4.0

4.0 | 1 rating Rate this file 98 Downloads (last 30 days) File Size: 351 KB File ID: #42445
image thumbnail

Improved Feedforward Neural Networks Using PSOGSA

by Seyedali Mirjalili

 

01 Jul 2013

This program is an improved Feedforward Neural Network using a hybrid algorithm called PSOGSA.

| Watch this File

File Information
Description

This work utilizes a hybrid of Particle Swarm Optimization (PSO) and Gravitational Search Algorithm (GSA) called PSOGSA for training Feedforward Neural Networks (FNNs). The algorithm is applied to the well-known Iris dataset.

The main paper: S. Mirjalili, S. Z. Mohd Hashim, and H. Moradian Sardroudi, "Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm," Applied Mathematics and Computation, vol. 218, pp. 11125-11137, 2012.

Required Products MATLAB
MATLAB release MATLAB 7.13 (R2011b)
Tags for This File  
Everyone's Tags
Tags I've Applied
Add New Tags Please login to tag files.
Please login to add a comment or rating.
Comments and Ratings (2)
30 Jul 2013 Seyedali Mirjalili

Hi Giovanni,

if actualvalue=[1 0 0] then Iris class = 1 (-1 normalized )
if actualvalue=[0 1 0] then Iris class = 2 (0 normalized)
if actualvalue=[0 0 1] then Iris class = 3 (1 normalized)

Let me know if I can be of more assistance.

Regards,
Ali

30 Jul 2013 Giovanni

Hi, the iris.txt contains 4x150 inputs and 1x150 outputs.Not clear to me how to build the calculated output, since "actualvalue" is a 1x3 vector, since the output is a single value, how do I compare, which formula I must apply to each row of the "actualvalue" matrix to compare with single output in iris.txt file ?

Contact us