabel('Loss')
# plt.title('Minibatch run vs. Training loss')
#
# plt.show()
5.使用模型
# print("x=0.2,z=",sess.run(z,feed_dict={X:0.2}))
6.使用保存的模型
with tf.Session() as sess2:
sess2.run(tf.global_variables_initializer())
saver.restore(sess2,savedir+"linermodel.cpkt")
print("x=0.2,z=",sess2.run(z,feed_dict={X:0.2}))