What is regression in Artificial Neural Network (ANN) ?

  • Permalink
  • submit to reddit
  • Email
  • Follow


Hi i tried to use nftool function of ANN. Where it shows plot regression after training.
Here i am confused what does regression means and What is R value in the plot ?? And from the plotting how can i understand what is my training performance.

Any help will be greatly appreciated :) Thank you.
0
Reply Tahlil 3/20/2011 10:35:14 AM

See related articles to this posting


On Mar 20, 6:35=A0am, "Tahlil Ahmed Chowdhury" <calicrati...@yahoo.com>
wrote:
> Hi i tried to use nftool function of ANN. Where it shows plot regression =
after training.
> Here i am confused what does regression means and What is R value in the =
plot ?? And from the plotting how can i understand what is my training perf=
ormance.
>
> Any help will be greatly appreciated :) Thank you.

http://en.wikipedia.org/wiki/Regression_analysis

Hope this helps.

Greg
0
Reply Greg 3/20/2011 12:56:38 PM

Thank you very much :)
0
Reply calicratis19 (2) 4/3/2011 5:06:05 AM
comp.soft-sys.matlab 203432 articles. 529 followers. Post

2 Replies
1287 Views

Similar Articles

[PageSpeed] 47


  • Permalink
  • submit to reddit
  • Email
  • Follow


Reply:

Similar Artilces:

What is network performance and error in sim function of Artificial Neural Network (ANN)
Hi im using ANN as a beginner and having a lot of trouble with it. I am usinng the sim function to test my input . My code is like this net = newff(ob,obb,hidden_layer_sz, {'radbas' 'radbas'},'traingd','learnlv1' ); net = train(net,ob,obb); [network_output,garbage_1,garbage_2,error,network_performance]=sim(net,ob,{},{},obb); It returns network_output and error and network_performance. But i dont understand what does those value stand for. Like I expected network_performance to be between 0 to 1. so that i can understand prediction accuracy fr...

Artificial Neural Networks
Hello Please let me know of an online site for simulating artificial neural networks. Aslam Atiq Start here: www.membrain.com On Nov 22, 11:32 am, aslat <as...@rediffmail.com> wrote: > Hello > Please let me know of an online site for simulating artificial neural > networks. > > Aslam Atiq http://citeseer.ist.psu.edu/kock96simulation.html http://citeseer.ist.psu.edu/serbedzija96simulating.html ...

neural networks regression
hi frds, i m doing project in neural networks regression.i hav to use feedforward back-propagation algorithm . if u have d code plz send it to kay.s2k@gmail.com ..... plz On Feb 8, 2:00=A0am, "sathish k" <kay....@gmail.com> wrote: > hi frds, > =A0 i m doing project in neural networks regression.i hav to > use feedforward back-propagation algorithm . if u have d > code plz send it to kay....@gmail.com =A0 =A0..... Is the Neural Network Toolbox available? Hope this helps. Greg ...

artificial neural network
hi everybody.. this is srikanth.. I am in need of some info about 'nntool'.. I would like to link this tool to some simulink model. can you please help me.. srikanth hai, This is BaijuMax, try to explain u r problem in detail. Any nntool or neural network related functions very easy to use. But difficult part is how to design Neural network that best suits u r problem. There many functions for training available . Type 'doc nnet' to get the matlab help. If u need further help ,mail me. ...

artificial neural networks
Anyone here using artificial neural network software in their robot? m -- "In theory, there is no difference between theory and practice. In practice, there is." http://annevolve.sourceforge.net is for those seeking intellectual challenges. (and artificial neural network software in C) Humans may know that my email address is zenguy at shaw dot ca. ...

Generalized Regression Neural Networks
hello! I have started playing with GRNN and have found it to be a good tool. I've used it for a redundant temperature sensor. Has anyone else used it with/out success? I haven't been able to find many papers on it so I'm wondering if it's "passe." Any insight would be appreciated. Thanks Maritimer wrote: > hello! > > I have started playing with GRNN and have found it to be a good tool. > I've used it for a redundant temperature sensor. > > Has anyone else used it with/out success? I haven't been able to find > many papers on it so ...

Artificial Neural Networks for Medicine
I am a Medicine student. I recently came to know about this idea of Artificial Neural Network. I just want your help to guide me how to study it, where to start as I dont have a strong Mathematics background. I have the respective understanding of cell biology and neuroscience. On 9 May 2006 02:35:23 -0700, "sadiq" <sadiqwaterrat@gmail.com> wrote: >I am a Medicine student. I recently came to know about this idea of >Artificial Neural Network. I just want your help to guide me how to >study it, where to start as I dont have a strong Mathematics >background. I have...

GA+Artificial Neural Network
Hello everybody, First I would like to ask how can I search the newsgroup archive, if there is one, with keywords? I am doing an turbomachinery optimization with Micro GA program from Carroll, Cu aerospace and fitness evaluations are performed with Artificial Neural Networks, trained using Navier-Stokes solutions (fluid dynamics). I have 27 input parameters to be optimized. The program uses -tournament selection and binary coding. I would like to have your opinions on: 1-the appropriateness of micro GA in the first place (population size 10?) 2-relation between the population size and the ...

General Regression Neural Networks
Hi, I'm learning neural networks and I have a little trouble understanding the second layer (summation layer) of the General Regression Neural Networks. Do I understand correctly that when I want a single output I have two units in the summation layer: - The first unit should sum the outputs of the pattern units (RBF-layer). - The second unit should multiply each output of the pattern unit with the associated weight? Sum after multiplication. Not sure what you mean by "units" the ouput activation function implements both operations. Hope this helps. Greg. ...

Simulating Artificial Neural Network
I have trained a neural network with 7x 600 input data and 8x600 target data matrices using feed forward back propagation(hidden layers 10) method in matlab with nntool.The performance is 0.0249,gradient is 0.00729.After training,simulated the network with the unknown data set of 7x60 . The network divided half of the data (0 0 1 0 0 0 0 0) well and other half it is showing all zero's.I retrained the network and this time some improvement is there.But some of the data are unclassified( 0 0 0 0 0 0 0 0 ).What could be the reason for this ?I f any one knows please share answer. "Sure...

Regression problem in neural networks.
Respected Sir/Madam, I want to calculate the regression through the neural networks. So plz guide me how to do the same. Kindly confirm which would be more suitable: Feed forward backpropagation or GRNN for this. I have used GRNN to some extent but I am not getting the figure (graph) which shows the regression line. Therefore, kindly let me know how to view the GRNN results in graphical form and finally interpret the results. With regards.....! Rakesh Kaundal ...

Artificial neural network in matlab
Does anyone can share some matlab code about Artificial neural network application? My study relative to water resources that prediction water level in the future. thank you ! "QU�N H?NG" <infohquan@gmail.com> wrote in message news:391fe545-3e5a-4c24-b2a9-153207a69731@googlegroups.com... > Does anyone can share some matlab code about Artificial neural network > application? My study relative to water resources that prediction water > level in the future. thank you ! http://www.mathworks.com/help/nnet/index.html and/or look at the User Stories linked to...

Artificial Neural network #2
Dear all; I am developing a Feedforward ANN using the ANN statistical toolbox. net= fitnet(hiddenLayerSize); [net,tr]=train(net,input,targets) I have normalized my inputs and targets data using the function "mapminmax", after the training i looked into the performance of the training (perf), validation (vperf) and testing( tperf) in the training record and the values are extremely low comapred to these obtained without normalizing the inputs and targets. My questions are: 1-So, when normalizing the inputs and targets, should we also, at the end, denormalize the mse values...

Artificial Neural Networks recommendation
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I'd like to use ANNs in python, especially Simple Recurrent Networks. Ideally I'd like to find a quick, pythonic module that is able to simulate different styles of network architectures including some types of recurrent networks (must: SRN/Elman, may: other recurrent networks like ESN or LSTM to experiment with them). So far I found the following Python ANN modules: - - PyNN (just a builder, requires external simulator) – http://neuralensemble.org/trac/PyNN/ - - Con-x (part of pyro) – http://pyrorobotics.org/?page=Conx - - PyNeurGe...

Regression
Hello, I'm trying to create a regression model using the NN toolbox but I can seem to get it. I have an input matrix X (20x7) with 20 observations and 7 "variables" and an output vector Y with 20 observations too. So, what I want is simply a vector containing the weights (w) calculated by the NN. Actually, I want to compare the results with those I've obtained using other regression models. thanks in advance. Sergio. On Sep 21, 10:37=A0am, "Sergio Silveira" <skett...@zipmail.com.br> wrote: > Hello, > > I'm trying to create a regression...

artificial neural network with matlabpool
Hello I'm trying to run a program contained data training with probabilistic neural network, using matlabpool parallel toolbox as the standard user on a PC (not the administrator) even though I could run the same program, on frequent occasion on my previous PC ( Intel R core(TM) i7 CPU 930@ 2.8 GHz) with lower amount of memory (8 GB ) but despite many toils I'm not successful in running the program on my current PC. even though I'm sure that the PC that I'm working with should be able to train and run larger amount of data but it can not. the Matlab version that I used now...

latest in artificial neural network
https://www.youtube.com/watch?v=JSNZA8jVcm4 www.youtube.com/watch?v=JSNZA8jVcm4 ...

General Regression Neural Networks #2
Hi, Being new to GRNNs I'm having trouble in understanding what the weights should be between the hidden layer and the output layer? Basically, the weights between hidden layer and the output layer are just the output values of the training data (Because there is the same number of hidden neurons as the number of training data, there is the same number of weights between the hidden layer and the output layer as the number of training data too.). Sami Lakka wrote: > Hi, > > > Being new to GRNNs I'm having trouble in understanding what the weights > should be bet...

Is neural network same as artificial intelligence (AI) ?
Is neural network same as artificial intelligence (AI) ? If my application is base on a neural network for recognizing things, for example objects in a picture, can I tell people that my application is base on artificial intelligence (AI) ?or is it something else? Just wanted to understand this point for once. Thanks :-) On Jul 11, 3:49=A0am, Sigmoid 2009 <sigmoid2...@gmail.com> wrote: > Is neural network same as artificial intelligence (AI) ? > > If my application is base on a neural network for recognizing things, > for example objects in a picture, can I tell people t...

Efficient arrays for artificial neural networks?
Hi group, I've been toying with the idea of spending some upcoming free time of mine this summer writing my own image recognition program, based primarily on neural networks. I've already written a simple one that counts the number of 1's & 0's in a binary string. After a bit of training, it gets it right most of the time. It's learning technique is simple - it starts off with a fixed topology (number of layers, neurons per layer) and randomly changes weights until the error is minimised. In the past I stored the neural data (number of neurons per layer, which is con...

How to compare the results of neural networks and regression?
Hello, I want to compare my results of an MLP with traditional (linear) regression and I'm not quite sure how to do that correctly. Is r-square a good validator or not and on which sets should the models be "trained" and validated. Which other validation possiblities can/should be used (i.e. MSE)? Read a lot of articles on comparisons between those models now and most of them give very little insights into the exact method of comparison. I'd be glad if anybody had some literature tips for me or some other help. Thank you very much. Bastian ...

Need Help with Artificial Neural Network
Can anyone help me in suggesting how can I extract the weight values obtained from the neural network training. A little help and suggestion in this regard would be highly appreciated. may be: net.IW net.LW net.b thanks a lot for the help....:D...btw can we extract the individual neurons output? "Sandro " <smgoni@unq.edu.ar> wrote in message <icjsvj$6nv$1@fred.mathworks.com>... > may be: > > net.IW > net.LW > net.b I do not know how to do it, ie, if exist any function to do it. However, tha output depend on the input. If the input is x,...

train general regression neural network
Hi, I want to use "nntool", the GUI function to create a GRNN network and train it. But I find that the train button is blocked. When I use feed-forward backprop to create the work, it seems work it again. Can anyone tell me why? On Aug 28, 2:39=A0am, "Huang HungWei" <h4046...@gmail.com> wrote: > Hi, I want to use "nntool", the GUI function to create a > GRNN network and train it. But I find that the train button > is blocked. When I use feed-forward backprop to create the > work, it seems work it again. Can anyone tell me why? My guess is t...

Is universal artificial neural network possible?
There are numerous algorithms or architectures of networks artificial intelligence at present, while each of them only deals with a narrow scope of applications, although many of them involves considerable efforts and complicated methods. The current achievements impress us that true intelligence is far to attain. It was the borrowed idea from biology decades ago that contributes to the advent of artificial neural network, which opened the new era although the the model is much simpler than real brain. But since then researchers have been concentrating on more and more complex mathematical me...