What is data in regression plot in neural network?
2 views (last 30 days)
Show older comments
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/156887/image.jpeg)
I have trained a neural network to recognize three different signs (stop sign, no-left sign and no-entry sign). I have taken 50 images for each class. Every picture has size of 25x25 matrix,so my input is 625x150 matrix and target - 1x150 matrix. After training for 1000 epochs, the network shows the following regression plot. But I do not understand what actually does the circle shaped data mean? And is it supposed to be concentrated as shown in the diagram?
0 Comments
Answers (2)
ALBERT ALEXANDER STONIER
on 4 Oct 2016
The circled shaped in the input data which you have used for the training of neural network.
1 Comment
Greg Heath
on 5 Oct 2016
The circles are your OUTPUT data. HOWEVR, You have 3 classes of target the plots only show outputs of 0 and 1.
Where are the points for the third class?
The standard approach is to have 3 dimensional (0,1) unit vectors for targets. That is, columns of eye(3).
Hope this helps.
Greg
2 Comments
Greg Heath
on 19 May 2018
I'm not offended that you forgot to include the word "please". So don't worry about it anymore.
The R and Rsquare (aka R^2)statistics are basic quantities used in statistical analysis curve-fitting. They are explained in all elementary statistics books as well as Google.
With regard to characterizing a nonlinear curve-fitting model, R is the slope in the linear fit
output = R*target + bias
Hope this helps
Greg
See Also
Categories
Find more on Sequence and Numeric Feature Data Workflows in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!