TestPro/test_python/test_tensorflow.py at master · 9527cpp/TestPro · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
#a = tf.constant(1)
#b = tf.Variable(0,dtype=tf.int32)
#b = tf.assign(b,a)
##b = a
#with tf.Session() as sess:
#    sess.run(tf.global_variables_initializer())
#    print(sess.run(a))
#    print(sess.run(b)

#a = tf.constant([[1,2],[3,4]])
#with tf.Session() as sess:
#    b = tf.reduce_max(a)
#    c = tf.reduce_max(a,0)
#    d = tf.reduce_sum(a)
#    e = tf.reduce_mean(a)
#    print(sess.run(c))
#    print(sess.run(c))
#    print(sess.run(b))
#    print(sess.run(a))
#    print(sess.run(d))
#    print(sess.run(e))

#a = tf.Variable(tf.random_normal([2,2]))
##w = tf.Variable([[1,1],[2,2],[3,3]])# 3 row 2 col
#x = tf.Variable([[1],[1]])#2 row 1 col
#b = tf.constant(1)
#y = tf.add(tf.matmul(w,x),b)
#z = tf.zeros(shape=(2,3))
#zz = tf.ones(shape=(2,3))
#with tf.Session() as sess:
#    sess.run(tf.global_variables_initializer())
#    print(sess.run(y))
#    print(sess.run(z))
#    print(sess.run(zz))


#x =tf.placeholder(tf.float32,[2,2])
#def run(_x):
#    with tf.name_scope("run"):
#        sum_summary = tf.summary.scalar("sum_sum",tf.reduce_sum(_x))
#        #return _x,tf.reduce_sum(_x), tf.nn.softmax(_x)
#        sm_summary = tf.summary.histogram("sm_sum",tf.nn.softmax(x))
#        return sum_summary,sm_summary
#def main():
#    with tf.Session() as sess:
#        x_op = run(x)
#        #summary = tf.summary.merge_all()
#        file_writer =  tf.summary.FileWriter("summary_dir",sess.graph)
#        sess.run(tf.global_variables_initializer())
#        for global_step in range(20):
#            #_x,_sum_x,_sm_x=sess.run(x_op,feed_dict={x:[[1,2],[3,4]]})
#            sum_summary ,sm_summary= sess.run(x_op,feed_dict={x:[[1,2],[3,4]]})
#            ##train_summary = sess.run(summary,feed_dict={x:[[1,2],[3,4]]})
#            ##file_writer.add_summary(train_summary,global_step)
#            file_writer.add_summary(sum_summary,global_step)
#            file_writer.add_summary(sm_summary,global_step)
#            #print(_x)
#            #print(_sum_x)
#            #print(_sm_x)
#        file_writer.close()
#    print("main")
#if __name__ == "__main__":
#    main()

#
# def variable_scope(name_scope,name):
#     with tf.name_scope(name_scope) as scope:
#         return tf.Variable(name,tf.float32)
#
#
# def input(_x):
#     name_scope = "input"
#     W = variable_scope(name_scope,"W")
#     W = tf.random_normal([2,2],stddev = 0.01)
#     b = variable_scope(name_scope,"b")
#     b = tf.random_normal([1,2],stddev = 0.01)
#     return tf.add(tf.matmul(_x,W),b)
#
# def main():
#     x = tf.placeholder(tf.float32,[None,2])
#
#     input_op = input(x)
#
#     with tf.Session() as sess:
#         sess.run(tf.global_variables_initializer())
#         _xWb = sess.run(input_op,feed_dict={x:[[1,1],[1,1],[1,1]]})
#         print(_xWb)
#
# if __name__ == "__main__":
#     main()
#

def add_layer(input,insize,outsize,active_fun = None):
    W = tf.Variable(tf.random_normal([insize,outsize]))
    b = tf.Variable(tf.zeros([1,outsize]))
    Wx_b = tf.matmul(input,W) + b
    if active_fun == None:
        return Wx_b
    else:
        return active_fun(Wx_b)


def gen_data(start,stop,count,step):
    x = np.linspace(start,stop,count)[:,np.newaxis]
    y = np.square(x) + np.random.normal(0,0.01,x.shape)
    x_test = x[::step]
    y_test = y[::step]

    x_train = []
    y_train = []
    count = 0
    for _x,_y in zip(x,y):
        if count % step != 0:
            x_train.append(_x)
            y_train.append(_y)
        count +=1
    return x,x_train,x_test,y,y_train,y_test


sess = tf.Session()
x_ph = tf.placeholder(tf.float32,[None,1])
y_ph = tf.placeholder(tf.float32,[None,1])
l1 = add_layer(x_ph,1,30,tf.tanh)
l2 = add_layer(l1,30,1,tf.tanh)
loss = tf.reduce_mean(tf.square(y_ph-l2))
train = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

#x,x_train,x_test,y,y_train,y_test = gen_data(-1,1,10,5)
x,x_train,x_test,y,y_train,y_test = gen_data(-0.5,0.5,500,5)

#plt.plot(x,y)
#plt.show()

#print(x,x_train,x_test,y,y_train,y_test)
sess.run(tf.global_variables_initializer())
for i in range(2000):
    sess.run(train,feed_dict={x_ph:x,y_ph:y})
    if i % 50==0:
        print(sess.run(loss,feed_dict={x_ph:x,y_ph:y}))

_l2 = sess.run(l2,feed_dict={x_ph:x})
plt.scatter(x,y)
plt.plot(x,_l2,'r')
plt.show()