Introduction

In many machine learning algorithms, input and output are both vectors(collection of numbers). The inputs that we deal with-picture, sound, color, etc, have to be converted to numbers so that they can be understood and processed by the computer. In ML, the vector is an array of numbers. One example of a vector would be X=[x1, x2, x3,……, xn]. Similarly, tensors can be defined as an array of numbers with a dimension greater than 2. The color image is an example of a tensor as it has a dimension of 3-length, width, channels. A scalar is considered as a 0th order tensor and the vector is considered as a 1st order tensor. A point to be noted here, often a 60X60 pixel image(60 taken as an example), or a matrix of 60X60 is unrolled into a single column matrix of size 3600. Matrix is considered to be a 2nd order tensor. Video data can be regarded as a 4th order tensor-Nx X Ny X 3 being the 3 dimensions and time being the 4th dimension.

#vector #artificial-intelligence #mathematics #statistics #machine-learning