对Tensorflow中的矩阵运算函数详解
tf.diag(diagonal,name=None) #生成对角矩阵
import tensorflowas tf; diagonal=[1,1,1,1] with tf.Session() as sess: print(sess.run(tf.diag(diagonal)))
#输出的结果为[[1 0 0 0]
[0 1 0 0]
[0 0 1 0]
[0 0 0 1]]
tf.diag_part(input,name=None) #功能与tf.diag函数相反,返回对角阵的对角元素
import tensorflow as tf; diagonal =tf.constant([[1,0,0,0],[0,1,0,0],[0,0,1,0],[0,0,0,1]]) with tf.Session() as sess: print(sess.run(tf.diag_part(diagonal)))
#输出结果为[1,1,1,1]
tf.trace(x,name=None) #求一个2维Tensor足迹，即为对角值diagonal之和
import tensorflow as tf; diagonal =tf.constant([[1,0,0,3],[0,1,2,0],[0,1,1,0],[1,0,0,1]]) with tf.Session() as sess: print(sess.run(tf.trace(diagonal)))#输出结果为4
tf.transpose(a,perm=None,name='transpose') #调换tensor的维度顺序，按照列表perm的维度排列调换tensor的顺序
import tensorflow as tf;
diagonal =tf.constant([[1,0,0,3],[0,1,2,0],[0,1,1,0],[1,0,0,1]])
with tf.Session() as sess:
print(sess.run(tf.transpose(diagonal))) #输出结果为[[1 0 0 1]
[0 1 1 0]
[0 2 1 0]
[3 0 0 1]]
tf.matmul(a,b,transpose_a=False,transpose_b=False,a_is_sparse=False,b_is_sparse=False,name=None) #矩阵相乘
transpose_a=False,transpose_b=False #运算前是否转置
a_is_sparse=False,b_is_sparse=False #a,b是否当作系数矩阵进行运算
import tensorflow as tf; A =tf.constant([1,0,0,3],shape=[2,2]) B =tf.constant([2,1,0,2],shape=[2,2]) with tf.Session() as sess: print(sess.run(tf.matmul(A,B)))
#输出结果为[[2 1] [0 6]]
tf.matrix_determinant(input,name=None) #计算行列式
import tensorflow as tf; A =tf.constant([1,0,0,3],shape=[2,2],dtype=tf.float32) with tf.Session() as sess: print(sess.run(tf.matrix_determinant(A)))
#输出结果为3.0
tf.matrix_inverse(input,adjoint=None,name=None)
adjoint决定计算前是否进行转置
import tensorflow as tf; A =tf.constant([1,0,0,2],shape=[2,2],dtype=tf.float64) with tf.Session() as sess: print(sess.run(tf.matrix_inverse(A)))
#输出结果为[[ 1. 0. ] [ 0. 0.5]]
tf.cholesky(input,name=None) #对输入方阵cholesky分解，即为将一个对称正定矩阵表示成一个下三角矩阵L和其转置的乘积德分解
import tensorflow as tf; A =tf.constant([1,0,0,2],shape=[2,2],dtype=tf.float64) with tf.Session() as sess: print(sess.run(tf.cholesky(A)))
#输出结果为[[ 1. 0. ] [ 0. 1.41421356]]
以上这篇对Tensorflow中的矩阵运算函数详解就是小编分享给大家的全部内容了，希望能给大家一个参考，也希望大家多多支持【听图阁-专注于Python设计】。
