tensolrflow之基础变量

#优化一个乘法算子

 

#coding:utf-8
__author__ = 'similarface'
import tensorflow as tf
sess=tf.Session()
#创建一个常量张量
a=tf.Variable(tf.constant(4.))
x_val=5.
x_data=tf.placeholder(dtype=tf.float32)

#添加计算图
multiplication=tf.multiply(a,x_data)
#我们将声明损失函数为输出与期望目标值100之间的L2距离：
loss = tf.square(tf.subtract(multiplication, 100.))

#初始化模型变量 现在我们并将我们的优化算法声明为标准梯度下降：
init = tf.initialize_all_variables()
sess.run(init)
#标准梯度下降
my_opt = tf.train.GradientDescentOptimizer(0.01)
train_step = my_opt.minimize(loss)

print('优化乘法输出100.')
for i in range(10):
   sess.run(train_step, feed_dict={x_data: x_val})
   a_val = sess.run(a)
   mult_output = sess.run(multiplication, feed_dict={x_data: x_val})
   print(str(a_val) + ' * ' + str(x_val) + ' = ' + str(mult_output))

 

 

__author__ = 'similarface'
from tensorflow.python.framework import ops
import tensorflow as tf
'''
y=a*x+b
'''
ops.reset_default_graph()
sess = tf.Session()
a = tf.Variable(tf.constant(1.))
b = tf.Variable(tf.constant(1.))
x_val = 5.
x_data = tf.placeholder(dtype=tf.float32)
two_gate = tf.add(tf.multiply(a, x_data), b)
loss = tf.square(tf.subtract(two_gate, 50.))
my_opt = tf.train.GradientDescentOptimizer(0.01)
train_step = my_opt.minimize(loss)
init = tf.initialize_all_variables()
sess.run(init)
print('\nOptimizing Two Gate Output to 50.')
for i in range(10):
   a_val, b_val = (sess.run(a), sess.run(b))
   # Run the train step
   sess.run(train_step, feed_dict={x_data: x_val})
   # Get the a and b values
   a_val, b_val = (sess.run(a), sess.run(b))
   # Run the two-gate graph output
   two_gate_output = sess.run(two_gate, feed_dict={x_data: x_val})
   print(str(a_val) + ' * ' + str(x_val) + ' + ' + str(b_val) + '= ' + str(two_gate_output))

'''
result:

10.4 * 5.0 + 2.88= 54.88
14.912 * 5.0 + 3.7824= 78.3424
17.0778 * 5.0 + 4.21555= 89.6043
18.1173 * 5.0 + 4.42347= 95.0101
18.6163 * 5.0 + 4.52326= 97.6048
18.8558 * 5.0 + 4.57117= 98.8503
18.9708 * 5.0 + 4.59416= 99.4482
19.026 * 5.0 + 4.6052= 99.7351
19.0525 * 5.0 + 4.61049= 99.8729
19.0652 * 5.0 + 4.61304= 99.939

'''

#coding:utf-8
__author__ = 'similarface'
'''
使用 Placeholders and Variables

Variables 变量是tensorflow 会跟踪并优化

Placeholders 占位符 类型，维度 占用
'''
import tensorflow as tf
import numpy as np

l_var=tf.Variable(tf.zeros([2,3]))
sess=tf.Session()
init_all=tf.global_variables_initializer()
sess.run(init_all)
print(l_var)

x=tf.placeholder(tf.float32,shape=[2,2])
#identity  x=y
y=tf.identity(x)
x_vals=np.random.rand(2,2)
#Placeholders 需要喂入数据
sess.run(y,feed_dict={x:x_vals})
print(y)

#返回一个给定对角值的对角tensor
'''
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
'''
id_matrix=tf.diag([1.0,1.0,1.0])

#tf.truncated_normal(shape, mean, stddev) :shape表示生成张量的维度，mean是均值，stddev是标准差。这个函数产生正太分布，均值和标准差自己设定
A=tf.truncated_normal([2,3])
#指定值填充矩阵
B = tf.fill([2,3], 5.0)
#均匀分布
C = tf.random_uniform([3,2])
#将np数组转化成tensor
D = tf.convert_to_tensor(np.array([[1., 2., 3.],[-3., -7.,-1.],[0., 5., -2.]]))

print("tf.diag: \n",sess.run(id_matrix))

print("truncated_normal: 2-2\n",sess.run(A))

print('fill:\n',sess.run(B))

print('random_uniform:\n',sess.run(C))

print('convert_to_tensor:\n',sess.run(D))

print("A+B\n",sess.run(A+B))


print('C:\n',sess.run(C))
#C转置
print("C'T:\n",sess.run(tf.transpose(C)))

#行列式
print(sess.run(tf.matrix_determinant(D)))

#就是得到逆矩阵
print(sess.run(tf.matrix_inverse(D)))

#对称正定矩阵
print(sess.run(tf.cholesky(id_matrix)))

#求解特征值和特征向量
print(sess.run(D))
print(sess.run(tf.self_adjoint_eig(D)))