DL6-多元线性回归：波士顿房价预测问题TensorFlow(1.x)实战
1、数据读取
# 导入相关库
%matplotlib notebook
# import tensorflow as tf V2.0用这个
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd # 数据读取
from sklearn.utils import shuffle #样本洗牌
import tensorflow.compat.v1 as tf
tf.disable_eager_execution()
tf.reset_default_graph()
#读取数据文件
df = pd.read_csv("data/boston.csv",header = 0)
#显示数据摘要描述信息
print(df.describe())
CRIM ZN INDUS CHAS NOX RM \
count 506.000000 506.000000 506.000000 506.000000 506.000000 506.000000
mean 3.613524 11.363636 11.136779 0.069170 0.554695 6.284634
std 8.601545 23.322453 6.860353 0.253994 0.115878 0.702617
min 0.006320 0.000000 0.460000 0.000000 0.385000 3.561000
25% 0.082045 0.000000 5.190000 0.000000 0.449000 5.885500
50% 0.256510 0.000000 9.690000 0.000000 0.538000 6.208500
75% 3.677082 12.500000 18.100000 0.000000 0.624000 6.623500
max 88.976200 100.000000 27.740000 1.000000 0.871000 8.780000
AGE DIS RAD TAX PTRATIO LSTAT \
count 506.000000 506.000000 506.000000 506.000000 506.000000 506.000000
mean 68.574901 3.795043 9.549407 408.237154 18.455534 12.653063
std 28.148861 2.105710 8.707259 168.537116 2.164946 7.141062
min 2.900000 1.129600 1.000000 187.000000 12.600000 1.730000
25% 45.025000 2.100175 4.000000 279.000000 17.400000 6.950000
50% 77.500000 3.207450 5.000000 330.000000 19.050000 11.360000
75% 94.075000 5.188425 24.000000 666.000000 20.200000 16.955000
max 100.000000 12.126500 24.000000 711.000000 22.000000 37.970000
MEDV
count 506.000000
mean 22.532806
std 9.197104
min 5.000000
25% 17.025000
50% 21.200000
75% 25.000000
max 50.000000
# 显示所有数据
print(df)
CRIM ZN INDUS CHAS NOX RM AGE DIS RAD TAX \
0 0.00632 18.0 2.31 0 0.538 6.575 65.2 4.0900 1 296
1 0.02731 0.0 7.07 0 0.469 6.421 78.9 4.9671 2 242
2 0.02729 0.0 7.07 0 0.469 7.185 61.1 4.9671 2 242
3 0.03237 0.0 2.18 0 0.458 6.998 45.8 6.0622 3 222
4 0.06905 0.0 2.18 0 0.458 7.147 54.2 6.0622 3 222
.. ... ... ... ... ... ... ... ... ... ...
501 0.06263 0.0 11.93 0 0.573 6.593 69.1 2.4786 1 273
502 0.04527 0.0 11.93 0 0.573 6.120 76.7 2.2875 1 273
503 0.06076 0.0 11.93 0 0.573 6.976 91.0 2.1675 1 273
504 0.10959 0.0 11.93 0 0.573 6.794 89.3 2.3889 1 273
505 0.04741 0.0 11.93 0 0.573 6.030 80.8 2.5050 1 273
PTRATIO LSTAT MEDV
0 15.3 4.98 24.0
1 17.8 9.14 21.6
2 17.8 4.03 34.7
3 18.7 2.94 33.4
4 18.7 5.33 36.2
.. ... ... ...
501 21.0 9.67 22.4
502 21.0 9.08 20.6
503 21.0 5.64 23.9
504 21.0 6.48 22.0
505 21.0 7.88 11.9
[506 rows x 13 columns]
2、准备建模
（1）准备工作–线性代数
import numpy as np
# 标量只是一个单一的数字
scalar_value = 18
print(scalar_value)
18
# 把标量转换为数组
sclar_np = np.array(scalar_value)
# 数组的shape不为空
print(sclar_np,sclar_np.shape)
18 ()
# 向量是一个有序的数字数组
# 这是一个列表
vector_value = [1,2,3]
# 转换为np中的数组array
vector_np = np.array(vector_value)
# shape显示为 一维数组， 其实这既不能算行向量，也不能算列向量：即不能区分（无法表示）行向量、列向量
print(vector_np, vector_np.shape)
[1 2 3] (3,)
# 矩阵是一个有序二维数组，它有两个索引。第一个指向该行，第二个指向该列
matrix_list = [[1,2,3],[4,5,6]]
matrix_np = np.array(matrix_list)
print("matrix_list = ",matrix_list,\
"\nmatrix_np =\n", matrix_np, \
"\nmatrix_np.shape = ", matrix_np.shape)
matrix_list = [[1, 2, 3], [4, 5, 6]]
matrix_np =
[[1 2 3]
[4 5 6]]
matrix_np.shape = (2, 3)
# 行向量矩阵表示
vector_row = np.array([[1,2,3]])
print(vector_row, "shape = ", vector_row.shape)
# 1行3列
[[1 2 3]] shape = (1, 3)
# 列向量矩阵表示
vector_column = np.array([[4],[5],[6]])
print(vector_column, "shape = ", vector_column.shape)
# 3行1列
[[4]
[5]
[6]] shape = (3, 1)
# 矩阵与标量运算
matrix_a = np.array([[1,2,3],[4,5,6]])
print(matrix_a,"\nshape = ", matrix_a.shape)
[[1 2 3]
[4 5 6]]
shape = (2, 3)
# 矩阵 * 标量
matrix_b = matrix_a * 2
print(matrix_b,"s\nhape = ", matrix_b.shape)
[[ 2 4 6]
[ 8 10 12]] s
hape = (2, 3)
# 矩阵 + 标量
matrix_c =matrix_a + 2
print(matrix_c,"\nshape = ", matrix_c.shape)
[[3 4 5]
[6 7 8]]
shape = (2, 3)
# 矩阵 ± 矩阵
# 要求矩阵的形状相同：对应位置相加减
matrix_a = np.array([[1,2,3],
[4,5,6]])
matrix_b = np.array([[-1,-2,-3],
[-4,-5,-6]])
print(matrix_a + matrix_b)
print(matrix_a - matrix_b)
[[0 0 0]
[0 0 0]]
[[ 2 4 6]
[ 8 10 12]]
# 矩阵 * 矩阵（点乘）
# 要求矩阵的形状相同：对应位置相乘
matrix_a = np.array([[1,2,3],
[4,5,6]])
matrix_b = np.array([[-1,-2,-3],
[-4,-5,-6]])
print(matrix_a * matrix_b)
#等价表述如下：
print(np.multiply(matrix_a,matrix_b))
[[ -1 -4 -9]
[-16 -25 -36]]
[[ -1 -4 -9]
[-16 -25 -36]]
# 矩阵 ※ 矩阵（叉乘）
# 要求第一个矩阵的列=第二个矩阵的行
matrix_a = np.array([[1,2,3],
[4,5,6]])#2行3列
matrix_b = np.array([[1,2,3,4],
[1,2,1,2],
[-1,0,1,1]])#3行4列
print(np.matmul(matrix_a,matrix_b))#输出为2行4列
[[ 0 6 8 11]
[ 3 18 23 32]]
# 矩阵转置
matrix_a = np.array([[1,2,3],
[4,5,6]])#2行3列
print(matrix_a,"\nshape = ", matrix_a.shape,'\n',matrix_a.T,'\n.Tshape=',matrix_a.T.shape)
[[1 2 3]
[4 5 6]]
shape = (2, 3)
[[1 4]
[2 5]
[3 6]]
.Tshape= (3, 2)
# 也可以通过reshape函数
matrix_b = np.array([[1,2,3],
[4,5,6]])#2行3列
matrix_c = matrix_b.reshape(6,1)
print(matrix_b,"\nshape = ", matrix_b.shape,'\n',matrix_c,'\n.Tshape=',matrix_c.shape)
[[1 2 3]
[4 5 6]]
shape = (2, 3)
[[1]
[2]
[3]
[4]
[5]
[6]]
.Tshape= (6, 1)
matrix_c = matrix_b.reshape(1,6)
print(matrix_b,"\nshape = ", matrix_b.shape,'\n',matrix_c,'\n.Tshape=',matrix_c.shape)
[[1 2 3]
[4 5 6]]
shape = (2, 3)
[[1 2 3 4 5 6]]
.Tshape= (1, 6)
（2）数据准备
# 获取df的值
df = df.values #只获取含数字的项，表头是字母不会包含进来
# 把 df 转换为 np 数组格式
df = np.array(df)
print(df)
[[6.3200e-03 1.8000e+01 2.3100e+00 ... 1.5300e+01 4.9800e+00 2.4000e+01]
[2.7310e-02 0.0000e+00 7.0700e+00 ... 1.7800e+01 9.1400e+00 2.1600e+01]
[2.7290e-02 0.0000e+00 7.0700e+00 ... 1.7800e+01 4.0300e+00 3.4700e+01]
...
[6.0760e-02 0.0000e+00 1.1930e+01 ... 2.1000e+01 5.6400e+00 2.3900e+01]
[1.0959e-01 0.0000e+00 1.1930e+01 ... 2.1000e+01 6.4800e+00 2.2000e+01]
[4.7410e-02 0.0000e+00 1.1930e+01 ... 2.1000e+01 7.8800e+00 1.1900e+01]]
# x_data 为前12列特征数据
x_data = df[:,:12] #行数：所有；列数：第0至11行，不包含第12行
# y_data 为最后1列标签数据
y_data = df[:,12]
print(x_data,"\n shape=",x_data.shape)
[[6.3200e-03 1.8000e+01 2.3100e+00 ... 2.9600e+02 1.5300e+01 4.9800e+00]
[2.7310e-02 0.0000e+00 7.0700e+00 ... 2.4200e+02 1.7800e+01 9.1400e+00]
[2.7290e-02 0.0000e+00 7.0700e+00 ... 2.4200e+02 1.7800e+01 4.0300e+00]
...
[6.0760e-02 0.0000e+00 1.1930e+01 ... 2.7300e+02 2.1000e+01 5.6400e+00]
[1.0959e-01 0.0000e+00 1.1930e+01 ... 2.7300e+02 2.1000e+01 6.4800e+00]
[4.7410e-02 0.0000e+00 1.1930e+01 ... 2.7300e+02 2.1000e+01 7.8800e+00]]
shape= (506, 12)
print(y_data,"\n shape=",y_data.shape)
[24. 21.6 34.7 33.4 36.2 28.7 22.9 27.1 16.5 18.9 15. 18.9 21.7 20.4
18.2 19.9 23.1 17.5 20.2 18.2 13.6 19.6 15.2 14.5 15.6 13.9 16.6 14.8
18.4 21. 12.7 14.5 13.2 13.1 13.5 18.9 20. 21. 24.7 30.8 34.9 26.6
25.3 24.7 21.2 19.3 20. 16.6 14.4 19.4 19.7 20.5 25. 23.4 18.9 35.4
24.7 31.6 23.3 19.6 18.7 16. 22.2 25. 33. 23.5 19.4 22. 17.4 20.9
24.2 21.7 22.8 23.4 24.1 21.4 20. 20.8 21.2 20.3 28. 23.9 24.8 22.9
23.9 26.6 22.5 22.2 23.6 28.7 22.6 22. 22.9 25. 20.6 28.4 21.4 38.7
43.8 33.2 27.5 26.5 18.6 19.3 20.1 19.5 19.5 20.4 19.8 19.4 21.7 22.8
18.8 18.7 18.5 18.3 21.2 19.2 20.4 19.3 22. 20.3 20.5 17.3 18.8 21.4
15.7 16.2 18. 14.3 19.2 19.6 23. 18.4 15.6 18.1 17.4 17.1 13.3 17.8
14. 14.4 13.4 15.6 11.8 13.8 15.6 14.6 17.8 15.4 21.5 19.6 15.3 19.4
17. 15.6 13.1 41.3 24.3 23.3 27. 50. 50. 50. 22.7 25. 50. 23.8
23.8 22.3 17.4 19.1 23.1 23.6 22.6 29.4 23.2 24.6 29.9 37.2 39.8 36.2
37.9 32.5 26.4 29.6 50. 32. 29.8 34.9 37. 30.5 36.4 31.1 29.1 50.
33.3 30.3 34.6 34.9 32.9 24.1 42.3 48.5 50. 22.6 24.4 22.5 24.4 20.
21.7 19.3 22.4 28.1 23.7 25. 23.3 28.7 21.5 23. 26.7 21.7 27.5 30.1
44.8 50. 37.6 31.6 46.7 31.5 24.3 31.7 41.7 48.3 29. 24. 25.1 31.5
23.7 23.3 22. 20.1 22.2 23.7 17.6 18.5 24.3 20.5 24.5 26.2 24.4 24.8
29.6 42.8 21.9 20.9 44. 50. 36. 30.1 33.8 43.1 48.8 31. 36.5 22.8
30.7 50. 43.5 20.7 21.1 25.2 24.4 35.2 32.4 32. 33.2 33.1 29.1 35.1
45.4 35.4 46. 50. 32.2 22. 20.1 23.2 22.3 24.8 28.5 37.3 27.9 23.9
21.7 28.6 27.1 20.3 22.5 29. 24.8 22. 26.4 33.1 36.1 28.4 33.4 28.2
22.8 20.3 16.1 22.1 19.4 21.6 23.8 16.2 17.8 19.8 23.1 21. 23.8 23.1
20.4 18.5 25. 24.6 23. 22.2 19.3 22.6 19.8 17.1 19.4 22.2 20.7 21.1
19.5 18.5 20.6 19. 18.7 32.7 16.5 23.9 31.2 17.5 17.2 23.1 24.5 26.6
22.9 24.1 18.6 30.1 18.2 20.6 17.8 21.7 22.7 22.6 25. 19.9 20.8 16.8
21.9 27.5 21.9 23.1 50. 50. 50. 50. 50. 13.8 13.8 15. 13.9 13.3
13.1 10.2 10.4 10.9 11.3 12.3 8.8 7.2 10.5 7.4 10.2 11.5 15.1 23.2
9.7 13.8 12.7 13.1 12.5 8.5 5. 6.3 5.6 7.2 12.1 8.3 8.5 5.
11.9 27.9 17.2 27.5 15. 17.2 17.9 16.3 7. 7.2 7.5 10.4 8.8 8.4
16.7 14.2 20.8 13.4 11.7 8.3 10.2 10.9 11. 9.5 14.5 14.1 16.1 14.3
11.7 13.4 9.6 8.7 8.4 12.8 10.5 17.1 18.4 15.4 10.8 11.8 14.9 12.6
14.1 13. 13.4 15.2 16.1 17.8 14.9 14.1 12.7 13.5 14.9 20. 16.4 17.7
19.5 20.2 21.4 19.9 19. 19.1 19.1 20.1 19.9 19.6 23.2 29.8 13.8 13.3
16.7 12. 14.6 21.4 23. 23.7 25. 21.8 20.6 21.2 19.1 20.6 15.2 7.
8.1 13.6 20.1 21.8 24.5 23.1 19.7 18.3 21.2 17.5 16.8 22.4 20.6 23.9
22. 11.9]
shape= (506,)
（3）模型定义
# 定义 特征数据和标签数据的占位符
x = tf.placeholder(tf.float32,[None,12],name = "X")
y = tf.placeholder(tf.float32,[None,1],name = "Y")
# shape中 None 表示行的数量未知，在实际训练时决定一次代入多少行样本，从一个样本的随机SDG到批量SDG都可以
# 定义模型函数
with tf.name_scope("Model"):
# w 初始值为shape = (12,1)的随机数
w = tf.Variable(tf.random_normal([12,1],stddev=0.01),name = "W")
# b 初始化值为1.0
b = tf.Variable(1.0,name = "b")
# w和x矩阵相乘，用matmul
def model(x,w,b):
return tf.matmul(x,w) + b
# 预测计算操作，前向计算节点
pred = model(x,w,b)
（4）模型训练
# 设置训练参数（超参数）
# 迭代轮次
train_epochs = 50
# 学习率
learning_rate = 0.01
# 定义均方差损失函数
with tf.name_scope("LossFunction"):
loss_function = tf.reduce_mean(tf.pow(y - pred,2))
# 创建梯度下降优化器
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss_function)
# 声明会话
sess = tf.Session()
init = tf.global_variables_initializer()
# 启动会话
sess.run(init)
# 迭代训练
for epoch in range(train_epochs):
loss_sum = 0
for xs,ys in zip(x_data,y_data):
# 数据变形，使其与spaceholder形状一致
xs = xs.reshape(1,12)
ys = ys.reshape(1,1)
_,loss = sess.run([optimizer,loss_function],feed_dict={x:xs,y:ys})
#将xs与ys利用占位符喂给x和y，然后得到梯度优化后的损失函数，此时w和b的值存在_里，但是我们并不需要知道具体的值，故用一个符号简单代替即可
loss_sum = loss_sum + loss
# 打乱数据顺序，防止记住顺序
xvalues, yvalues = shuffle(x_data,y_data)
b0temp = b.eval(session=sess)
w0temp = w.eval(session=sess)
loss_average = loss_sum / len(y_data)
print("epoch=",epoch + 1,"loss = ",loss_average,"b = ",b0temp,"w = ",w0temp)
epoch= 50 loss = nan b = nan w = [[nan]
[nan]
[nan]
[nan]
[nan]
[nan]
[nan]
[nan]
[nan]
[nan]
[nan]
[nan]]
出现问题：单变量与多变量的梯度下降有别！
归一化：特征值/（特征值最大值-特征值最小值）
3、版本2：进行归一化处理
# 导入相关库
%matplotlib notebook
# import tensorflow as tf V2.0用这个
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd # 数据读取
from sklearn.utils import shuffle #样本洗牌
import tensorflow.compat.v1 as tf
tf.disable_eager_execution()
tf.reset_default_graph()
#读取数据文件
df = pd.read_csv("data/boston.csv",header = 0)
#显示数据摘要描述信息
print(df.describe())
CRIM ZN INDUS CHAS NOX RM \
count 506.000000 506.000000 506.000000 506.000000 506.000000 506.000000
mean 3.613524 11.363636 11.136779 0.069170 0.554695 6.284634
std 8.601545 23.322453 6.860353 0.253994 0.115878 0.702617
min 0.006320 0.000000 0.460000 0.000000 0.385000 3.561000
25% 0.082045 0.000000 5.190000 0.000000 0.449000 5.885500
50% 0.256510 0.000000 9.690000 0.000000 0.538000 6.208500
75% 3.677082 12.500000 18.100000 0.000000 0.624000 6.623500
max 88.976200 100.000000 27.740000 1.000000 0.871000 8.780000
AGE DIS RAD TAX PTRATIO LSTAT \
count 506.000000 506.000000 506.000000 506.000000 506.000000 506.000000
mean 68.574901 3.795043 9.549407 408.237154 18.455534 12.653063
std 28.148861 2.105710 8.707259 168.537116 2.164946 7.141062
min 2.900000 1.129600 1.000000 187.000000 12.600000 1.730000
25% 45.025000 2.100175 4.000000 279.000000 17.400000 6.950000
50% 77.500000 3.207450 5.000000 330.000000 19.050000 11.360000
75% 94.075000 5.188425 24.000000 666.000000 20.200000 16.955000
max 100.000000 12.126500 24.000000 711.000000 22.000000 37.970000
MEDV
count 506.000000
mean 22.532806
std 9.197104
min 5.000000
25% 17.025000
50% 21.200000
75% 25.000000
max 50.000000
# 获取df的值
df = df.values #只获取含数字的项，表头是字母不会包含进来
# 把 df 转换为 np 数组格式
df = np.array(df)
print(df)
[[6.3200e-03 1.8000e+01 2.3100e+00 ... 1.5300e+01 4.9800e+00 2.4000e+01]
[2.7310e-02 0.0000e+00 7.0700e+00 ... 1.7800e+01 9.1400e+00 2.1600e+01]
[2.7290e-02 0.0000e+00 7.0700e+00 ... 1.7800e+01 4.0300e+00 3.4700e+01]
...
[6.0760e-02 0.0000e+00 1.1930e+01 ... 2.1000e+01 5.6400e+00 2.3900e+01]
[1.0959e-01 0.0000e+00 1.1930e+01 ... 2.1000e+01 6.4800e+00 2.2000e+01]
[4.7410e-02 0.0000e+00 1.1930e+01 ... 2.1000e+01 7.8800e+00 1.1900e+01]]
特征数据归一化
# 对特征数据【0到11】列做（0-1）归一化
# 对第i列的所有数据进行归一化
for i in range(12):
df[:,i] = df[:,i] / (df[:,i].max() - df[:,i].min())
# x_data 为归一化后的前12列特征数据
x_data = df[:,:12]
# y_data 为最后一列标签数据：无需归一化
y_data = df[:,12]
print(x_data,"\n shape=",x_data.shape)
print(y_data,"\n shape=",y_data.shape)
[[7.10352762e-05 1.80000000e-01 8.46774194e-02 ... 5.64885496e-01
1.62765957e+00 1.37417219e-01]
[3.06957815e-04 0.00000000e+00 2.59164223e-01 ... 4.61832061e-01
1.89361702e+00 2.52207506e-01]
[3.06733020e-04 0.00000000e+00 2.59164223e-01 ... 4.61832061e-01
1.89361702e+00 1.11203091e-01]
...
[6.82927750e-04 0.00000000e+00 4.37316716e-01 ... 5.20992366e-01
2.23404255e+00 1.55629139e-01]
[1.23176518e-03 0.00000000e+00 4.37316716e-01 ... 5.20992366e-01
2.23404255e+00 1.78807947e-01]
[5.32876969e-04 0.00000000e+00 4.37316716e-01 ... 5.20992366e-01
2.23404255e+00 2.17439294e-01]]
shape= (506, 12)
[24. 21.6 34.7 33.4 36.2 28.7 22.9 27.1 16.5 18.9 15. 18.9 21.7 20.4
18.2 19.9 23.1 17.5 20.2 18.2 13.6 19.6 15.2 14.5 15.6 13.9 16.6 14.8
18.4 21. 12.7 14.5 13.2 13.1 13.5 18.9 20. 21. 24.7 30.8 34.9 26.6
25.3 24.7 21.2 19.3 20. 16.6 14.4 19.4 19.7 20.5 25. 23.4 18.9 35.4
24.7 31.6 23.3 19.6 18.7 16. 22.2 25. 33. 23.5 19.4 22. 17.4 20.9
24.2 21.7 22.8 23.4 24.1 21.4 20. 20.8 21.2 20.3 28. 23.9 24.8 22.9
23.9 26.6 22.5 22.2 23.6 28.7 22.6 22. 22.9 25. 20.6 28.4 21.4 38.7
43.8 33.2 27.5 26.5 18.6 19.3 20.1 19.5 19.5 20.4 19.8 19.4 21.7 22.8
18.8 18.7 18.5 18.3 21.2 19.2 20.4 19.3 22. 20.3 20.5 17.3 18.8 21.4
15.7 16.2 18. 14.3 19.2 19.6 23. 18.4 15.6 18.1 17.4 17.1 13.3 17.8
14. 14.4 13.4 15.6 11.8 13.8 15.6 14.6 17.8 15.4 21.5 19.6 15.3 19.4
17. 15.6 13.1 41.3 24.3 23.3 27. 50. 50. 50. 22.7 25. 50. 23.8
23.8 22.3 17.4 19.1 23.1 23.6 22.6 29.4 23.2 24.6 29.9 37.2 39.8 36.2
37.9 32.5 26.4 29.6 50. 32. 29.8 34.9 37. 30.5 36.4 31.1 29.1 50.
33.3 30.3 34.6 34.9 32.9 24.1 42.3 48.5 50. 22.6 24.4 22.5 24.4 20.
21.7 19.3 22.4 28.1 23.7 25. 23.3 28.7 21.5 23. 26.7 21.7 27.5 30.1
44.8 50. 37.6 31.6 46.7 31.5 24.3 31.7 41.7 48.3 29. 24. 25.1 31.5
23.7 23.3 22. 20.1 22.2 23.7 17.6 18.5 24.3 20.5 24.5 26.2 24.4 24.8
29.6 42.8 21.9 20.9 44. 50. 36. 30.1 33.8 43.1 48.8 31. 36.5 22.8
30.7 50. 43.5 20.7 21.1 25.2 24.4 35.2 32.4 32. 33.2 33.1 29.1 35.1
45.4 35.4 46. 50. 32.2 22. 20.1 23.2 22.3 24.8 28.5 37.3 27.9 23.9
21.7 28.6 27.1 20.3 22.5 29. 24.8 22. 26.4 33.1 36.1 28.4 33.4 28.2
22.8 20.3 16.1 22.1 19.4 21.6 23.8 16.2 17.8 19.8 23.1 21. 23.8 23.1
20.4 18.5 25. 24.6 23. 22.2 19.3 22.6 19.8 17.1 19.4 22.2 20.7 21.1
19.5 18.5 20.6 19. 18.7 32.7 16.5 23.9 31.2 17.5 17.2 23.1 24.5 26.6
22.9 24.1 18.6 30.1 18.2 20.6 17.8 21.7 22.7 22.6 25. 19.9 20.8 16.8
21.9 27.5 21.9 23.1 50. 50. 50. 50. 50. 13.8 13.8 15. 13.9 13.3
13.1 10.2 10.4 10.9 11.3 12.3 8.8 7.2 10.5 7.4 10.2 11.5 15.1 23.2
9.7 13.8 12.7 13.1 12.5 8.5 5. 6.3 5.6 7.2 12.1 8.3 8.5 5.
11.9 27.9 17.2 27.5 15. 17.2 17.9 16.3 7. 7.2 7.5 10.4 8.8 8.4
16.7 14.2 20.8 13.4 11.7 8.3 10.2 10.9 11. 9.5 14.5 14.1 16.1 14.3
11.7 13.4 9.6 8.7 8.4 12.8 10.5 17.1 18.4 15.4 10.8 11.8 14.9 12.6
14.1 13. 13.4 15.2 16.1 17.8 14.9 14.1 12.7 13.5 14.9 20. 16.4 17.7
19.5 20.2 21.4 19.9 19. 19.1 19.1 20.1 19.9 19.6 23.2 29.8 13.8 13.3
16.7 12. 14.6 21.4 23. 23.7 25. 21.8 20.6 21.2 19.1 20.6 15.2 7.
8.1 13.6 20.1 21.8 24.5 23.1 19.7 18.3 21.2 17.5 16.8 22.4 20.6 23.9
22. 11.9]
shape= (506,)
# 定义 特征数据和标签数据的占位符
x = tf.placeholder(tf.float32,[None,12],name = "X")
y = tf.placeholder(tf.float32,[None,1],name = "Y")
# 定义模型函数
with tf.name_scope("Model"):
# w 初始值为shape = (12,1)的随机数
w = tf.Variable(tf.random_normal([12,1],stddev=0.01),name = "W")
# b 初始化值为1.0
b = tf.Variable(1.0,name = "b")
# w和x矩阵相乘，用matmul
def model(x,w,b):
return tf.matmul(x,w) + b
# 预测计算操作，前向计算节点
pred = model(x,w,b)
# 设置训练参数（超参数）
# 迭代轮次
train_epochs = 50
# 学习率
learning_rate = 0.01
# 定义均方差损失函数
with tf.name_scope("LossFunction"):
loss_function = tf.reduce_mean(tf.pow(y - pred,2))
# 创建梯度下降优化器
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss_function)
# 声明会话
sess = tf.Session()
init = tf.global_variables_initializer()
# 启动会话
sess.run(init)
# 迭代训练
for epoch in range(train_epochs):
loss_sum = 0
for xs,ys in zip(x_data,y_data):
# 数据变形，使其与spaceholder形状一致
xs = xs.reshape(1,12)
ys = ys.reshape(1,1)
_,loss = sess.run([optimizer,loss_function],feed_dict={x:xs,y:ys})
#将xs与ys利用占位符喂给x和y，然后得到梯度优化后的损失函数，此时w和b的值存在_里，但是我们并不需要知道具体的值，故用一个符号简单代替即可
loss_sum = loss_sum + loss
# 打乱数据顺序，防止记住顺序
xvalues, yvalues = shuffle(x_data,y_data)
b0temp = b.eval(session=sess)
w0temp = w.eval(session=sess)
loss_average = loss_sum / len(y_data)
print("epoch=",epoch + 1,"loss = ",loss_average,"b = ",b0temp,"w = ",w0temp)
epoch= 1 loss = 44.337393957753434 b = 3.6102986 w = [[-0.61777246]
[ 1.3829758 ]
[-0.7958472 ]
[ 0.5048962 ]
[ 2.5224879 ]
[ 7.1773787 ]
[-0.05069519]
[ 0.8032707 ]
[ 0.3868867 ]
[ 0.33171538]
[ 2.305457 ]
[-4.3866787 ]]
epoch= 2 loss = 32.036438212210605 b = 3.9981961 w = [[-1.1673872 ]
[ 1.963905 ]
[-1.5239497 ]
[ 0.856303 ]
[ 2.896329 ]
[10.614244 ]
[-0.81843406]
[ 0.36249873]
[ 0.6352789 ]
[-0.25952753]
[ 1.1442716 ]
[-8.10359 ]]
epoch= 3 loss = 27.34001518933999 b = 4.278993 w = [[ -1.6652527 ]
[ 2.1797404 ]
[ -1.9723119 ]
[ 1.063931 ]
[ 3.1755688 ]
[ 13.210045 ]
[ -1.1952971 ]
[ -0.17106298]
[ 0.94141656]
[ -0.76077545]
[ 0.21417993]
[-10.960338 ]]
epoch= 4 loss = 24.672504632380633 b = 4.500205 w = [[ -2.1225903 ]
[ 2.2053273 ]
[ -2.2477956 ]
[ 1.1574029 ]
[ 3.3579588 ]
[ 15.199795 ]
[ -1.3577286 ]
[ -0.71559364]
[ 1.2651606 ]
[ -1.1928074 ]
[ -0.518474 ]
[-13.182013 ]]
epoch= 5 loss = 23.104833186311204 b = 4.6888976 w = [[ -2.546973 ]
[ 2.1386263]
[ -2.4139462]
[ 1.1708477]
[ 3.4531305]
[ 16.742895 ]
[ -1.4043621]
[ -1.2321597]
[ 1.5853091]
[ -1.5699434]
[ -1.0881089]
[-14.9242115]]
epoch= 6 loss = 22.160490104904515 b = 4.860485 w = [[ -2.9436767]
[ 2.0337877]
[ -2.509238 ]
[ 1.1334271]
[ 3.4749079]
[ 17.950644 ]
[ -1.3902718]
[ -1.704704 ]
[ 1.8913547]
[ -1.9023851]
[ -1.5261304]
[-16.297922 ]]
epoch= 7 loss = 21.580403804739078 b = 5.023739 w = [[ -3.3164914]
[ 1.9199821]
[ -2.5574613]
[ 1.0670242]
[ 3.437573 ]
[ 18.902704 ]
[ -1.346074 ]
[ -2.1290264]
[ 2.1787562]
[ -2.1976423]
[ -1.85953 ]
[-17.384504 ]]
epoch= 8 loss = 21.21740849930612 b = 5.183575 w = [[ -3.6682456 ]
[ 1.8121917 ]
[ -2.5736852 ]
[ 0.98677564]
[ 3.3541694 ]
[ 19.657541 ]
[ -1.2886808 ]
[ -2.5067797 ]
[ 2.4461915 ]
[ -2.4614594 ]
[ -2.1107812 ]
[-18.244934 ]]
epoch= 9 loss = 20.98522323375682 b = 5.3427134 w = [[ -4.0011134]
[ 1.7173829]
[ -2.5677123]
[ 0.9024809]
[ 3.2359138]
[ 20.258835 ]
[ -1.227302 ]
[ -2.8422096]
[ 2.6940217]
[ -2.6983678]
[ -2.2981663]
[-18.92591 ]]
epoch= 10 loss = 20.832208754840092 b = 5.5026116 w = [[ -4.316819 ]
[ 1.6380858]
[ -2.5460882]
[ 0.8200648]
[ 3.0920887]
[ 20.739676 ]
[ -1.1668166]
[ -3.140446 ]
[ 2.9234095]
[ -2.9120276]
[ -2.436327 ]
[-19.463686 ]]
epoch= 11 loss = 20.727108475022284 b = 5.663972 w = [[ -4.616779 ]
[ 1.5744346]
[ -2.5133312]
[ 0.7427979]
[ 2.9301863]
[ 21.125591 ]
[ -1.1097097]
[ -3.4066348]
[ 3.1358395]
[ -3.1054864]
[ -2.5369139]
[-19.886688 ]]
epoch= 12 loss = 20.65087399468323 b = 5.827078 w = [[ -4.9021707 ]
[ 1.5253503 ]
[ -2.472628 ]
[ 0.67222476]
[ 2.7561388 ]
[ 21.436354 ]
[ -1.0571206 ]
[ -3.6455274 ]
[ 3.3328485 ]
[ -3.2813058 ]
[ -2.609074 ]
[-20.217522 ]]
epoch= 13 loss = 20.591912843748347 b = 5.9919696 w = [[ -5.1740084]
[ 1.4892125]
[ -2.4262955]
[ 0.608814 ]
[ 2.5745807]
[ 21.687426 ]
[ -1.0094529]
[ -3.8613236]
[ 3.5158975]
[ -3.4416852]
[ -2.6599605]
[-20.474157 ]]
epoch= 14 loss = 20.543204317078576 b = 6.1585617 w = [[ -5.4331717 ]
[ 1.4642256 ]
[ -2.3760529 ]
[ 0.5524116 ]
[ 2.3890753 ]
[ 21.890976 ]
[ -0.96671915]
[ -4.057625 ]
[ 3.6863337 ]
[ -3.5885103 ]
[ -2.6951325 ]
[-20.671051 ]]
epoch= 15 loss = 20.500552776940673 b = 6.3266907 w = [[ -5.680442 ]
[ 1.448629 ]
[ -2.323208 ]
[ 0.50252676]
[ 2.2023227 ]
[ 22.056625 ]
[ -0.92872787]
[ -4.2374883 ]
[ 3.84534 ]
[ -3.7234159 ]
[ -2.7188869 ]
[-20.819813 ]]
epoch= 16 loss = 20.46148518907175 b = 6.496175 w = [[ -5.9165115 ]
[ 1.4407969 ]
[ -2.268759 ]
[ 0.45851797]
[ 2.0163589 ]
[ 22.191914 ]
[ -0.8951785 ]
[ -4.4034514 ]
[ 3.9939713 ]
[ -3.8478165 ]
[ -2.7344937 ]
[-20.92986 ]]
epoch= 17 loss = 20.424589754659305 b = 6.6668124 w = [[ -6.1420093]
[ 1.4392806]
[ -2.2134895]
[ 0.4197028]
[ 1.8326569]
[ 22.302906 ]
[ -0.865742 ]
[ -4.5576286]
[ 4.1331363]
[ -3.9629526]
[ -2.7444816]
[-21.008703 ]]
epoch= 18 loss = 20.389084714326096 b = 6.8383956 w = [[ -6.3574996 ]
[ 1.4428155 ]
[ -2.158008 ]
[ 0.38542095]
[ 1.6522747 ]
[ 22.394419 ]
[ -0.8400744 ]
[ -4.701762 ]
[ 4.2636395 ]
[ -4.069903 ]
[ -2.7507432 ]
[-21.062569 ]]
epoch= 19 loss = 20.354564370924116 b = 7.010729 w = [[ -6.5635095 ]
[ 1.4503263 ]
[ -2.102796 ]
[ 0.35506266]
[ 1.4759431 ]
[ 22.470253 ]
[ -0.81783974]
[ -4.837295 ]
[ 4.3861775 ]
[ -4.1695976 ]
[ -2.7547047 ]
[-21.096367 ]]
epoch= 20 loss = 20.32083476652019 b = 7.183645 w = [[ -6.7605214 ]
[ 1.4609047 ]
[ -2.0482311 ]
[ 0.32808915]
[ 1.3041507 ]
[ 22.533424 ]
[ -0.79871434]
[ -4.9653964 ]
[ 4.5013595 ]
[ -4.262861 ]
[ -2.7574115 ]
[-21.114166 ]]
epoch= 21 loss = 20.28783740242552 b = 7.3569617 w = [[ -6.948975 ]
[ 1.4737933 ]
[ -1.9946077 ]
[ 0.3040303 ]
[ 1.1371793 ]
[ 22.586391 ]
[ -0.78240204]
[ -5.0870357 ]
[ 4.6097302 ]
[ -4.350404 ]
[ -2.759646 ]
[-21.119238 ]]
epoch= 22 loss = 20.255557866654083 b = 7.5305247 w = [[ -7.1292915 ]
[ 1.4883684 ]
[ -1.9421594 ]
[ 0.28248435]
[ 0.9751797 ]
[ 22.631052 ]
[ -0.7686177 ]
[ -5.2030044 ]
[ 4.7117853 ]
[ -4.4328394 ]
[ -2.761948 ]
[-21.114258 ]]
epoch= 23 loss = 20.224017936810313 b = 7.704196 w = [[ -7.301845 ]
[ 1.5041173 ]
[ -1.8910623 ]
[ 0.2631091 ]
[ 0.81818926]
[ 22.668938 ]
[ -0.75710267]
[ -5.3139567 ]
[ 4.8079605 ]
[ -4.510705 ]
[ -2.7647078 ]
[-21.10137 ]]
epoch= 24 loss = 20.193252790668627 b = 7.877838 w = [[ -7.467009 ]
[ 1.5206252 ]
[ -1.8414457 ]
[ 0.24561568]
[ 0.6661698 ]
[ 22.70127 ]
[ -0.74761546]
[ -5.420425 ]
[ 4.898644 ]
[ -4.5844703 ]
[ -2.7681818 ]
[-21.082333 ]]
epoch= 25 loss = 20.163283503926376 b = 8.051342 w = [[ -7.6251125 ]
[ 1.5375584 ]
[ -1.7934122 ]
[ 0.22976038]
[ 0.51901996]
[ 22.729036 ]
[ -0.7399396 ]
[ -5.5228634 ]
[ 4.9841967 ]
[ -4.654546 ]
[ -2.7725394 ]
[-21.058584 ]]
epoch= 26 loss = 20.134130553488504 b = 8.224596 w = [[ -7.776479 ]
[ 1.5546496 ]
[ -1.7470256 ]
[ 0.21534127]
[ 0.3766105 ]
[ 22.752962 ]
[ -0.73386884]
[ -5.6216326 ]
[ 5.06495 ]
[ -4.7212744 ]
[ -2.777862 ]
[-21.031239 ]]
epoch= 27 loss = 20.105817040083704 b = 8.3974905 w = [[ -7.92141 ]
[ 1.5716902 ]
[ -1.7023295 ]
[ 0.20218605]
[ 0.23877427]
[ 22.773718 ]
[ -0.729224 ]
[ -5.717039 ]
[ 5.141205 ]
[ -4.784975 ]
[ -2.784202 ]
[-21.001255 ]]
epoch= 28 loss = 20.078329738080154 b = 8.56994 w = [[ -8.06019 ]
[ 1.5885159 ]
[ -1.659346 ]
[ 0.1901504 ]
[ 0.10533985]
[ 22.791744 ]
[ -0.7258338 ]
[ -5.809349 ]
[ 5.213251 ]
[ -4.8459077 ]
[ -2.7915494 ]
[-20.969383 ]]
epoch= 29 loss = 20.05167671847701 b = 8.741887 w = [[ -8.193069 ]
[ 1.6050023 ]
[ -1.6180803 ]
[ 0.17911133]
[ -0.02388395]
[ 22.807428 ]
[ -0.7235491 ]
[ -5.8987784 ]
[ 5.281348 ]
[ -4.9043016 ]
[ -2.7998886 ]
[-20.93623 ]]
epoch= 30 loss = 20.02585880453415 b = 8.913233 w = [[ -8.320327 ]
[ 1.6210549 ]
[ -1.5785275 ]
[ 0.16896608]
[ -0.14908668]
[ 22.82112 ]
[ -0.7222309 ]
[ -5.9855075 ]
[ 5.345736 ]
[ -4.9603696 ]
[ -2.809181 ]
[-20.902306 ]]
epoch= 31 loss = 20.00084181768566 b = 9.08393 w = [[ -8.442204 ]
[ 1.6366044 ]
[ -1.5406682 ]
[ 0.15962708]
[ -0.2704553 ]
[ 22.833027 ]
[ -0.72175264]
[ -6.0697007 ]
[ 5.4066467 ]
[ -5.0142846 ]
[ -2.8193712 ]
[-20.867983 ]]
epoch= 32 loss = 19.976618240610218 b = 9.253911 w = [[ -8.558926 ]
[ 1.651602 ]
[ -1.504473 ]
[ 0.15101798]
[ -0.38817817]
[ 22.843365 ]
[ -0.7220032 ]
[ -6.151498 ]
[ 5.464299 ]
[ -5.0662036 ]
[ -2.8304014 ]
[-20.833578 ]]
epoch= 33 loss = 19.95316042354514 b = 9.423108 w = [[ -8.670715 ]
[ 1.6660159 ]
[ -1.4699101 ]
[ 0.14307168]
[ -0.50243527]
[ 22.852308 ]
[ -0.72288346]
[ -6.2310157 ]
[ 5.518886 ]
[ -5.1162667 ]
[ -2.8422055 ]
[-20.799343 ]]
epoch= 34 loss = 19.930450114194123 b = 9.591489 w = [[ -8.777779 ]
[ 1.679826 ]
[ -1.4369352 ]
[ 0.13573325]
[ -0.6133945 ]
[ 22.85996 ]
[ -0.7242962 ]
[ -6.3083735 ]
[ 5.5705957 ]
[ -5.164591 ]
[ -2.854715 ]
[-20.765516 ]]
epoch= 35 loss = 19.90846017309892 b = 9.759006 w = [[ -8.880324 ]
[ 1.6930271 ]
[ -1.4055077 ]
[ 0.12895079]
[ -0.7212243 ]
[ 22.866415 ]
[ -0.7261652 ]
[ -6.3836584 ]
[ 5.6196127 ]
[ -5.211291 ]
[ -2.867862 ]
[-20.73217 ]]
epoch= 36 loss = 19.887165716818934 b = 9.925617 w = [[ -8.978544 ]
[ 1.7056178 ]
[ -1.3755795 ]
[ 0.12268099]
[ -0.8260821 ]
[ 22.871796 ]
[ -0.7284178 ]
[ -6.45696 ]
[ 5.666094 ]
[ -5.2564545 ]
[ -2.881594 ]
[-20.699532 ]]
epoch= 37 loss = 19.866540418133514 b = 10.091272 w = [[ -9.072603 ]
[ 1.7176012 ]
[ -1.3471004 ]
[ 0.11688287]
[ -0.92811525]
[ 22.87616 ]
[ -0.730987 ]
[ -6.5283713 ]
[ 5.7102056 ]
[ -5.300173 ]
[ -2.8958457 ]
[-20.667597 ]]
epoch= 38 loss = 19.84655585635008 b = 10.255973 w = [[ -9.16269 ]
[ 1.7289914]
[ -1.3200164]
[ 0.1115229]
[ -1.02746 ]
[ 22.87953 ]
[ -0.7338145]
[ -6.5979457]
[ 5.752091 ]
[ -5.342516 ]
[ -2.910552 ]
[-20.636497 ]]
epoch= 39 loss = 19.827194230022545 b = 10.419676 w = [[ -9.248955 ]
[ 1.7398028 ]
[ -1.2942772 ]
[ 0.10656872]
[ -1.1242509 ]
[ 22.881989 ]
[ -0.7368505 ]
[ -6.6657662 ]
[ 5.7918835 ]
[ -5.383562 ]
[ -2.9256687 ]
[-20.606283 ]]
epoch= 40 loss = 19.80842495173308 b = 10.582332 w = [[ -9.331573 ]
[ 1.7500488 ]
[ -1.2698294 ]
[ 0.1019915 ]
[ -1.218611 ]
[ 22.883596 ]
[ -0.74004924]
[ -6.7318883 ]
[ 5.82971 ]
[ -5.4233704 ]
[ -2.9411438 ]
[-20.576973 ]]
epoch= 41 loss = 19.79022036244805 b = 10.743951 w = [[ -9.410695 ]
[ 1.7597514 ]
[ -1.2466215 ]
[ 0.09776533]
[ -1.3106596 ]
[ 22.884367 ]
[ -0.74337256]
[ -6.7963815 ]
[ 5.865695 ]
[ -5.461994 ]
[ -2.9569368 ]
[-20.548565 ]]
epoch= 42 loss = 19.772557086552826 b = 10.90449 w = [[ -9.486461 ]
[ 1.7689276 ]
[ -1.224602 ]
[ 0.09386475]
[ -1.4005071 ]
[ 22.884365 ]
[ -0.74678785]
[ -6.8592973 ]
[ 5.8999496 ]
[ -5.499495 ]
[ -2.973007 ]
[-20.521082 ]]
epoch= 43 loss = 19.755417098352222 b = 11.063955 w = [[ -9.559009 ]
[ 1.7776009 ]
[ -1.2037163 ]
[ 0.09026939]
[ -1.4882517 ]
[ 22.883583 ]
[ -0.75026137]
[ -6.9206896 ]
[ 5.9325867 ]
[ -5.535919 ]
[ -2.989303 ]
[-20.494566 ]]
epoch= 44 loss = 19.738778096573792 b = 11.222289 w = [[ -9.628478 ]
[ 1.785791 ]
[ -1.1839144 ]
[ 0.08695884]
[ -1.5739897 ]
[ 22.882086 ]
[ -0.7537668 ]
[ -6.980616 ]
[ 5.963705 ]
[ -5.571313 ]
[ -3.0057926 ]
[-20.468973 ]]
epoch= 45 loss = 19.722612199989562 b = 11.3795185 w = [[ -9.694994 ]
[ 1.7935212 ]
[ -1.1651485 ]
[ 0.08391371]
[ -1.6578104 ]
[ 22.879871 ]
[ -0.7572823 ]
[ -7.039118 ]
[ 5.9933887 ]
[ -5.605718 ]
[ -3.0224457 ]
[-20.444311 ]]
epoch= 46 loss = 19.706914213606513 b = 11.535623 w = [[ -9.75869 ]
[ 1.8008089 ]
[ -1.1473718 ]
[ 0.081117 ]
[ -1.7398065 ]
[ 22.877012 ]
[ -0.76079136]
[ -7.096251 ]
[ 6.0217304 ]
[ -5.6391726 ]
[ -3.0392458 ]
[-20.420582 ]]
epoch= 47 loss = 19.69165743188988 b = 11.69058 w = [[ -9.819683 ]
[ 1.8076748 ]
[ -1.130533 ]
[ 0.07855233]
[ -1.8200492 ]
[ 22.87351 ]
[ -0.76426905]
[ -7.152051 ]
[ 6.0488124 ]
[ -5.6717143 ]
[ -3.0561526 ]
[-20.397795 ]]
epoch= 48 loss = 19.67681932402663 b = 11.844392 w = [[ -9.87808 ]
[ 1.814139 ]
[ -1.1145889 ]
[ 0.07620447]
[ -1.8986154 ]
[ 22.869387 ]
[ -0.7677054 ]
[ -7.2065682 ]
[ 6.0747123 ]
[ -5.703382 ]
[ -3.073144 ]
[-20.37588 ]]
epoch= 49 loss = 19.662384319130293 b = 11.997031 w = [[ -9.933989 ]
[ 1.8202229 ]
[ -1.0994942 ]
[ 0.07405989]
[ -1.9755716 ]
[ 22.864666 ]
[ -0.77108556]
[ -7.25984 ]
[ 6.0995016 ]
[ -5.7342043 ]
[ -3.0901952 ]
[-20.354834 ]]
epoch= 50 loss = 19.648338488437822 b = 12.148509 w = [[ -9.987511 ]
[ 1.8259448]
[ -1.085209 ]
[ 0.0721045]
[ -2.0509896]
[ 22.859385 ]
[ -0.7744012]
[ -7.3119183]
[ 6.1232414]
[ -5.764211 ]
[ -3.1072927]
[-20.33465 ]]
4、模型应用
模型一般应该用来预测新的样本的值
本例506 条数据都用来训练了，暂时没有新的数据
# 指定某一条看看效果
n = 348
x_test = x_data[n]
x_test = x_test.reshape(1,12)
predict = sess.run(pred,feed_dict={x:x_test})
print("预测值：%f"%predict)
target = y_data[n]
print("标签值：%f"%target)
预测值：23.972691
标签值：24.500000
# 随机指定某一条数据
n = np.random.randint(506)
print(n)
x_test = x_data[n]
x_test = x_test.reshape(1,12)
predict = sess.run(pred,feed_dict={x:x_test})
print("预测值：%f"%predict)
target = y_data[n]
print("标签值：%f"%target)
448
预测值：14.384734
标签值：14.100000
版本3：可视化训练过程中的损失值
loss_list = []
# 迭代训练
for epoch in range(train_epochs):
loss_sum = 0
for xs,ys in zip(x_data,y_data):
# 数据变形，使其与spaceholder形状一致
xs = xs.reshape(1,12)
ys = ys.reshape(1,1)
_,loss = sess.run([optimizer,loss_function],feed_dict={x:xs,y:ys})
#将xs与ys利用占位符喂给x和y，然后得到梯度优化后的损失函数，此时w和b的值存在_里，但是我们并不需要知道具体的值，故用一个符号简单代替即可
loss_sum = loss_sum + loss
# 打乱数据顺序，防止记住顺序
xvalues, yvalues = shuffle(x_data,y_data)
b0temp = b.eval(session=sess)
w0temp = w.eval(session=sess)
loss_average = loss_sum / len(y_data)
loss_list.append(loss_average)
print("epoch=",epoch + 1,"loss = ",loss_average,"b = ",b0temp,"w = ",w0temp)
plt.plot(loss_list)
epoch= 50 loss = 19.233434925812812 b = 18.298111 w = [[-11.041447 ]
[ 1.8939558 ]
[ -0.8549625 ]
[ 0.07317752]
[ -4.6470685 ]
[ 22.261425 ]
[ -0.8367696 ]
[ -8.972372 ]
[ 6.701183 ]
[ -6.657288 ]
[ -3.8620994 ]
[-19.961855 ]]
[]
训练次数不是越多越好，有可能出现过拟合，即只对训练集表现出较好的预测能力，但是对新数据的预测能力会下降
版本4：加上TensorBoard
# 声明会话
sess = tf.Session()
init = tf.global_variables_initializer()
# 启动会话
sess.run(init)
# 为TensorBoard可视化准备数据
# 设置日志存储目录
logdir = 'd:/log'
# 创建一个操作，用于记录损失值loss，后面在TensorBoard中SCALARS栏可见
sum_loss_op = tf.summary.scalar("loss",loss_function)
# 把所有需要记录摘要日志文件的合并，方便一次性写入
merged = tf.summary.merge_all()
# 启动会话
sess.run(init)
# 创建摘要文件写入器（FileWriter），将计算图写入摘要文件，后面在TensorBoard中GRAPHS栏可见
writer = tf.summary.FileWriter(logdir,sess.graph)
# 迭代训练
for epoch in range(train_epochs):
loss_sum = 0.0
for xs,ys in zip(x_data,y_data):
# 数据变形，使其与spaceholder形状一致
xs = xs.reshape(1,12)
ys = ys.reshape(1,1)
_,summary_str,loss = sess.run([optimizer,sum_loss_op,loss_function],feed_dict={x:xs,y:ys})
#将xs与ys利用占位符喂给x和y，然后得到梯度优化后的损失函数，此时w和b的值存在_里，但是我们并不需要知道具体的值，故用一个符号简单代替即可
writer.add_summary(summary_str,epoch)
loss_sum = loss_sum + loss
# 打乱数据顺序，防止记住顺序
xvalues, yvalues = shuffle(x_data,y_data)
b0temp = b.eval(session=sess)
w0temp = w.eval(session=sess)
loss_average = loss_sum / len(y_data)
print("epoch=",epoch + 1,"loss = ",loss_average,"b = ",b0temp,"w = ",w0temp)
epoch= 1 loss = 44.346984727579645 b = 3.6101098 w = [[-0.6028991 ]
[ 1.3716252 ]
[-0.7876034 ]
[ 0.49574542]
[ 2.5262067 ]
[ 7.176039 ]
[-0.05296665]
[ 0.8115916 ]
[ 0.3888658 ]
[ 0.3218941 ]
[ 2.3052237 ]
[-4.386189 ]]
epoch= 2 loss = 32.04108791965448 b = 3.9981675 w = [[-1.1532013 ]
[ 1.9551501 ]
[-1.5167841 ]
[ 0.85052747]
[ 2.8998208 ]
[10.613817 ]
[-0.8205696 ]
[ 0.37133965]
[ 0.6370229 ]
[-0.2686866 ]
[ 1.1436037 ]
[-8.103704 ]]
epoch= 3 loss = 27.34222933747573 b = 4.279024 w = [[ -1.6516982 ]
[ 2.1728437 ]
[ -1.9659317 ]
[ 1.0603381 ]
[ 3.178882 ]
[ 13.210131 ]
[ -1.1971354 ]
[ -0.16200757]
[ 0.9431066 ]
[ -0.7692719 ]
[ 0.21318811]
[-10.960832 ]]
epoch= 4 loss = 24.673796940525087 b = 4.50023 w = [[ -2.1096234]
[ 2.1997828]
[ -2.242004 ]
[ 1.1552098]
[ 3.3611088]
[ 15.200149 ]
[ -1.3592125]
[ -0.706517 ]
[ 1.2668993]
[ -1.2006631]
[ -0.5196944]
[-13.182718 ]]
epoch= 5 loss = 23.105802523715468 b = 4.688873 w = [[ -2.5345595]
[ 2.1340857]
[ -2.4086115]
[ 1.1695445]
[ 3.456134 ]
[ 16.743366 ]
[ -1.4054959]
[ -1.2231828]
[ 1.5871502]
[ -1.5771949]
[ -1.089483 ]
[-14.925047 ]]
epoch= 6 loss = 22.16134066072484 b = 4.8603954 w = [[ -2.9317865]
[ 2.0300012]
[ -2.5042703]
[ 1.1326815]
[ 3.4777637]
[ 17.95115 ]
[ -1.3910842]
[ -1.6959045]
[ 1.8933216]
[ -1.9090716]
[ -1.5275977]
[-16.298841 ]]
epoch= 7 loss = 21.58121336979395 b = 5.023566 w = [[ -3.3051083]
[ 1.9167767]
[ -2.5527997]
[ 1.0666223]
[ 3.4402902]
[ 18.903196 ]
[ -1.3466065]
[ -2.1204534]
[ 2.1808507]
[ -2.2038078]
[ -1.8610458]
[-17.385468 ]]
epoch= 8 loss = 21.218201108343276 b = 5.183318 w = [[ -3.657341 ]
[ 1.8094352]
[ -2.5692813]
[ 0.9865797]
[ 3.3567495]
[ 19.657984 ]
[ -1.2889739]
[ -2.4984608]
[ 2.4484055]
[ -2.4671469]
[ -2.1123066]
[-18.245943 ]]
epoch= 9 loss = 20.985988155955273 b = 5.3423796 w = [[ -3.9906657]
[ 1.7149824]
[ -2.5635278]
[ 0.9024041]
[ 3.2383654]
[ 20.259172 ]
[ -1.2273903]
[ -2.8341599]
[ 2.6963465]
[ -2.703608 ]
[ -2.299666 ]
[-18.926945 ]]
epoch= 10 loss = 20.83295870112948 b = 5.5021973 w = [[ -4.306814 ]
[ 1.6359707 ]
[ -2.542102 ]
[ 0.82005054]
[ 3.0944092 ]
[ 20.73996 ]
[ -1.1667447 ]
[ -3.1326795 ]
[ 2.92583 ]
[ -2.916861 ]
[ -2.4378006 ]
[-19.464714 ]]
epoch= 11 loss = 20.727827887255383 b = 5.6634765 w = [[ -4.607194 ]
[ 1.5725503]
[ -2.5095208]
[ 0.7428122]
[ 2.9323804]
[ 21.125816 ]
[ -1.1095084]
[ -3.3991563]
[ 3.1383364]
[ -3.1099505]
[ -2.5383441]
[-19.887714 ]]
epoch= 12 loss = 20.651557802794766 b = 5.8265033 w = [[ -4.892995 ]
[ 1.523658 ]
[ -2.4689786 ]
[ 0.67224526]
[ 2.7582135 ]
[ 21.436516 ]
[ -1.056813 ]
[ -3.638334 ]
[ 3.3354006 ]
[ -3.2854328 ]
[ -2.6104448 ]
[-20.218552 ]]
epoch= 13 loss = 20.592560407845724 b = 5.991331 w = [[ -5.16522 ]
[ 1.4876795]
[ -2.4227934]
[ 0.6088305]
[ 2.5765417]
[ 21.687523 ]
[ -1.0090615]
[ -3.8544097]
[ 3.5184896]
[ -3.445503 ]
[ -2.6612668]
[-20.475204 ]]
epoch= 14 loss = 20.543824973743366 b = 6.1578693 w = [[ -5.424757 ]
[ 1.4628285 ]
[ -2.3726912 ]
[ 0.5524165 ]
[ 2.3909328 ]
[ 21.891 ]
[ -0.96625865]
[ -4.0509844 ]
[ 3.6889513 ]
[ -3.5920413 ]
[ -2.696368 ]
[-20.672127 ]]
epoch= 15 loss = 20.501146773610966 b = 6.32595 w = [[ -5.672386 ]
[ 1.44735 ]
[ -2.319976 ]
[ 0.5025167]
[ 2.20409 ]
[ 22.056557 ]
[ -0.9282105]
[ -4.2311087]
[ 3.8479712]
[ -3.7266824]
[ -2.7200403]
[-20.820915 ]]
epoch= 16 loss = 20.462054043941045 b = 6.4953823 w = [[ -5.9088006 ]
[ 1.4396188 ]
[ -2.265652 ]
[ 0.45849186]
[ 2.0180266 ]
[ 22.191826 ]
[ -0.8946232 ]
[ -4.397331 ]
[ 3.9965947 ]
[ -3.8508472 ]
[ -2.735589 ]
[-20.930954 ]]
epoch= 17 loss = 20.425134463581962 b = 6.6659684 w = [[ -6.134625 ]
[ 1.4381878 ]
[ -2.2105005 ]
[ 0.41966116]
[ 1.8342335 ]
[ 22.302794 ]
[ -0.8651574 ]
[ -4.5517592 ]
[ 4.1357436 ]
[ -3.965766 ]
[ -2.7455106 ]
[-21.009815 ]]
epoch= 18 loss = 20.389595499823777 b = 6.837511 w = [[ -6.3504333 ]
[ 1.4418002 ]
[ -2.155134 ]
[ 0.38536286]
[ 1.6537652 ]
[ 22.394276 ]
[ -0.83946913]
[ -4.696138 ]
[ 4.2662187 ]
[ -4.07251 ]
[ -2.7517076 ]
[-21.063667 ]]
epoch= 19 loss = 20.355049912910186 b = 7.0098114 w = [[ -6.556744 ]
[ 1.4493804 ]
[ -2.1000302 ]
[ 0.35499093]
[ 1.4773606 ]
[ 22.470066 ]
[ -0.8172184 ]
[ -4.8319025 ]
[ 4.3887243 ]
[ -4.172024 ]
[ -2.7556016 ]
[-21.097462 ]]
epoch= 20 loss = 20.321303464425828 b = 7.18269 w = [[ -6.754046 ]
[ 1.4600201 ]
[ -2.045567 ]
[ 0.32800516]
[ 1.305491 ]
[ 22.533237 ]
[ -0.79808664]
[ -4.960228 ]
[ 4.5038643 ]
[ -4.265119 ]
[ -2.7582548 ]
[-21.115255 ]]
epoch= 21 loss = 20.288277413960977 b = 7.3559747 w = [[ -6.94278 ]
[ 1.4729656]
[ -1.992043 ]
[ 0.3039353]
[ 1.1384501]
[ 22.586176 ]
[ -0.7817691]
[ -5.0820823]
[ 4.612189 ]
[ -4.352504 ]
[ -2.760428 ]
[-21.120298 ]]
epoch= 22 loss = 20.255976657753536 b = 7.529512 w = [[ -7.123358 ]
[ 1.4875929 ]
[ -1.9396926 ]
[ 0.28238067]
[ 0.9763845 ]
[ 22.630833 ]
[ -0.767987 ]
[ -5.1982617 ]
[ 4.714188 ]
[ -4.434795 ]
[ -2.7626817 ]
[-21.115284 ]]
epoch= 23 loss = 20.22442039234261 b = 7.7031617 w = [[ -7.296164 ]
[ 1.5033908]
[ -1.8886874]
[ 0.2629979]
[ 0.8193346]
[ 22.668709 ]
[ -0.7564766]
[ -5.30941 ]
[ 4.8103 ]
[ -4.512533 ]
[ -2.7653892]
[-21.102367 ]]
epoch= 24 loss = 20.193632152690782 b = 7.8767877 w = [[ -7.4615693 ]
[ 1.5199423 ]
[ -1.8391632 ]
[ 0.24549845]
[ 0.66725415]
[ 22.701052 ]
[ -0.74700016]
[ -5.4160767 ]
[ 4.9009175 ]
[ -4.5861807 ]
[ -2.7688181 ]
[-21.083317 ]]
epoch= 25 loss = 20.16363997791389 b = 8.05026 w = [[ -7.6199126 ]
[ 1.5369147 ]
[ -1.791215 ]
[ 0.22963867]
[ 0.5200528 ]
[ 22.728811 ]
[ -0.73933184]
[ -5.5186944 ]
[ 4.986406 ]
[ -4.6561403 ]
[ -2.7731261 ]
[-21.059525 ]]
epoch= 26 loss = 20.134472406776553 b = 8.223491 w = [[ -7.7715034 ]
[ 1.5540417 ]
[ -1.7449114 ]
[ 0.21521527]
[ 0.37759066]
[ 22.75276 ]
[ -0.7332736 ]
[ -5.6176376 ]
[ 5.067092 ]
[ -4.7227654 ]
[ -2.7784088 ]
[-21.03216 ]]
epoch= 27 loss = 20.106140428734335 b = 8.396376 w = [[ -7.91665 ]
[ 1.5711161 ]
[ -1.7002957 ]
[ 0.20205604]
[ 0.23970994]
[ 22.773497 ]
[ -0.7286397 ]
[ -5.7132163 ]
[ 5.143287 ]
[ -4.7863665 ]
[ -2.7847047 ]
[-21.002142 ]]
epoch= 28 loss = 20.078638766050148 b = 8.5688305 w = [[ -8.055634 ]
[ 1.5879754 ]
[ -1.6573887 ]
[ 0.19001842]
[ 0.10623693]
[ 22.791494 ]
[ -0.72526073]
[ -5.8056846 ]
[ 5.215267 ]
[ -4.8472004 ]
[ -2.7920082 ]
[-20.970247 ]]
epoch= 29 loss = 20.05197920028582 b = 8.740764 w = [[ -8.188714 ]
[ 1.6044916 ]
[ -1.6162013 ]
[ 0.17897943]
[ -0.02302845]
[ 22.807207 ]
[ -0.72299093]
[ -5.8952627 ]
[ 5.2832932 ]
[ -4.9055157 ]
[ -2.8003173 ]
[-20.937078 ]]
epoch= 30 loss = 20.026142806785725 b = 8.91211 w = [[ -8.3161545 ]
[ 1.6205711 ]
[ -1.5767202 ]
[ 0.16883452]
[ -0.14826903]
[ 22.820906 ]
[ -0.72168666]
[ -5.9821444 ]
[ 5.347619 ]
[ -4.961507 ]
[ -2.80958 ]
[-20.90313 ]]
epoch= 31 loss = 20.00111694905495 b = 9.0828 w = [[ -8.438215 ]
[ 1.636145 ]
[ -1.5389304 ]
[ 0.15949494]
[ -0.26967722]
[ 22.832846 ]
[ -0.72122526]
[ -6.0664797 ]
[ 5.4084606 ]
[ -5.015347 ]
[ -2.819747 ]
[-20.8688 ]]
epoch= 32 loss = 19.976876023260523 b = 9.252759 w = [[ -8.555112 ]
[ 1.6511654 ]
[ -1.5028042 ]
[ 0.15088564]
[ -0.38743532]
[ 22.843216 ]
[ -0.7214917 ]
[ -6.148412 ]
[ 5.4660387 ]
[ -5.067201 ]
[ -2.8307493 ]
[-20.834362 ]]
epoch= 33 loss = 19.953411039013904 b = 9.421962 w = [[ -8.667068 ]
[ 1.665602 ]
[ -1.4683019 ]
[ 0.14294076]
[ -0.50171745]
[ 22.852142 ]
[ -0.72238165]
[ -6.2280664 ]
[ 5.520563 ]
[ -5.117201 ]
[ -2.8425179 ]
[-20.800135 ]]
epoch= 34 loss = 19.930691390567286 b = 9.590336 w = [[ -8.774292 ]
[ 1.6794354 ]
[ -1.4353894 ]
[ 0.13560352]
[ -0.61270636]
[ 22.859802 ]
[ -0.7238109 ]
[ -6.305546 ]
[ 5.572204 ]
[ -5.1654673 ]
[ -2.8550014 ]
[-20.766256 ]]
epoch= 35 loss = 19.908689436425135 b = 9.757842 w = [[ -8.87699 ]
[ 1.6926562 ]
[ -1.404023 ]
[ 0.12882215]
[ -0.7205627 ]
[ 22.866287 ]
[ -0.7256954 ]
[ -6.3809533 ]
[ 5.621156 ]
[ -5.21211 ]
[ -2.8681278 ]
[-20.732916 ]]
epoch= 36 loss = 19.887386695063512 b = 9.924463 w = [[ -8.97536 ]
[ 1.7052654 ]
[ -1.3741522 ]
[ 0.12255323]
[ -0.82544595]
[ 22.871664 ]
[ -0.7279629 ]
[ -6.4543743 ]
[ 5.667572 ]
[ -5.2572203 ]
[ -2.881837 ]
[-20.700218 ]]
epoch= 37 loss = 19.86675028833604 b = 10.090131 w = [[ -9.069558 ]
[ 1.7172685 ]
[ -1.3457288 ]
[ 0.116758 ]
[ -0.927498 ]
[ 22.876017 ]
[ -0.73054504]
[ -6.5258846 ]
[ 5.7116237 ]
[ -5.300886 ]
[ -2.8960645 ]
[-20.668264 ]]
epoch= 38 loss = 19.846758902883902 b = 10.254826 w = [[ -9.159769 ]
[ 1.7286767 ]
[ -1.3187001 ]
[ 0.11140099]
[ -1.026865 ]
[ 22.879416 ]
[ -0.73338926]
[ -6.595566 ]
[ 5.7534475 ]
[ -5.3431883 ]
[ -2.9107597 ]
[-20.63712 ]]
epoch= 39 loss = 19.82738176727769 b = 10.418527 w = [[ -9.246163 ]
[ 1.7395029 ]
[ -1.293014 ]
[ 0.10644902]
[ -1.1236793 ]
[ 22.881903 ]
[ -0.736444 ]
[ -6.663486 ]
[ 5.793184 ]
[ -5.3841906 ]
[ -2.925864 ]
[-20.60686 ]]
epoch= 40 loss = 19.80859844872877 b = 10.581193 w = [[ -9.328893 ]
[ 1.7497646 ]
[ -1.2686185 ]
[ 0.10187446]
[ -1.2180638 ]
[ 22.883526 ]
[ -0.73966265]
[ -6.729703 ]
[ 5.8309593 ]
[ -5.423962 ]
[ -2.9413295 ]
[-20.577477 ]]
epoch= 41 loss = 19.790382144316983 b = 10.742806 w = [[ -9.408131 ]
[ 1.759481 ]
[ -1.2454603 ]
[ 0.09765093]
[ -1.310128 ]
[ 22.88431 ]
[ -0.74300146]
[ -6.794283 ]
[ 5.866894 ]
[ -5.4625535 ]
[ -2.9571 ]
[-20.549078 ]]
epoch= 42 loss = 19.7727154651905 b = 10.903362 w = [[ -9.484003 ]
[ 1.7686719 ]
[ -1.2234856 ]
[ 0.09375397]
[ -1.3999892 ]
[ 22.88429 ]
[ -0.74643093]
[ -6.857287 ]
[ 5.9011016 ]
[ -5.5000234 ]
[ -2.973147 ]
[-20.521568 ]]
epoch= 43 loss = 19.755573752456257 b = 11.062811 w = [[ -9.556662 ]
[ 1.7773588 ]
[ -1.2026455 ]
[ 0.0901618 ]
[ -1.4877502 ]
[ 22.883541 ]
[ -0.74991935]
[ -6.9187627 ]
[ 5.933685 ]
[ -5.536415 ]
[ -2.9894311 ]
[-20.495026 ]]
epoch= 44 loss = 19.73892638734157 b = 11.221154 w = [[ -9.626234 ]
[ 1.7855614]
[ -1.1828885]
[ 0.0868539]
[ -1.5735025]
[ 22.882063 ]
[ -0.7534398]
[ -6.9787636]
[ 5.9647517]
[ -5.571777 ]
[ -3.0059128]
[-20.469427 ]]
epoch= 45 loss = 19.72275971981826 b = 11.378388 w = [[ -9.692853 ]
[ 1.7933004 ]
[ -1.1641629 ]
[ 0.08381241]
[ -1.6573385 ]
[ 22.879875 ]
[ -0.7569692 ]
[ -7.037346 ]
[ 5.994391 ]
[ -5.6061606 ]
[ -3.0225582 ]
[-20.44474 ]]
epoch= 46 loss = 19.70705129104678 b = 11.534488 w = [[ -9.756643 ]
[ 1.8005998 ]
[ -1.1464226 ]
[ 0.08101843]
[ -1.7393415 ]
[ 22.877008 ]
[ -0.76048803]
[ -7.094551 ]
[ 6.0226994 ]
[ -5.6395864 ]
[ -3.0393372 ]
[-20.421003 ]]
epoch= 47 loss = 19.691786841337766 b = 11.68945 w = [[ -9.817729 ]
[ 1.8074781 ]
[ -1.1296227 ]
[ 0.07845551]
[ -1.8195949 ]
[ 22.87351 ]
[ -0.7639785 ]
[ -7.1504245 ]
[ 6.049739 ]
[ -5.6721025 ]
[ -3.0562296 ]
[-20.398193 ]]
epoch= 48 loss = 19.676944714699722 b = 11.843252 w = [[ -9.876206 ]
[ 1.8139527 ]
[ -1.1137139 ]
[ 0.07610977]
[ -1.898168 ]
[ 22.86938 ]
[ -0.76742446]
[ -7.2050123 ]
[ 6.075606 ]
[ -5.7037435 ]
[ -3.073202 ]
[-20.376253 ]]
epoch= 49 loss = 19.662505444781683 b = 11.99591 w = [[ -9.932195 ]
[ 1.8200471 ]
[ -1.0986578 ]
[ 0.07396667]
[ -1.9751372 ]
[ 22.864668 ]
[ -0.77081823]
[ -7.2583547 ]
[ 6.1003575 ]
[ -5.7345405 ]
[ -3.0902467 ]
[-20.35518 ]]
epoch= 50 loss = 19.64845156447943 b = 12.147389 w = [[ -9.985801 ]
[ 1.8257802 ]
[ -1.0844069 ]
[ 0.07201444]
[ -2.0505586 ]
[ 22.85937 ]
[ -0.77414376]
[ -7.3104887 ]
[ 6.1240644 ]
[ -5.7645226 ]
[ -3.1073253 ]
[-20.334978 ]]
