2019独角兽企业重金招聘Python工程师标准>>>
张量
定义
Tensor is just a REPRESENTATION for a d-way “matrix” (called “array” in Matlab, e.g., A(2,2,1,2) ).
In fact tensors are merely a generalisation of scalars and vectors; a scalar is a zero rank tensor, and a vector is a first rank tensor. The rank (or order) of a tensor is defined by the number of directions (and hence the dimensionality of the array) required to describe it.
应用前景
Big data,Signal processing,Chemometrics,Model reduction,Nonlinear system modeling
例如:图像识别,人脸识别(不同表情、不同状态),一张照片是一个二维的矩阵(矩阵-二维数组-二阶张量),多张照片(10种表情)摞起来。三阶张量。100个人,四个维度-四阶张量。定义了数据的不同表示。随着大数据时代的来临,数据量越来越大了,越来越复杂,维度越来越多,数据的表示相应也会相对复杂一些。
张量计算
张量计算遵循的运算规则是沿着某一个维度,与矩阵或者向量的对应项进行运算即可,尤其需要注意运算的顺序和细节。