Manhattandist: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Benjamin
(Created page with "===Purpose=== Calculates ===Synopsis=== :distances = manhattandist(x) :distances = manhattandist(x,basis) :distances = manhattandist(x,options) :distances = manhattandist(x...")
 
imported>Benjamin
No edit summary
Line 1: Line 1:
===Purpose===
===Purpose===
Calculates  
Calculates Manhattan Distance between Samples (rows) of a Dataset Object (DSO) or a matrix.


===Synopsis===
===Synopsis===
Line 10: Line 10:


===Description===
===Description===
Calculates the Manhattan Distance, sum of the absolute value differences, from each row to every other row in the supplied matrix or, optionally, all rows of (x) to all rows in a second matrix (basis).


====Inputs====
====Inputs====


* '''x''' = A dataset object.
* '''x''' = A DSO or a matrix.


====Optional Inputs====
====Optional Inputs====


* '''basis''' = A second dataset object to compare the first dataset object to when calculating Manhattan distance.
* '''basis''' = A second DSO/matrix to compare the first DSO/matrix against when calculating Manhattan distance.
* '''options''' =
* '''options''' = Discussed below.


====Outputs====
====Outputs====


* '''distances''' =  
* '''distances''' = A m-by-m matrix containing the comprehensive calculated Manhattan distances between samples.
 
====Options====
 
options = A structure array with the following fields:
 
* '''waitbar''': [{'auto'}| 'on' | 'off' ], display waitbar. 'Auto' setting will automatically display a waitbar if computation takes longer than 3 seconds.
* '''diag''': {default: 0} Defines the values to be used when comparing a sample to itself. Technically this distance is zero however in some instances, using an alternate value (e.g.: inf) is useful for flagging these self-calculated distances.


===See Also===
===See Also===

Revision as of 13:16, 15 August 2017

Purpose

Calculates Manhattan Distance between Samples (rows) of a Dataset Object (DSO) or a matrix.

Synopsis

distances = manhattandist(x)
distances = manhattandist(x,basis)
distances = manhattandist(x,options)
distances = manhattandist(x,basis,options)

Description

Calculates the Manhattan Distance, sum of the absolute value differences, from each row to every other row in the supplied matrix or, optionally, all rows of (x) to all rows in a second matrix (basis).

Inputs

  • x = A DSO or a matrix.

Optional Inputs

  • basis = A second DSO/matrix to compare the first DSO/matrix against when calculating Manhattan distance.
  • options = Discussed below.

Outputs

  • distances = A m-by-m matrix containing the comprehensive calculated Manhattan distances between samples.

Options

options = A structure array with the following fields:

  • waitbar: [{'auto'}| 'on' | 'off' ], display waitbar. 'Auto' setting will automatically display a waitbar if computation takes longer than 3 seconds.
  • diag: {default: 0} Defines the values to be used when comparing a sample to itself. Technically this distance is zero however in some instances, using an alternate value (e.g.: inf) is useful for flagging these self-calculated distances.

See Also