[R-es] Gráficas 3D

Francisco Rodríguez fjroar en hotmail.com
Lun Feb 19 22:21:02 CET 2018 

Os mando un ejemplo que ten�a a mano, un saludo

w1 = seq(-10,10,length=50)
w2 = w1

# Definimos la funci�n que dibujaremos
f1 = function(w1,w2)
 {(1 - (2*w1 + w2*3))^2+ (1 -(w1 + w2*4))^2 + (0 - ((-1)*w1 + w2))^2 +
     (0 - ((-1)*w1 + (-1)*w2))^2 + (0 - (2*w1 + (-1)*w2))^2}

# La funci�n outer evalua la funci�n f en cada
z = outer(w1,w2,f1)
#z = f(x, y)

# Un gr�fico en perspectiva
persp(w1,w2,z, theta = 30, phi = 30, expand = 0.5, col = "lightblue")





Enviado desde Correo<https://go.microsoft.com/fwlink/?LinkId=550986> para Windows 10

De: Andr�s Hirigoyen<mailto:andreshirigoyen en gmail.com>
Enviado: lunes, 19 de febrero de 2018 22:17
Para: Carlos Ortega<mailto:cof en qualityexcellence.es>
CC: Lista R<mailto:r-help-es en r-project.org>
Asunto: Re: [R-es] Gr�ficas 3D

Manos a la.obra

El lun., 19 de feb. de 2018 18:15, Carlos Ortega <cof en qualityexcellence.es>
escribi�:

> Hola,
>
> S�, mira estos ejemplos de c�mo hacerlo...
> Busca las ecuaciones param�tricas de cada una de estas figuras geom�tricas
> (lo puedes encontrar en Wikipedia o en Wolfram Mathematica) y ya lo tendr�s
> casi todo...
>
> https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Falstatr.blogspot.com.es%2F2014%2F02%2Fr-fun-with-surf3d-function.html&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=bH4T4%2FWCruiEO5iumCe7Jrwar0Fu1iOKkLIdNeGuTHc%3D&reserved=0
>
> Saludos,
> Carlos Ortega
> https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>
> El 19 de febrero de 2018, 21:32, Andr�s Hirigoyen <
> andreshirigoyen en gmail.com> escribi�:
>
>> Gracias Carlos, mi idea es construir un cono, un cilindro  u otros cuerpos
>> geom�trico y luego graficarlos. Alguna idea de como empezar?
>>
>> Muchas gracias como siempre
>>
>> El lun., 19 de feb. de 2018 15:06, <r-help-es-request en r-project.org>
>> escribi�:
>>
>> > Env�e los mensajes para la lista R-help-es a
>> >         r-help-es en r-project.org
>> >
>> > Para subscribirse o anular su subscripci�n a trav�s de la WEB
>> >         https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-help-es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=HGYecrba7vZLzE8xbzcDWE3BX%2BFhL2Uq6lKTDHrmYOw%3D&reserved=0
>> >
>> > O por correo electr�nico, enviando un mensaje con el texto "help" en
>> > el asunto (subject) o en el cuerpo a:
>> >         r-help-es-request en r-project.org
>> >
>> > Puede contactar con el responsable de la lista escribiendo a:
>> >         r-help-es-owner en r-project.org
>> >
>> > Si responde a alg�n contenido de este mensaje, por favor, edite la
>> > linea del asunto (subject) para que el texto sea mas especifico que:
>> > "Re: Contents of R-help-es digest...". Adem�s, por favor, incluya en
>> > la respuesta s�lo aquellas partes del mensaje a las que est�
>> > respondiendo.
>> >
>> >
>> > Asuntos del d�a:
>> >
>> >    1. Gr�fica 3D (Andr�s Hirigoyen)
>> >    2. Re:  gbm.step para clasificaci�n no binaria (Manuel Mendoza)
>> >    3. Re: Gr�fica 3D (Carlos Ortega)
>> >    4. Re:  gbm.step para clasificaci�n no binaria (Carlos Ortega)
>> >    5. Re:  gbm.step para clasificaci�n no binaria (Manuel Mendoza)
>> >
>> > ----------------------------------------------------------------------
>> >
>> > Message: 1
>> > Date: Mon, 19 Feb 2018 16:44:00 +0000
>> > From: Andr�s Hirigoyen <andreshirigoyen en gmail.com>
>> > To: Lista R <r-help-es en r-project.org>
>> > Subject: [R-es] Gr�fica 3D
>> > Message-ID:
>> >         <
>> > CAOaTypVZASi_LTfrSwcsz3SoXzntnUtXsDHaBT2yzv64R0n-zg en mail.gmail.com>
>> > Content-Type: text/plain; charset="utf-8"
>> >
>> > Buenas tardes �me recomiendan alg�n paquete para graficar objetos
>> > geom�tricos en 3D? Alguien ya hizo este tipo de gr�ficos?
>> >
>> > Muchas gracias
>> >
>> >         [[alternative HTML version deleted]]
>> >
>> >
>> >
>> >
>> > ------------------------------
>> >
>> > Message: 2
>> > Date: Mon, 19 Feb 2018 18:01:21 +0100
>> > From: Manuel Mendoza <mmendoza en mncn.csic.es>
>> > To: Carlos Ortega <cof en qualityexcellence.es>
>> > Cc: "r-help-es en r-project.org" <r-help-es en r-project.org>
>> > Subject: Re: [R-es]  gbm.step para clasificaci�n no binaria
>> > Message-ID:
>> >         <20180219180121.Horde.4LStqNNI0EIN9hWJghBdZg4 en webmail.csic.es>
>> > Content-Type: text/plain; charset="utf-8"; Format="flowed";
>> >         DelSp="Yes"
>> >
>> >
>> > Gracias Carlos. Hasta donde yo entiendo si las hay:
>> >
>> > El argumento family puede ser:
>> >
>> > "gaussian" (for minimizing squared error); por lo que tiene que ser
>> > num�rica
>> > "bernoulli" (logistic regression for 0-1 out-comes); binaria por narices
>> > "poisson" (count outcomes; requires the response to be a positive
>> > integer); num�rica tambi�n, pues.
>> >
>> > La �nica podr�a ser "laplace" (for minimizing absolute loss), pero me
>> > da este error: Error in while (delta.deviance > tolerance.test &
>> > n.fitted < max.trees) { :
>> >    missing value where TRUE/FALSE needed
>> >
>> > Supongo que loss se refiere a la funci�n de p�rdida, y como habla de
>> > deviance (la funci�n de p�rdida por defecto en gbm) pienso que tambi�n
>> > es para variable respuesta num�rica, aunque no lo encontr� por ning�n
>> > lado. Por eso pregunt�.
>> >
>> > Probar� sin indicar la family, a ver si funciona.
>> > Gracias,
>> > Manuel
>> >
>> >
>> > Quoting Carlos Ortega <cof en qualityexcellence.es>:
>> >
>> > > Hola,
>> > >
>> > > No hay ninguna limitaci�n en la ayuda de la funci�n en este sentido.
>> > > Tan solo se indica que han de existir dos niveles en la variable
>> > > predictora, vaya que al menos sea binaria...
>> > > En la funci�n en el par�metro "gbm.y" es donde indicas qu� columna es
>> la
>> > > predictora. No hay otro par�metro donde por otro lado le indiques si
>> es
>> > un
>> > > modelo binario o multinominal...
>> > >
>> > > Saludos,
>> > > Carlos Ortega
>> > > https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> > >
>> > >
>> > > 2018-02-19 14:03 GMT+01:00 Manuel Mendoza <mmendoza en mncn.csic.es>:
>> > >
>> > >> Hola erreros, �sab�is si gbm.step puede usarse para clasificaci�n no
>> > >> binaria?
>> > >> Gracias
>> > >> --
>> > >> Dr Manuel Mendoza
>> > >> Department of Biogeography and Global Change
>> > >> National Museum of Natural History (MNCN)
>> > >> Spanish Scientific Council (CSIC)
>> > >> C/ Serrano 115bis, 28006 MADRID
>> > >> Spain
>> > >>
>> > >> _______________________________________________
>> > >> R-help-es mailing list
>> > >> R-help-es en r-project.org
>> > >> https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-help-es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=HGYecrba7vZLzE8xbzcDWE3BX%2BFhL2Uq6lKTDHrmYOw%3D&reserved=0
>> > >>
>> > >
>> > >
>> > >
>> > > --
>> > > Saludos,
>> > > Carlos Ortega
>> > > https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> >
>> >
>> > --
>> > Dr Manuel Mendoza
>> > Department of Biogeography and Global Change
>> > National Museum of Natural History (MNCN)
>> > Spanish Scientific Council (CSIC)
>> > C/ Serrano 115bis, 28006 MADRID
>> > Spain
>> >
>> >
>> >
>> >
>> > ------------------------------
>> >
>> > Message: 3
>> > Date: Mon, 19 Feb 2018 18:07:12 +0100
>> > From: Carlos Ortega <cof en qualityexcellence.es>
>> > To: Andr�s Hirigoyen <andreshirigoyen en gmail.com>
>> > Cc: Lista R <r-help-es en r-project.org>
>> > Subject: Re: [R-es] Gr�fica 3D
>> > Message-ID:
>> >         <CAOKbq8iWN+D0jKWwXXwQE_ax=
>> > Q4mUA9SdTY7tEqBZMMJucH_rg en mail.gmail.com>
>> > Content-Type: text/plain; charset="utf-8"
>> >
>> > S�, mira esto...
>> > "plot3D" en la vi�eta te ofrece hasta 50 posibilidades diferentes de
>> > representar un gr�fico 3D...
>> >
>> > https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fcran.r-project.org%2Fweb%2Fpackages%2Fplot3D%2Fvignettes%2Fvolcano.pdf&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=0zMWFCxkj4ghVSiGo%2FpacmPhYOl4oE3IRrJ%2B6hDxXJo%3D&reserved=0
>> >
>> > Saludos,
>> > Carlos Ortega
>> > https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> >
>> > El 19 de febrero de 2018, 17:44, Andr�s Hirigoyen <
>> > andreshirigoyen en gmail.com
>> > > escribi�:
>> >
>> > > Buenas tardes �me recomiendan alg�n paquete para graficar objetos
>> > > geom�tricos en 3D? Alguien ya hizo este tipo de gr�ficos?
>> > >
>> > > Muchas gracias
>> > >
>> > >         [[alternative HTML version deleted]]
>> > >
>> > > _______________________________________________
>> > > R-help-es mailing list
>> > > R-help-es en r-project.org
>> > > https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-help-es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=HGYecrba7vZLzE8xbzcDWE3BX%2BFhL2Uq6lKTDHrmYOw%3D&reserved=0
>> > >
>> >
>> >
>> >
>> > --
>> > Saludos,
>> > Carlos Ortega
>> > https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> >
>> >         [[alternative HTML version deleted]]
>> >
>> >
>> >
>> >
>> > ------------------------------
>> >
>> > Message: 4
>> > Date: Mon, 19 Feb 2018 18:09:47 +0100
>> > From: Carlos Ortega <cof en qualityexcellence.es>
>> > To: Manuel Mendoza <mmendoza en mncn.csic.es>
>> > Cc: "r-help-es en r-project.org" <r-help-es en r-project.org>
>> > Subject: Re: [R-es]  gbm.step para clasificaci�n no binaria
>> > Message-ID:
>> >         <
>> > CAOKbq8gbiJwxxRgbKnj0rbiHZ4g2bAO8oLyQbVEm+3bKCB8mhw en mail.gmail.com>
>> > Content-Type: text/plain; charset="utf-8"
>> >
>> > Hola,
>> >
>> > S�, tienes raz�n...
>> > �No puedes usar la propia funci�n "gbm" del paquete "gbm"?...
>> >
>> > Gracias,
>> > Carlos Ortega
>> > https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> >
>> > El 19 de febrero de 2018, 18:01, Manuel Mendoza <mmendoza en mncn.csic.es>
>> > escribi�:
>> >
>> > >
>> > > Gracias Carlos. Hasta donde yo entiendo si las hay:
>> > >
>> > > El argumento family puede ser:
>> > >
>> > > "gaussian" (for minimizing squared error); por lo que tiene que ser
>> > > num�rica
>> > > "bernoulli" (logistic regression for 0-1 out-comes); binaria por
>> narices
>> > > "poisson" (count outcomes; requires the response to be a positive
>> > > integer); num�rica tambi�n, pues.
>> > >
>> > > La �nica podr�a ser "laplace" (for minimizing absolute loss), pero me
>> da
>> > > este error: Error in while (delta.deviance > tolerance.test &
>> n.fitted <
>> > > max.trees) { :
>> > >   missing value where TRUE/FALSE needed
>> > >
>> > > Supongo que loss se refiere a la funci�n de p�rdida, y como habla de
>> > > deviance (la funci�n de p�rdida por defecto en gbm) pienso que
>> tambi�n es
>> > > para variable respuesta num�rica, aunque no lo encontr� por ning�n
>> lado.
>> > > Por eso pregunt�.
>> > >
>> > > Probar� sin indicar la family, a ver si funciona.
>> > > Gracias,
>> > > Manuel
>> > >
>> > >
>> > >
>> > > Quoting Carlos Ortega <cof en qualityexcellence.es>:
>> > >
>> > > Hola,
>> > >>
>> > >> No hay ninguna limitaci�n en la ayuda de la funci�n en este sentido.
>> > >> Tan solo se indica que han de existir dos niveles en la variable
>> > >> predictora, vaya que al menos sea binaria...
>> > >> En la funci�n en el par�metro "gbm.y" es donde indicas qu� columna
>> es la
>> > >> predictora. No hay otro par�metro donde por otro lado le indiques si
>> es
>> > un
>> > >> modelo binario o multinominal...
>> > >>
>> > >> Saludos,
>> > >> Carlos Ortega
>> > >> https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> > >>
>> > >>
>> > >> 2018-02-19 14:03 GMT+01:00 Manuel Mendoza <mmendoza en mncn.csic.es>:
>> > >>
>> > >> Hola erreros, �sab�is si gbm.step puede usarse para clasificaci�n no
>> > >>> binaria?
>> > >>> Gracias
>> > >>> --
>> > >>> Dr Manuel Mendoza
>> > >>> Department of Biogeography and Global Change
>> > >>> National Museum of Natural History (MNCN)
>> > >>> Spanish Scientific Council (CSIC)
>> > >>> C/ Serrano 115bis, 28006 MADRID
>> > >>> Spain
>> > >>>
>> > >>> _______________________________________________
>> > >>> R-help-es mailing list
>> > >>> R-help-es en r-project.org
>> > >>> https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-help-es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=HGYecrba7vZLzE8xbzcDWE3BX%2BFhL2Uq6lKTDHrmYOw%3D&reserved=0
>> > >>>
>> > >>>
>> > >>
>> > >>
>> > >> --
>> > >> Saludos,
>> > >> Carlos Ortega
>> > >> https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> > >>
>> > >
>> > >
>> > > --
>> > > Dr Manuel Mendoza
>> > > Department of Biogeography and Global Change
>> > > National Museum of Natural History (MNCN)
>> > > Spanish Scientific Council (CSIC)
>> > > C/ Serrano 115bis, 28006 MADRID
>> > > Spain
>> > >
>> > >
>> >
>> >
>> > --
>> > Saludos,
>> > Carlos Ortega
>> > https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> >
>> >         [[alternative HTML version deleted]]
>> >
>> >
>> >
>> >
>> > ------------------------------
>> >
>> > Message: 5
>> > Date: Mon, 19 Feb 2018 18:57:39 +0100
>> > From: Manuel Mendoza <mmendoza en mncn.csic.es>
>> > To: Carlos Ortega <cof en qualityexcellence.es>
>> > Cc: "r-help-es en r-project.org" <r-help-es en r-project.org>
>> > Subject: Re: [R-es]  gbm.step para clasificaci�n no binaria
>> > Message-ID:
>> >         <20180219185739.Horde.AFdT7V8spGDZzyOKkCIZwQ2 en webmail.csic.es>
>> > Content-Type: text/plain; charset="utf-8"; Format="flowed";
>> >         DelSp="Yes"
>> >
>> >
>> > Bueno, primero te comento que si no le indico la family me hace
>> > Bernuilli y da error por no ser binaria.
>> >
>> > La raz�n de aplicar gbm.step del paquete dismo es que da una
>> > informaci�n fundamental, como la interacci�n entre las variables o los
>> > partial plots. La interacci�n te la representa en 3D y es super
>> > explicativo.
>> >
>> > El paquete randomforest tambi�n me da los partialplots y sale muy
>> > bien, pero me gustar�a probar tambi�n con boosted. Lo har� con gbm,
>> > aunque no obtenga las interacciones.
>> >
>> > Gracias una vez m�s,
>> > Manuel
>> >
>> >
>> > Quoting Carlos Ortega <cof en qualityexcellence.es>:
>> >
>> > > Hola,
>> > >
>> > > S�, tienes raz�n...
>> > > �No puedes usar la propia funci�n "gbm" del paquete "gbm"?...
>> > >
>> > > Gracias,
>> > > Carlos Ortega
>> > > https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> > >
>> > > El 19 de febrero de 2018, 18:01, Manuel Mendoza <
>> mmendoza en mncn.csic.es>
>> > > escribi�:
>> > >
>> > >>
>> > >> Gracias Carlos. Hasta donde yo entiendo si las hay:
>> > >>
>> > >> El argumento family puede ser:
>> > >>
>> > >> "gaussian" (for minimizing squared error); por lo que tiene que ser
>> > >> num�rica
>> > >> "bernoulli" (logistic regression for 0-1 out-comes); binaria por
>> narices
>> > >> "poisson" (count outcomes; requires the response to be a positive
>> > >> integer); num�rica tambi�n, pues.
>> > >>
>> > >> La �nica podr�a ser "laplace" (for minimizing absolute loss), pero
>> me da
>> > >> este error: Error in while (delta.deviance > tolerance.test &
>> n.fitted <
>> > >> max.trees) { :
>> > >>   missing value where TRUE/FALSE needed
>> > >>
>> > >> Supongo que loss se refiere a la funci�n de p�rdida, y como habla de
>> > >> deviance (la funci�n de p�rdida por defecto en gbm) pienso que
>> tambi�n
>> > es
>> > >> para variable respuesta num�rica, aunque no lo encontr� por ning�n
>> lado.
>> > >> Por eso pregunt�.
>> > >>
>> > >> Probar� sin indicar la family, a ver si funciona.
>> > >> Gracias,
>> > >> Manuel
>> > >>
>> > >>
>> > >>
>> > >> Quoting Carlos Ortega <cof en qualityexcellence.es>:
>> > >>
>> > >> Hola,
>> > >>>
>> > >>> No hay ninguna limitaci�n en la ayuda de la funci�n en este sentido.
>> > >>> Tan solo se indica que han de existir dos niveles en la variable
>> > >>> predictora, vaya que al menos sea binaria...
>> > >>> En la funci�n en el par�metro "gbm.y" es donde indicas qu� columna
>> es
>> > la
>> > >>> predictora. No hay otro par�metro donde por otro lado le indiques si
>> > es un
>> > >>> modelo binario o multinominal...
>> > >>>
>> > >>> Saludos,
>> > >>> Carlos Ortega
>> > >>> https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> > >>>
>> > >>>
>> > >>> 2018-02-19 14:03 GMT+01:00 Manuel Mendoza <mmendoza en mncn.csic.es>:
>> > >>>
>> > >>> Hola erreros, �sab�is si gbm.step puede usarse para clasificaci�n no
>> > >>>> binaria?
>> > >>>> Gracias
>> > >>>> --
>> > >>>> Dr Manuel Mendoza
>> > >>>> Department of Biogeography and Global Change
>> > >>>> National Museum of Natural History (MNCN)
>> > >>>> Spanish Scientific Council (CSIC)
>> > >>>> C/ Serrano 115bis, 28006 MADRID
>> > >>>> Spain
>> > >>>>
>> > >>>> _______________________________________________
>> > >>>> R-help-es mailing list
>> > >>>> R-help-es en r-project.org
>> > >>>> https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-help-es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=HGYecrba7vZLzE8xbzcDWE3BX%2BFhL2Uq6lKTDHrmYOw%3D&reserved=0
>> > >>>>
>> > >>>>
>> > >>>
>> > >>>
>> > >>> --
>> > >>> Saludos,
>> > >>> Carlos Ortega
>> > >>> https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> > >>>
>> > >>
>> > >>
>> > >> --
>> > >> Dr Manuel Mendoza
>> > >> Department of Biogeography and Global Change
>> > >> National Museum of Natural History (MNCN)
>> > >> Spanish Scientific Council (CSIC)
>> > >> C/ Serrano 115bis, 28006 MADRID
>> > >> Spain
>> > >>
>> > >>
>> > >
>> > >
>> > > --
>> > > Saludos,
>> > > Carlos Ortega
>> > > https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>> >
>> >
>> > --
>> > Dr Manuel Mendoza
>> > Department of Biogeography and Global Change
>> > National Museum of Natural History (MNCN)
>> > Spanish Scientific Council (CSIC)
>> > C/ Serrano 115bis, 28006 MADRID
>> > Spain
>> >
>> >
>> >
>> >
>> > ------------------------------
>> >
>> > Subject: Pi� de p�gina del digest
>> >
>> > _______________________________________________
>> > R-help-es mailing list
>> > R-help-es en r-project.org
>> > https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-help-es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=HGYecrba7vZLzE8xbzcDWE3BX%2BFhL2Uq6lKTDHrmYOw%3D&reserved=0
>> >
>> >
>> > ------------------------------
>> >
>> > Fin de Resumen de R-help-es, Vol 108, Env�o 30
>> > **********************************************
>> >
>>
>>         [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-help-es mailing list
>> R-help-es en r-project.org
>> https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-help-es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=HGYecrba7vZLzE8xbzcDWE3BX%2BFhL2Uq6lKTDHrmYOw%3D&reserved=0
>>
>
>
>
> --
> Saludos,
> Carlos Ortega
> https://nam02.safelinks.protection.outlook.com/?url=www.qualityexcellence.es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=eFpfPaoll6t1iBl5rZeN2XBBz5M%2FvI6el7puiIij7bE%3D&reserved=0
>

        [[alternative HTML version deleted]]

_______________________________________________
R-help-es mailing list
R-help-es en r-project.org
https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-help-es&data=02%7C01%7C%7C784dc452db1548f8542d08d577de3fb8%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636546718785681990&sdata=HGYecrba7vZLzE8xbzcDWE3BX%2BFhL2Uq6lKTDHrmYOw%3D&reserved=0


	[[alternative HTML version deleted]]

Más información sobre la lista de distribución R-help-es