Texture and Faq more info on R Squared statistic: Difference between pages

From Eigenvector Research Documentation Wiki
(Difference between pages)
Jump to navigation Jump to search
imported>Scott
 
imported>Lyle
No edit summary
 
Line 1: Line 1:
===Purpose===
===Issue:===
Run texture analysis on an image or group of images.


===Synopsis===
Can you give me more information on the R-Squared statistic?


:[y,varargout]= texture(img, method, options, varargin)
===Possible Solutions:===


===Description===
R-Squared (R<sup>2</sup>) is an assessment of how well the model does the prediction (it is similar to RMSEC except that it doesn't show if there is a bias).  
TEXTURE will perform an analysis on any 2 dimensional (m x n pixels) layer of an image or group of images. It loops through input 'img' where mode 3 is considered the variables (slabs of an image) mode and mode 4 is considered the images mode. Any additional inputs to the chosen 'method' must be added after the options input.  


Outputs are 'y' a dataset with each row corresponding to an image and varargout containing any secondary output from the selected method. Outputs for a 'method' must be designated by the user with options.numout. Outputs will be appended together forming one row vector per image. E.G. for 'anglemt' if options.numout = [2 1] then the y{2} and y{1} outputs of 'anglemt' will be concatenated to form the row. Only the first output of a function is used. If there is a second output of a function it is assigned to varargout.  
You can access the R<sup>2</sup> by right-clicking on a scores plot of predicted vs. measured. It is one of the items which show up in the information box ("Show on figure" puts it on the figure).  


Aggregation can be performed across variables (slabs/layers) and or across images. For example, setting options.varmode = 'sum' will sum the layer of an image into one layer.  
Note: in other software, R<sup>2</sup> is for the MODELED data only. In PLS_Toolbox we calculate it for the DISPLAYED data. That means that if you show excluded data, or if you show predicted/test data with calibration data ("Show Cal with Test") the R<sup>2</sup> will be for what is shown and will be different from the calibration data. Turn off the "Show Cal with Test" checkbox on the Plot Controls window to view the R<sup>2</sup> for only the test data.  


====Inputs====
R<sup>2</sup> is calculated as the square of the correlation coefficient between the X and Y axes plotted in the figure. If the only data shown is the estimation of the calibration Y data vs. the actual calibration Y data, this is nearly the same as the standard R<sup>2</sup> for a model as defined by, e.g. Martens and Naes.  
* '''img''' = image data (matrix or DataSet). The first two modes must be pixels, the third mode must be variables, and the fourth mode must be images.
* '''method''' = texture method ('anglemt' 'autocor_img' 'svd_texture' 'fft_texture' 'semivar').


====Optional Inputs====
* '''varargin''' = additional input needed for selected 'method'.


====Outputs====
'''Still having problems? Please contact our helpdesk at [mailto:helpdesk@eigenvector.com helpdesk@eigenvector.com]'''
* '''firstout''' = first output is this.
* '''y''' = dataset where each row (spectra) corresponds to an image.
* '''varargout''' = second ouput of a function (e.g. for anglemt this would be 's', the distance scales). Will transform into row vector of cells.


===Options===
[[Category:FAQ]]
 
options =  a structure array with the following fields:
 
* '''numout''': [{1}] 1xK vector. Number and order of outputs from selected 'method' to use.
* '''imgmode''': [ {'none'} | 'sum' | 'mean' ] aggregation across images.
* '''varmode''': [ {'none'} | 'sum' | 'mean' ] aggregation across variables.
 
TEXTURE FUNCTION SUB-STRUCTURES
These options can be assigned in their nested form. See the particular function for a full description of available options. For example:
: <tt>options.anglemt.xscale = [1:100];</tt>
 
===See Also===
 
[[anglemt]], [[autocor_img]], [[detrend_img]], [[fft_texture]], [[semivar]], [[svd_texture]]

Latest revision as of 13:23, 2 January 2019

Issue:

Can you give me more information on the R-Squared statistic?

Possible Solutions:

R-Squared (R2) is an assessment of how well the model does the prediction (it is similar to RMSEC except that it doesn't show if there is a bias).

You can access the R2 by right-clicking on a scores plot of predicted vs. measured. It is one of the items which show up in the information box ("Show on figure" puts it on the figure).

Note: in other software, R2 is for the MODELED data only. In PLS_Toolbox we calculate it for the DISPLAYED data. That means that if you show excluded data, or if you show predicted/test data with calibration data ("Show Cal with Test") the R2 will be for what is shown and will be different from the calibration data. Turn off the "Show Cal with Test" checkbox on the Plot Controls window to view the R2 for only the test data.

R2 is calculated as the square of the correlation coefficient between the X and Y axes plotted in the figure. If the only data shown is the estimation of the calibration Y data vs. the actual calibration Y data, this is nearly the same as the standard R2 for a model as defined by, e.g. Martens and Naes.


Still having problems? Please contact our helpdesk at helpdesk@eigenvector.com

Retrieved from "https://www.wiki.eigenvector.com/index.php?title=Texture&oldid=10045"