vec2ind

(To be removed) Convert vectors to indices

vec2ind will be removed in a future release. For more information, see Transition Legacy Neural Network Code to dlnetwork Workflows.

For advice on updating your code, see Version History.

Syntax

[ind,N] = vec2ind(vec)

Description

[ind,N] = vec2ind(vec) takes a matrix of vectors, each containing a single 1 and returns the indices of the ones, ind, and the number of rows in vec, N.

ind2vec and vec2ind allow indices to be represented either by themselves or as vectors containing a 1 in the row of the index they represent.

Examples

Convert Three Vectors to Indices and Back

This example shows how to convert three vectors to indices and back, using both the ind2vec and vec2ind functions.

Define three vector with all zeros in the last row and convert it to indices.

vec = [0 0 1 0; 1 0 0 0; 0 1 0 0]'
[ind,n] = vec2ind(vec)

vec =
     0     1     0
     0     0     1
     1     0     0
     0     0     0
ind =
     3     1     2

n =
     4

Convert the indices to vector, while preserving the number of rows.

vec2 = full(ind2vec(ind,n)) 

vec2 =
     0     1     0
     0     0     1
     1     0     0
     0     0     0

Input Arguments

vec — Matrix of vectors

matrix

Vector representation of the indices, specified as a matrix of vectors, each containing a single 1.

Output Arguments

ind — Indices

row vector

Indices, returned as a row vector.

N — Number of rows

scalar

Number of rows of the input matrix, returned as a scalar.

Version History

Introduced before R2006a

R2026a: vec2ind will be removed

vec2ind will be removed in a future release.

To learn more about data processing for neural networks, see these examples and topics:

• Preprocess Data for Domain-Specific Deep Learning Applications

• Datastores for Deep Learning

• Choose an App to Label Ground Truth Data

To convert one-hot encoded vectors to integers, use the onehotdecode function instead.

For more information, see Transition Legacy Neural Network Code to dlnetwork Workflows.