ndarray-unsqueeze

Add singleton dimensions to an ndarray

Introduction

This module takes an input ndarray and either appends a singleton dimension (a dimension of length one) or inserts it before a specific dimension.

Examples

var ndarray = require('ndarray')
var unsqueeze = require('ndarray-unsqueeze')
 
unsqueeze(ndarray([1, 2, 3, 4], [2, 2]))
// => ndarray([1, 2, 3, 4], [2, 2, 1])
 
unsqueeze(ndarray([1, 2, 3, 4], [2, 2]), 0)
// => ndarray([1, 2, 3, 4], [1, 2, 2])

Note that ndarrays have no concept of row or column vectors. If you need a matrix explicitly representing a row or column vector, you can use unsqueeze:

var show = require('ndarray-show')
 
// Create a 3 x 1 matrix by appending a singleton dimension:
show(unsqueeze(ndarray([1,2,3])))
// => 1.000
//    2.000
//    3.000
 
// Create a 1 x 3 matrix by prepending a singleton dimension:
show(unsqueeze(ndarray([1,2,3]), 0))
// => 1.000  2.000  3.000
 

Installation

$ npm install ndarray-unsqueeze

API

require('ndarray-unsqueeze')(input[, axis])

Arguments:

• input: The input ndarray to be unsqueeze
• axes (optional): An integer index of the dimension at which to insert the singleton dimension. If unspecified, singleton dimension is appended to the shape.

Returns: A new array view of the unsqueezed ndarray (i.e. a new ndarray object with the same underlying data).

See Also

• ndarray-squeeze

License

© 2016 Ricky Reusser. MIT License.