tf.diag(diagonal,name=None) #生成對(duì)角矩陣
|
1
2
3
4
|
import tensorflowas tf;diagonal=[1,1,1,1]with tf.Session() as sess: print(sess.run(tf.diag(diagonal))) |
|
1
2
3
4
|
#輸出的結(jié)果為[[1 0 0 0] [0 1 0 0] [0 0 1 0] [0 0 0 1]] |
tf.diag_part(input,name=None) #功能與tf.diag函數(shù)相反,返回對(duì)角陣的對(duì)角元素
|
1
2
3
4
|
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))) |
|
1
|
#輸出結(jié)果為[1,1,1,1] |
tf.trace(x,name=None) #求一個(gè)2維Tensor足跡，即為對(duì)角值diagonal之和
|
1
2
3
4
|
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)))#輸出結(jié)果為4 |
tf.transpose(a,perm=None,name='transpose') #調(diào)換tensor的維度順序，按照列表perm的維度排列調(diào)換tensor的順序
|
1
2
3
4
5
6
7
|
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))) #輸出結(jié)果為[[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) #矩陣相乘
transpose_a=False,transpose_b=False #運(yùn)算前是否轉(zhuǎn)置
a_is_sparse=False,b_is_sparse=False #a,b是否當(dāng)作系數(shù)矩陣進(jìn)行運(yùn)算
|
1
2
3
4
5
|
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))) |
|
1
2
|
#輸出結(jié)果為[[2 1] [0 6]] |
tf.matrix_determinant(input,name=None) #計(jì)算行列式
|
1
2
3
4
|
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))) |
|
1
|
#輸出結(jié)果為3.0 |
tf.matrix_inverse(input,adjoint=None,name=None)
adjoint決定計(jì)算前是否進(jìn)行轉(zhuǎn)置
|
1
2
3
4
|
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))) |
|
1
2
|
#輸出結(jié)果為[[ 1. 0. ] [ 0. 0.5]] |
tf.cholesky(input,name=None) #對(duì)輸入方陣cholesky分解，即為將一個(gè)對(duì)稱正定矩陣表示成一個(gè)下三角矩陣L和其轉(zhuǎn)置的乘積德分解
|
1
2
3
4
|
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))) |
|
1
2
|
#輸出結(jié)果為[[ 1. 0. ] [ 0. 1.41421356]] |
以上這篇對(duì)Tensorflow中的矩陣運(yùn)算函數(shù)詳解就是小編分享給大家的全部?jī)?nèi)容了，希望能給大家一個(gè)參考，也希望大家多多支持服務(wù)器之家。
原文鏈接：https://blog.csdn.net/zSean/article/details/75154118
