What is regression in Artificial Neural Network (ANN) ?

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
Tahlil
3/20/2011 10:35:14 AM
comp.soft-sys.matlab 207270 articles. 1 followers. lunamoonmoon (258) is leader. Post Follow

2 Replies
1601 Views

Similar Articles

[PageSpeed] 42
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
Greg
3/20/2011 12:56:38 PM
Thank you very much :)
0
4/3/2011 5:06:05 AM
Reply:
Similar Artilces:

Artificial star spikes and/or sparkle
Hello all, I'm looking for a plugin/script/tutorial to create artificial spikes, like they are typical for astronomic photographs of bright stars, eg: http://www.robgendlerastropics.com/M45STLmosaic.html http://www.robgendlerastropics.com/ConemosaicNM.html (please take a look at the bigger image versions also) The Gimp filter under "Filter -> Light and Shadows -> Sparkle" is not usefull for this kind of look. Any recommendations welcome. TIA, Frank Fr@nk Stef@ni schrieb: > Hello all, > > I'm looking for a plugin/script/tutorial to create artificial ...

Data Analysis With Regression Solver
Hi To rediscover The Ohm's Law, I = V / R, by the analysis of a set of data(I,V) we need a Linear Feet to get a graph between experimental points, which contains the origin of the axis. We created a program, which uses Regression Solver to get the Linear Fit, but the graph we get, does not contain the axis origin (I = 0 mA, V = 0 V) You can download our program from the next location http://www.credislabviewtoolkit.home.ro/Download/Ohm's Law - Data Analysis with Regression Solver.llb Please help us to get the best Linear Feet to get a graph, which obtai...

Re: DNS on private network #5
> some DNS's, which are in 2nd level into my domain tree. The changes made in > the file are ... > . 3600000 IN NS dns01.mydomain. > dns01.mydomain. 3600000 A 132.145.160.128 > . 3600000 IN NS dns02.mydomain. > dns02.mydomain. 3600000 A 132.145.160.129 > > In wich dns02 and dns01 are the roots for my private network. So all > the queries, that DNS's of 2nd level can't be resolved, are sent to the > roots of my domain (dns01 or dns02).This process works fine...

Network speed detected as 10mbps not 100mbps
Hi, I have just upgraded a switch to Fast Ethernet speed, and one of my computers is detecting the speed as 10mbps for some reason- why is this? cheers Some systems do not properly negotiate the speed and/or duplex settings. Also, if the swicth is hard set to 100 or 10 and the NIC card is set to auto-negotiate, it may not properly negotiate the duplex. The best thing you can do is to make sure that both the switch port and the NIC card are hard coded to 100 Mbps at Full duplex. Here's a good chart of the problem. It's talking about Cisco devices but it also holds true fo...

Re: Interpreting Regression Intercept w Centered Predictors
On Thu, 2 Apr 2009 09:39:55 -0700, Anthony Pitruzzello <tonypit45@YAHOO.CA> wrote: >Any regression model can be represented such that mean(y) = an >intercept plus a linear combination of b1-k(Mean(X1 – Xk), and the intercept >can be represented as intercept = mean(y) – Summation(b1-k(Mean(X1 – Xk)).� Center the Xs and the intercept = mean(y).� In the output below, mean(y) = 2.48413 while >the intercept = 2.41836.� I can think of >no reason why the intercept should not exactly match mean(y).� Does anyone know? The output was difficult to read due to alignment and wrap...

network packet capture
Hi I wan to capture network traffic so I've been looking at pcaplet. It seems to work but people are saying that its quite old so I was wondering if there was anything newer available. If there is nothing better, I've got two issues with it, the first is I get this message three times when I start up a script: /usr/lib/ruby/site_ruby/1.8/i686-linux/pcap.so: warning: do not use Fixnums as Symbols I've read this is because of something in ruby becoming deprecated, what can I do to stop these warnings? The other problem is that once it has captured a packet it starts flooding my n...

Problems with a network printer
Hi - I have a Canon i550 connected to a Win XP system. I have a MAc Ibook running Tiger. The Mac can connect to the home network via a wireless router. I am trying to print using the Canon hooked into the Win XP system. I believe I have the accounts working properly. I configured the Canon on the Mac using a "generic driver" on the Mac. When I print from the Mac, the job shows up in the XP system print Queue, and the light blinks on the Canon printer, but nothing prints. The job completes through the queue and shows as complete on the Mac , but nothing actually...

Re: QR:Logistic regression (1 values are much more less than the #3
Thanks, My total sample is 7350, and I have got 16 categorical variables, most of them are binary (2 values) others with 3 to 4 categories, I am interested for the moment to the main effect. I will be interested to some interactions What will be the problems if I consider the logistic regression in the situation that I have described? Any comments are very welcome Adel Peter Flom <flom@NDRI.ORG> a �crit : >>> "adel F." 1/23/2006 8:40 am >>> wrote <<< Hi, I have a simple question, I want to model a binary variable using proc logistic, my...

Connecting to Office network
Hi, I am using boardband at home. Is there any way for me to connect instead of dial in to my office's pc or server. If your office has a permanent internet connection you can setup a VPN (Virtual Private Network). Tonnes of information about this topic. First port of call is probably MS website. -- Darwood - MCP, MCSE "hoo" <hooligan7128@yahoo.com> wrote in message news:bgsvvd$no5$1@nobel2.pacific.net.sg... > Hi, > > I am using boardband at home. Is there any way for me to connect instead of > dial in to my office's pc or server. ...

ANN: Artist III
http://www.worldofspectrum.org/infoseekid.cgi?id=0022693 Based on the Arist II 128, the changes are: * Runs on 48K machines as well * Uses the RAMdisc as the default disk drive (modify BASIC to use other systems) * Easier to read menus * Bold, Italic and Bold-Italic fonts * One extra brush * Better patterns * Nicer icons * Starts with blank screen with grid turned on * Restored ZX Printer suport * Hidden easter egg On Jan 25, 10:48=A0pm, OwenBot <cheve...@gmail.com> wrote: > http://www.worldofspectrum.org/infoseekid.cgi?id=3D0022693 > > Based on the Arist II 128, the changes ...

networking reference
I have several linux books with chapters devoted to networking in a general way likw Linux the complete reference and Redhat linux 9 Bible plus others. I would like to find a more complete networking reference. Any suggestions? On Sat, 10 Jan 2004 13:50:13 +0000, jstaff474 wrote: > I have several linux books with chapters devoted to networking in a general > way likw Linux the complete reference and Redhat linux 9 Bible plus others. > > I would like to find a more complete networking reference. Any suggestions? Andrew S. Tanenbaum Computer Networks - Third Edition Prentice Hal...

small network debian windows
On a small network(company), with mostly windows machines is it possible for a debian system to through off that network. I know it is silly, kind of on the lines of I broke the internet, but just checking, I am not used to debian systems. We have had strange activity, post-debian install, the only thing that might be the culprit is I installed a DNS-server(which I know I didnt) or this application(dhclient-2.2.x) is not syncing properly with the cisco router? Here are my netstats, I dont think I am running anything dangerous. tcp 0 0 *:www *:* ...

regression
Hello, I have a big data and I want to construct a regression equation but in the equation (Y=a+b1*X1+b2*X2), I have a problem that a<0 and it is uncompatible with my results so I must have a>0. How can I to do to make a>0 ? Thank you very much mou na wrote: > Hello, > I have a big data and I want to construct a regression equation but in the > equation (Y=a+b1*X1+b2*X2), I have a problem that a<0 and it is uncompatible > with my results so I must have a>0. How can I to do to make a>0 ? > Thank you very much I just Googled on "const...

Re: Q: Generate artificial data for sample size simulation
Jim, I'm not a statistician, but I think you need to define the disribution you want to draw from. It obviously is not normal. Regards, Dennis Diskin "Groeneveld, Jim" <jim.groeneveld@VITATRON.COM> Sent by: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> 03/02/2004 03:36 AM Please respond to "Groeneveld, Jim" To: SAS-L@LISTSERV.UGA.EDU cc: Subject: Q: Generate artificial data for sample size simulation Hi friends, Does anyone of you have (tested) code to generate an artificial dataset of data with (intege...

Re: Cox Regression Time*Predictor
On Fri, 8 May 2009 10:33:18 -0700, Bminer <b_miner@LIVE.COM> wrote: >Hi All- > >I'm dipping my toe into the Cox Regression pool and finding it a >little chilly....I'm leaving the cozy world of the tidy but <perhaps> >too assumption filled parametric survival model to the brave *new* >world....and would like some confirmation from the accomplished L'ers. > >I have a simple model to start off with one nominal and one binary >predictor. The predictors * time are all significant (my luck) though. > > >proc format; >value country 3=...

Permission denied while trying to access another pc in window network.
I have two computer setup in my home for file and internet sharing through a linksys router. one pc is connected to the router using Lan cable while the other pc using wireless adapter. Both pc are running on winxp. My problem is i'm able to see the 2 pc name located in my network workgroup, but i'm unable to access into the pc directory. It's prompt me to change the permission, which i have already set a folder to permission shared "Everyone". Please give me advice on how can i resolve this issue. What operating System are you using on both your PC? &...

[ANN] icalc 0.0.4 is released
URLS Home: http://icalc.rubyforge.org/ Download: http://rubyforge.org/projects/icalc/ ABSTRACT The �icalc� is versatile tool for network/system administators as well as general users (implemented in a single file pure Ruby code). Give it one of the <ip_address> <netmask> | <ip_address/prifix> | <ip_range>, e.g. <start_ipaddr-end_ipaddr> pairs, it compute rest of the things like network, broadcast, prefix, netmask, class and, or subnet charts divided by a valid divider. USAGE $ icalc <address/prefix> [divisor] $ icalc <address-r...

Multiple regression in Oracle
I know Oracle 10gR2 can do linear regression. It does it quite well. But has anyone ever accomplished multiple regression? I found some definitions for multiple regression on the web and it looks like greek. Serious! There were so many alpha's and betas that I thought I was watching some random version of Stargate SG1. Any help would be appreciated. Thanks, -- Andreas Sheriff Oracle 9i Certified Professional Oracle 10g Certified Professional Oracle 9i Certified PL/SQL Developer ---- "If you don't eat your meat, you cannot have any pudding. "How can you have any pud...

Nominal multinomial regression
Hello all SAS-users, Does anybody know whether SAS has a procedure for estimating nominal multinomial models with continuous covariates? If I have understood it right, both GENMOD and LOGISTIC can handle only ordinal multinomial models. CATMOD can handle also nominal ones, but it has limitations for predictor variables (I have continuous predictors, CATMOD does not allow them). Any help is greatly appreciated. Mikko After I mailed the question it took about two minutes to find a solution from SAS-help... I just wonder why I did not see it before sending the mail. Anyway, the estimation of ...

distinguishing between slopes of regression lines
Hello, I have a question about distinguishing between the slopes of regression lines. The problem is like so: I have 10 data points (Y-axis and X-axis) am fitting a regression line using the Matlab function "regress" to the point 1-4 and another regression line to the point 2-9 (note that the sizes of each regression line are different). I want to ask the following question: are the slopes of the two lines are significantly different? I know it should be done using the t-test somehow where I use the b values (the regression slopes) but I am not sure exactly how? What formula should I...

stepwise regression
Hello everybody, Do you know how SAS proceeds when it runs a stepwise regression, please? How does it choose the order to introduce the independent variables? Which are the criteria? Thanks! Nathalie The criteria used is in the documentation for PROC REG for linear stepwise regression; there are too many choices to list here (and the documentaiton is now online at SAS). Doc Muhlbaier Duke Doc "Nathlepine" <nathlepine@aol.com> wrote in message news:20040811171757.25017.00000243@mb-m28.aol.com... > Hello everybody, > > Do you know how SAS proceeds when it runs a st...

[ANN] MacSOUP 2.8.1 available
The 2.8.1 version of MacSOUP is available. This is a bugfix release, it fixes the following problems: - No longer uses 100% CPU while connecting to the server. - Fixed crash on startup when the address book contains groups with an empty name. (It isn't possible to create such groups with the Address Book application, but third-party sync tools can apparently create them.) - Correctly reports Mac OS version in UserAgent header for versions 10.4.10 or newer. - Minor fixes. Available from: <http://home.snafu.de/stk/macsoup/> -- Stefan Haller Berlin, Ger...

[ANN]: Scipad-8.67
ANNOUNCE: Scipad version 8.67 ============================= A new version of the Scipad text editor for Scilab and Scicoslab is available. CHANGES SINCE PREVIOUS VERSION ============================== NEW: * Added incremental search feature (a la emacs) * Added "Load into Scilab (to cursor)" feature * Library functions, Scicos functions and user functions are directly clickable to jump to their source code * Added "Jump to (open file)" action for ctrl + shift + right click on an unknown name in the text - this looks for a file of...

Hyperbolic transfer functions in neural network toolbox
Hi Everybody, i got a problem with implementing my neural network system, because I need tangent hyperbolic transfer function. I use a back propagation feed forward and I need this transfer function. NN tool box hadn't it. Must I generate my own transfer function with custom function commands or could I find an easier way to solve this problem? If I am not wrong, back propagation algorithm can also use hyperbolic transfer functions. Why doesn't the NN tool box posses it? Thanx in advance Evrim Evrim Atilla wrote: > Hi Everybody, > > i got a problem with implementing my ne...

Three strikes on network install of Solaris 11 Express
I believe I've set up AI correctly on my Solaris 11 Express machine. However, when I tried an install with three different client, they all failed in different ways. Can anybody suggest a fix of any of these? The first one I tried had an ASUS motherboard with the ATI chipset. It failed at the TFTP stage. The client sent the boot file name correctly, but in.tftpd sent only a `tsize' option in response. The client never responded to this, but just repeated the boot file name. It was using the Reaktek RTL8111B/8168B Gigabit boot agent. Next I tried an ASUS laptop with an Atom 450 CPU...