tensorflow常用函数

1、numpy数组转为张量

import tensorflow as tf
import numpy as np
a = np.arange(0, 5)
b = tf.convert_to_tensor(a, dtype=tf.int64)  #numpy 转为张量，本机默认dtype为int32
print("a:", a)
print("b:", b)

2、创建张量

import tensorflow as tf
import numpy as np
x=np.zeros([2,3],dtype=np.int64)
a = tf.zeros([2, 3])
b = tf.ones(4)
c = tf.fill([2, 2], 9)
d=tf.convert_to_tensor(x)
print("a:", a)
print("b:", b)
print("c:", c)
print('d',d)

3、使用函数创建张量

import tensorflow as tf

d = tf.random.normal([2, 2], mean=0.5, stddev=1)#d为2*2的张量，均值为0.5，标准差为1
print("d:", d)
e = tf.random.truncated_normal([2, 2], mean=0.5, stddev=1,dtype=tf.float64)#正态分布e，shape2*2，均值0.5，标准差1，
print("e:", e)

4、创建均匀分布随机浮点数张量

import tensorflow as tf
import numpy as np
a = tf.random.uniform([2, 2], minval=0, maxval=1)
b = tf.random.uniform((1,10), minval=1, maxval=10,dtype=tf.float64)#范围为[low, high]的均匀分布随机浮点数张量
print("f:", a)
print('a:',b)

5、类型转换

import tensorflow as tf

x1 = tf.constant([1.1, 2.2, 3.3], dtype=tf.float64)
print("x1:", x1)
x2 = tf.cast(x1, tf.int32)
print("x2", x2)
print("minimum of x2：", tf.reduce_min(x2))
print("maxmum of x2:", tf.reduce_max(x2))

6、求均值，求和

import tensorflow as tf
import numpy  as np

x=tf.constant([[1,2,3],[2,3,4]])
print('x:',x)
print('均值：',tf.reduce_sum(x))
print('行求和：',tf.reduce_sum(x,axis=1))

7、加减乘除

import tensorflow as tf
import numpy  as np
a=tf.ones((1,5),dtype=tf.int32)
b=tf.fill((1,5),4)
c=tf.fill((5,1),6)
print('a+b:',tf.add(a,b))
print('a-b:',tf.subtract(a,b))
print('a*c:',tf.multiply(a,c))
print('a/b:',tf.divide(a,b))

8、指数运算

import numpy  as np
a=tf.fill([1,2],9.)
print('a的立方:',tf.pow(a,3))
print('a的平方:',tf.square(a))
print('a的开方:',tf.sqrt(a))
print('a的开方:',tf.pow(a,0.5))

 9、求导

import tensorflow as tf
with tf.GradientTape() as tape:
    x=tf.Variable(tf.constant(3.0))
    y=tf.pow(x,2)
grad=tape.gradient(y,x)
print(grad.numpy())

10

import tensorflow as tf

classes = 3
labels = tf.constant([1, 0, 2])  # 输入的元素值最小为0，最大为2
output = tf.one_hot(labels, depth=classes)
print("result of labels1:", output)

11、

import tensorflow as tf
import math
y = tf.constant([1.01, 2.01, -2.01])
y_pro = tf.nn.softmax(y)# y_pro 符合概率分布
a1=math.exp(1.01)/(math.exp(1.01)+math.exp(2.01)+math.exp(-2.01))
print( y_pro)  # y_pro 符合概率分布y_pro[0]=a1
print("The sum of y_pro:", tf.reduce_sum(y_pro))  # 通过softmax后，所有概率加起来和为1