similar to: Package AMORE

Hi, I want to use the neural network package AMORE and I don't find in the documentation the weight decay option. Could someone tell if it is possible to add a regularization parameter (also known as a weight decay) to the training method. Is it possible to alter the gradient descent rule for that? Thanks, Ron

I am using the iris example came with nnet package to test AMORE. I can see the outcomes are similar to nnet with adaptative gradient descent. However, when I changed the method in the newff to the batch gradient descent, even by setting the epoch numbers very large, I still found all the iris expected class=2 being classified as class=3. In addition, all those records in the outcomes (y) are the

Is there any function in AMORE switching output into logistical or linear one, like linout=TRUE in nnet. Please give me some help, thanks. -- View this message in context: http://r.789695.n4.nabble.com/Logistical-or-Linear-Output-in-AMORE-tp4302187p4302187.html Sent from the R help mailing list archive at Nabble.com. [[alternative HTML version deleted]]

Hi again, thank you for your reply but i wanted something a bit diferent. I have R in linux but cant get a graphical interface working so i am running from the console. How do i run a program i've writen. I have it saved as a file but cant seem to find out how to execute it. Is it in R, or do i use other of the programs in R's folder and pass the name of the file as arg.? Thanks for your

Hi All, I try to test the neural network package AMORE, I normalized my data first, the input data is X [x1,x2,x3] where x1,x2,x3 each is 100 row 1 column vector. the output data Y is 100 row 1 column vector. my network has neurons=c(3,2,2,1) which 2 hidden layers, 3 node in the input layer while 1 in the output layer. Once the network is trained. I use sim (result$net, z) to

Hello, Just a quick message to say that, ------------------------------------------------------------------------ ------- when choosing (with or without the tcltk window) "Bern 1 (swiss)" as CRAN mirror to install a package (function install.packages) for R-2.5.0, the list of available packages for osx seems not complete: AMORE, AdaptFit, AlgDesign, AnalyzeFMRI, SemiPar,

Ok, makes sense. My suggestion is that we patch the IR Verifier to ensure that the mask is indeed a vector of constants and/or undefs. Right now it only runs the standard checks for instructions. We will also run Alive2 on the test suite to make sure undef is never replaced in practice. Thanks, Nuno -----Original Message----- From: Eli Friedman <efriedma at quicinc.com> Sent: 27 de

Dear colleagues, ?arima? returns directly the 1-step ahead errors but I am interested in obtaining other h-step ahead errors for several ARIMA models I have fitted. Is there any way I can obtain this with R? Any help would be appreciated. Sincerely, Nuno Prista _________________________ CO - FCUL, Lisboa, Portugal CQFE - ODU, Norfolk, USA

Hello All, I have samba (Version 3.4.7) installed on a Ubuntu Server 10.04 (64-bit) using apt. I'm attempting to authenticate users connecting to the samba share over LDAP following the documentation https://help.ubuntu.com/10.04/serverguide/C/samba-ldap.html, but the authentication over LDAP fails. The OpenLDAP server was already configured to include the samba.schema, so i have skipped all

Hello everyone, I am employing the waba method from the multilevel package for obtaining a within-group correlation (Description: http://bg9.imslab.co.jp/Rhelp/R-2.4.0/src/library/multilevel/man/waba.html). Does anybody know a way or a calculation for obtaining a significance value for that correlation? And another question: Does anybody know whether it is possible to save individual

When I first configure the client, in the Received Items, it shows all = items correctly, but then, when starting to donwload, they simply = disappear. All other folders are OK. Problem is just in inbox. I thing (although my tests were not exaustive) that deleting the .imap = folder solves the problem, but I just can't understand why, because it = is recreated again... Any help? Thank you,

On 11/27/19 2:10 AM, Eli Friedman via llvm-dev wrote: The shuffle mask of a shufflevector is special: it's required to be a constant in a specific form. From LangRef: "The shuffle mask operand is required to be a constant vector with either constant integer or undef values." So really, we can resolve this any way we want; "undef" in this context doesn't have to mean

Hello, when I run packageStatus(), I get the following results: > packageStatus() Number of installed packages: ok upgrade unavailable /home/fernando/my_library 38 0 1 /usr/local/lib/R/library 28 0 0 Number of available packages (each package/bundle counted only once): installed

Howdy! the setup: RH-VALE 6.2.4 running samba2.2.2 cvs from 11/24/2001... user running NT 4.0 SP6a the problem: I have a user that cannot print - She can add the printer but can't print. log for that machine shows the following: ... normal looking connections to obtain drivers - all look OK - no complaints.. .. then this - libsmb/namequery.c:name_query(415) Got a positive name query

Hi, So SCEV cannot simplify that expression to -1 because of overflow issues (take e.g. src=0x7FFF..). If you compile with 64 bits integers, then the sexts disappear and so SCEV can do the simplification. Depending on what you want to do, you may want to either do not extend src and dst to 64 bits or add a NSW flag to the computations. That will hopefully make SCEV push the sext outside and

Hi, All I'm trying to save the workspace as a Check Point for possible future come back. The "save.image" function works if the whole coding is in a main function: rm(list=ls()) t = 1 t = t+1 print(paste("before",t)) # [1] "before 2" save.image("tt.RData") t = t+1 print(paste("after",t)) # [1] "after 3"

Hi everyone, I am facing a major trouble with my Samba4 domain Just installed Samba 4.0.6 on Centos 6.4, using internal DNS, when I add an W7 client it is a little bit slow to join the domain, but when I login as as Samba user it takes over 15 minutes. is there any trick to get it faster? I am using an internal DNS server, but it hapens the same with Google or OpenDNS. My resolv.conf is pointing

Dear R users, I have been trying to estimate global sensitivity indices such as the sobol 1st and 2nd order indices. I managed to obtain the PRCC. The example presented in the sensitivity package on sobol2002 seems to work well for linear models: for example: calculate y for given x values. However, when trying to apply this technique to dynamic models (SIR type), the error messages just keep

Hi, after updating to 2.1 (see below) I am no longer able to install all R packages as mentioned at <http://support.stat.ucla.edu/view.php?supportid=30>. After finishing the download, I received the following error: [...] trying URL 'http://www.stats.ox.ac.uk/pub/RWin/bin/windows/contrib/2.1/xgobi_1.2- 13.zip' Content type 'application/zip' length 102623 bytes opened

Not sure how off-topic this is, but should we consider/have we considered porting our InstCombines to Alive? The PLDI '15 paper even demos C++ extraction from Alive theorems. I think it'd be a small step from that to extracting tightly optimized VM code, not unlike what Tablegen emits. Everything would be so clean and readable and organized. And edge cases can still be handled manually,