From lailaliana78 at yahoo.com  Mon Nov  8 19:40:18 2010
From: lailaliana78 at yahoo.com (lailaliana78 at yahoo.com)
Date: Mon, 08 Nov 2010 19:40:18 
Subject: [Biomod-commits] biomod
Message-ID: d26bc2dc06d9b989a0b8fcec40479abc@[192.168.1.1]

they helped get I stumbled on u will be paid to make between $288 and $5,535 nobody believes this can happen until I saw the software since I started money has only increased. u are only required get in on the new program <a href="http://bit.ly/cWr1St">go here to check it out</a>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101108/d566472a/attachment.htm>

From bpetitpierre at gmail.com  Tue Nov  9 10:09:59 2010
From: bpetitpierre at gmail.com (Blaise Petitpierre)
Date: Tue, 9 Nov 2010 10:09:59 +0100
Subject: [Biomod-commits] Interaction between variables in GBM
Message-ID: <AANLkTikN_duz9J_nSyjz-zXG+Oz2-WqLHqtf7kJchHHK@mail.gmail.com>

Dear BIOMODER,

I had a look on the Biomod.Models() function and I was wondering why
interaction depth was set to 7 in the GBM function.

gbm(eval(parse(text = paste(SpNames[i],
                  paste(scopeExpSyst(DataBIOMOD[1:10, 1:NbVar],
                    "GBM"), collapse = "")))), data =
DataBIOMOD[calib.lines,
                  ], distribution = "bernoulli", var.monotone = rep(0,
                  length = NbVar), w = Yweights[calib.lines,
                  i], *interaction.depth = 7*, shrinkage = 0.001,
                  bag.fraction = 0.5, train.fraction = 1, verbose = F,
                  cv.folds = 5)



Interaction depth is the maximum depth of variable interactions. 1 implies
an additive model, 2 implies a model with up to 2-way interactions, etc... I
was wondering if it was because 7 variables were used in the tutorial but in
this case, wouldn't be better to set the interaction depth to the number of
variables (NbVar) ?

However I was wondering the effect of interaction depth on overfitting.

Cheers,
Blaise Petitpierre
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101109/22be7a9c/attachment.htm>

From wilfried.thuiller at ujf-grenoble.fr  Tue Nov  9 10:24:58 2010
From: wilfried.thuiller at ujf-grenoble.fr (Wilfried Thuiller)
Date: Tue, 9 Nov 2010 10:24:58 +0100
Subject: [Biomod-commits] Interaction between variables in GBM
In-Reply-To: <AANLkTikN_duz9J_nSyjz-zXG+Oz2-WqLHqtf7kJchHHK@mail.gmail.com>
References: <AANLkTikN_duz9J_nSyjz-zXG+Oz2-WqLHqtf7kJchHHK@mail.gmail.com>
Message-ID: <A90B9DF6-AB85-4AFB-90BA-86D4C4AA9384@ujf-grenoble.fr>

Dear Blaise,

interaction.depth reflects the size of the trees to fit at each  
iteration. One means means only one node and two leaves, and so on.
There is no magic value and one should select the optimal number using  
independant datasets or cross-validation. However, interaction.deph is  
also influencing the learning rate (skrinkage).
Given the BRT is aready quite time-consuming I did not write such a  
function which would find the optimal numbers.
Instead, I followed Elith et al. Journal of Animal Ecology 2009 where  
7 for interaction.depth and skrinkage of 0.001 were the most suitable  
numbers.

If some of you would like to change this parameter, I cold easily add  
it to the Models call so that each of you can change the default value  
for interaction.depth and shrinkage

All the best,
Wilfried




Le 9 nov. 10 ? 10:09, Blaise Petitpierre a ?crit :

> Dear BIOMODER,
>
> I had a look on the Biomod.Models() function and I was wondering why  
> interaction depth was set to 7 in the GBM function.
>
> gbm(eval(parse(text = paste(SpNames[i],
>                   paste(scopeExpSyst(DataBIOMOD[1:10, 1:NbVar],
>                     "GBM"), collapse = "")))), data =  
> DataBIOMOD[calib.lines,
>                   ], distribution = "bernoulli", var.monotone = rep(0,
>                   length = NbVar), w = Yweights[calib.lines,
>                   i], interaction.depth = 7, shrinkage = 0.001,
>                   bag.fraction = 0.5, train.fraction = 1, verbose = F,
>                   cv.folds = 5)
>
>
>
> Interaction depth is the maximum depth of variable interactions. 1  
> implies an additive model, 2 implies a model with up to 2-way  
> interactions, etc... I was wondering if it was because 7 variables  
> were used in the tutorial but in this case, wouldn't be better to  
> set the interaction depth to the number of variables (NbVar) ?
>
> However I was wondering the effect of interaction depth on  
> overfitting.
>
> Cheers,
> Blaise Petitpierre
> _______________________________________________
> Biomod-commits mailing list
> Biomod-commits at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/biomod-commits

-------------------------
Dr. Wilfried Thuiller
Laboratoire d'Ecologie Alpine, UMR CNRS 5553
Universit? Joseph Fourier
BP53, 38041 Grenoble cedex 9, France
tel: +33 (0)4 76 51 44 97
fax: +33 (0)4 76 51 42 79

Email: wilfried.thuiller at ujf-grenoble.fr
Home page: http://www.will.chez-alice.fr
Website: http://www-leca.ujf-grenoble.fr/equipes/tde.htm

FP6 European MACIS project: http://www.macis-project.net
FP6 European EcoChange project: http://www.ecochange-project.eu






-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101109/506ec2fe/attachment.htm>

From lailaliana78 at yahoo.com  Wed Nov 10 03:32:41 2010
From: lailaliana78 at yahoo.com (lailaliana78 at yahoo.com)
Date: Wed, 10 Nov 2010 03:32:41 
Subject: [Biomod-commits] why are you so hard to reach biomod
Message-ID: 2990d247f501165e6eb25826e0407126@[192.168.1.1]

i saw u recieve payment <a href="http://bit.ly/9L5omq">left click on this</a>



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101110/30b523ed/attachment.htm>

From barantes at gmail.com  Mon Nov 15 19:14:00 2010
From: barantes at gmail.com (Bruno Arantes)
Date: Mon, 15 Nov 2010 16:14:00 -0200
Subject: [Biomod-commits] Error projecting models to another area
Message-ID: <AANLkTi=_Sx+6m6+DZ8Vb5DEUGK10z26QQJfOZZXTx4Gy@mail.gmail.com>

Hi All.

I'm trying to project models built with BIOMOD but I'm getting the following
error:

Error in predict.nnet(object, Proj, type = "raw") : missing values in 'x'

If I set ANN=F in the Projection call, then the error changes to:
Error in if (sum(ProbData) != 0) { :
  missing value where TRUE/FALSE needed

Does anyone know why this is happening and how to fix it? I suspect it could
have something to do with my dataset containing NA values... is there some
way to set na.action?

Here's info on my environment and data:

R version 2.11.1 (2010-05-31) running on Win XP SP3 (32-bit)
BIOMOD 1.1-5 2010-05-04

Command being used for projection:
> Projection(Proj = other.area[,3:11], Proj.name="test", GLM = T, GBM = T,
GAM = T, CTA = T, ANN = T, SRE = T, FDA =T, MARS = T, RF = T, BinRoc=T,
BinKappa=T, BinTSS=T, FiltRoc=F, FiltKappa=F, FiltTSS=F,
repetition.models=F)

> summary(other.area)
       x                y                alt                bio1
 bio10              bio11              bio12            bio15
bio16
 Min.   :-73.79   Min.   :-33.625   Min.   :     0.0   Min.   :     0.0
Min.   :     0.0   Min.   :     0.0   Min.   :     0   Min.   :     0.00
Min.   :     0.0
 1st Qu.:-64.12   1st Qu.:-23.875   1st Qu.:   121.0   1st Qu.:   234.0
1st Qu.:   248.0   1st Qu.:   212.0   1st Qu.:  1342   1st Qu.:    47.00
1st Qu.:   585.0
 Median :-54.37   Median :-14.125   Median :   259.0   Median :   253.0
Median :   260.0   Median :   244.0   Median :  1721   Median :    65.00
Median :   797.0
 Mean   :-54.37   Mean   :-14.125   Mean   :   326.8   Mean   :   244.4
Mean   :   256.2   Mean   :   229.5   Mean   :  1754   Mean   :    60.87
Mean   :   762.3
 3rd Qu.:-44.62   3rd Qu.: -4.375   3rd Qu.:   464.0   3rd Qu.:   262.0
3rd Qu.:   269.0   3rd Qu.:   256.0   3rd Qu.:  2207   3rd Qu.:    78.00
3rd Qu.:   933.0
 Max.   :-34.96   Max.   :  5.375   Max.   :  2054.0   Max.   :   280.0
Max.   :   292.0   Max.   :   272.0   Max.   :  3883   Max.   :   128.00
Max.   :  1628.0
                                    NA's   :117187.0   NA's   :117187.0
NA's   :117187.0   NA's   :117187.0   NA's   :117187   NA's   :117187.00
NA's   :117187.0
     bio17               bio4
 Min.   :     0.0   Min.   :     0
 1st Qu.:    33.0   1st Qu.:   450
 Median :    87.0   Median :   718
 Mean   :   141.7   Mean   :  1075
 3rd Qu.:   215.0   3rd Qu.:  1517
 Max.   :   771.0   Max.   :  4480
 NA's   :117187.0   NA's   :117187


Model training seems to have completed successfully, the only warnings that
appeared were:
glm.fit: fitted probabilities numerically 0 or 1 occurred

Let me know if you need more details about these issues.

Thanks.

-- 
Bruno Arantes de Andrade Bueno
http://twitter.com/barantes
http://barantesbirdwatching.blogspot.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101115/a43ad6ff/attachment.htm>

From barantes at gmail.com  Mon Nov 15 19:47:49 2010
From: barantes at gmail.com (Bruno Arantes)
Date: Mon, 15 Nov 2010 16:47:49 -0200
Subject: [Biomod-commits] Error projecting models to another area
In-Reply-To: <AANLkTi=_Sx+6m6+DZ8Vb5DEUGK10z26QQJfOZZXTx4Gy@mail.gmail.com>
References: <AANLkTi=_Sx+6m6+DZ8Vb5DEUGK10z26QQJfOZZXTx4Gy@mail.gmail.com>
Message-ID: <AANLkTinJ76ZHUk2h_qgsc1nW+UpCc3uHO0iJKc_Wdo+Y@mail.gmail.com>

I solved the issues by using subset() on my dataset to select only those
rows at which there are no NA values.

Thanks.

-- 
Bruno Arantes de Andrade Bueno
http://twitter.com/barantes
http://barantesbirdwatching.blogspot.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101115/7d0f8c31/attachment.htm>

From wilfried.thuiller at ujf-grenoble.fr  Mon Nov 15 20:03:48 2010
From: wilfried.thuiller at ujf-grenoble.fr (Wilfried Thuiller)
Date: Mon, 15 Nov 2010 20:03:48 +0100
Subject: [Biomod-commits] Error projecting models to another area
In-Reply-To: <AANLkTinJ76ZHUk2h_qgsc1nW+UpCc3uHO0iJKc_Wdo+Y@mail.gmail.com>
References: <AANLkTi=_Sx+6m6+DZ8Vb5DEUGK10z26QQJfOZZXTx4Gy@mail.gmail.com>
	<AANLkTinJ76ZHUk2h_qgsc1nW+UpCc3uHO0iJKc_Wdo+Y@mail.gmail.com>
Message-ID: <4945D3BB-40B1-4D8E-9538-C24ECD779471@ujf-grenoble.fr>

Yep, that was the reason.
You can also use na.exclude or na.omit which does the same thing.


Le 15 nov. 2010 ? 19:47, Bruno Arantes a ?crit :

> I solved the issues by using subset() on my dataset to select only those rows at which there are no NA values.
> 
> Thanks.
> 
> -- 
> Bruno Arantes de Andrade Bueno
> http://twitter.com/barantes
> http://barantesbirdwatching.blogspot.com
> 
> _______________________________________________
> Biomod-commits mailing list
> Biomod-commits at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/biomod-commits

--------------------------
Dr. Wilfried Thuiller
Laboratoire d'Ecologie Alpine, UMR CNRS 5553
Universit? Joseph Fourier
BP53, 38041 Grenoble cedex 9, France
tel: +33 (0)4 76 51 44 97
fax: +33 (0)4 76 51 42 79

Email: wilfried.thuiller at ujf-grenoble.fr
Home page: http://www.will.chez-alice.fr
Website: http://www-leca.ujf-grenoble.fr/equipes/tde.htm

FP6 European MACIS project: http://www.macis-project.net
FP6 European EcoChange project: http://www.ecochange-project.eu





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101115/22963b10/attachment.htm>

From cris_a12 at hotmail.com  Mon Nov 15 20:28:49 2010
From: cris_a12 at hotmail.com (Cris Cris)
Date: Mon, 15 Nov 2010 17:28:49 -0200
Subject: [Biomod-commits] ANN and VarImportance
In-Reply-To: <4945D3BB-40B1-4D8E-9538-C24ECD779471@ujf-grenoble.fr>
References: <AANLkTi=_Sx+6m6+DZ8Vb5DEUGK10z26QQJfOZZXTx4Gy@mail.gmail.com>,
	<AANLkTinJ76ZHUk2h_qgsc1nW+UpCc3uHO0iJKc_Wdo+Y@mail.gmail.com>,
	<4945D3BB-40B1-4D8E-9538-C24ECD779471@ujf-grenoble.fr>
Message-ID: <COL115-W64116D6027875A45530927F9360@phx.gbl>


When I include the presence-absence of a host specie as a variable, the VarImportance of ANN shows only zeros associated with the host layer. 
All other algorithms show the opposite: the host is the most important variable.
All the rest is ok: I can plot the ANN's results and reach some good values of AUC but this result of VarImportance seems strange. 
Thank you for the support.
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101115/13cdbe6e/attachment.htm>

From wilfried.thuiller at ujf-grenoble.fr  Mon Nov 15 20:36:25 2010
From: wilfried.thuiller at ujf-grenoble.fr (Wilfried Thuiller)
Date: Mon, 15 Nov 2010 20:36:25 +0100
Subject: [Biomod-commits] ANN and VarImportance
In-Reply-To: <COL115-W64116D6027875A45530927F9360@phx.gbl>
References: <AANLkTi=_Sx+6m6+DZ8Vb5DEUGK10z26QQJfOZZXTx4Gy@mail.gmail.com>,
	<AANLkTinJ76ZHUk2h_qgsc1nW+UpCc3uHO0iJKc_Wdo+Y@mail.gmail.com>,
	<4945D3BB-40B1-4D8E-9538-C24ECD779471@ujf-grenoble.fr>
	<COL115-W64116D6027875A45530927F9360@phx.gbl>
Message-ID: <171223A1-D50A-4354-B04E-53D7BCF2AE27@ujf-grenoble.fr>

yes but this probably the way it is for this algorithm. 
The method to estimate variable importance is the same for all algorithms

I gather that host presence-absence is also correlated to other variables, right? 


Le 15 nov. 2010 ? 20:28, Cris Cris a ?crit :

> When I include the presence-absence of a host specie as a variable, the VarImportance of ANN shows only zeros associated with the host layer.
> All other algorithms show the opposite: the host is the most important variable.
> All the rest is ok: I can plot the ANN's results and reach some good values of AUC but this result of VarImportance seems strange.
> Thank you for the support.
> _______________________________________________
> Biomod-commits mailing list
> Biomod-commits at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/biomod-commits

--------------------------
Dr. Wilfried Thuiller
Laboratoire d'Ecologie Alpine, UMR CNRS 5553
Universit? Joseph Fourier
BP53, 38041 Grenoble cedex 9, France
tel: +33 (0)4 76 51 44 97
fax: +33 (0)4 76 51 42 79

Email: wilfried.thuiller at ujf-grenoble.fr
Home page: http://www.will.chez-alice.fr
Website: http://www-leca.ujf-grenoble.fr/equipes/tde.htm

FP6 European MACIS project: http://www.macis-project.net
FP6 European EcoChange project: http://www.ecochange-project.eu





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101115/bf874f14/attachment-0001.htm>

From lailaliana78 at yahoo.com  Thu Nov 18 10:34:25 2010
From: lailaliana78 at yahoo.com (lailaliana78 at yahoo.com)
Date: Thu, 18 Nov 2010 10:34:25 
Subject: [Biomod-commits] hey biomod
Message-ID: 0ea4ce11408756cb7cd8c47767e56583@[192.168.1.1]

we got my co workers and I found u get munny to make between $523 and $8,799 i wasn't aware it was possible since day 1 the gains have only gained more. u are only required check out the news <a href="http://is.gd/hmz8H">click here to view it</a>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101118/91324670/attachment.htm>

From lailaliana78 at yahoo.com  Thu Nov 18 23:31:06 2010
From: lailaliana78 at yahoo.com (lailaliana78 at yahoo.com)
Date: Thu, 18 Nov 2010 23:31:06 
Subject: [Biomod-commits] deaths in my family biomod
Message-ID: 7fdd695a73eb8559a31e7af225f82b30@[192.168.1.1]

we got u get $$ <a href="http://bit.ly/aCZiYW">click here</a>



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20101118/5248623d/attachment.htm>

From lailaliana78 at yahoo.com  Sat Nov 20 18:31:20 2010
From: lailaliana78 at yahoo.com (lailaliana78 at yahoo.com)
Date: Sat, 20 Nov 2010 18:31:20 
Subject: [Biomod-commits] SO MUCH CRAZINESS biomod
Message-ID: 4fb510dacdc09fc0c28b0d075bbab1da@[192.168.1.1]

Hello! biomod I was exploring the many websites on this internet thing the other evening when I came across something interesting it was an on screen news cast about a single mistress who made a business on the many websites on the internet and ended up making 8000 a month from home.  I was completely set on on finding an alternative way to make money, the recession situation can be felt from all classes of the spectrum. I was very sceptical at first but after a little research I realized that
this was very much not a lie.  By following the abc news' fortune 500 company hand written instructions laid out in the article I have started a home business that nets me over 4000 dollars a month.  You really ought to look into it.  Here's the page y:<a href="http://bit.ly/aNShwv">click this</a>.  you can email me questions at lailaliana78%40yahoo.com.  good luck.   diandra ceasing corozal
__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com