<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <p>Obrigado novamente Fernando,</p>
    <p>Mas se eu rodo um modelo para o subgrupo Feature ==A  e um modelo
      para o subgrupo Featura==B e obtenho os parametros do modelo a1 e
      a2, e b1 e b2 respectivamente. Eu vou utilizar esses parâmetros
      para ajustar esses modelo novamente sobre os mesmos dados 999? Faz
      sentido isso? Acredito que você queira utilizar esses parâmetros
      como valores iniciais  para outro conjunto de dados. Acredito que 
      afunção nlsLM conte o número de interação para obter o melhor
      ajuste e limita um valor máximo para nao ficar rodando
      indefinidamente. Esta contagem é crescente não? Desculpe a
      pergunta mas estou tentando enteder a questão para propor um
      solução</p>
    <p>Resposta:<br>
    </p>
    <p>São valores iniciais para o mesmo conjunto de dados o que eu 
      quero, porque a função nls() faz se ajustada uma única vez:</p>
    <p>1) Uso um intervalo de valores iniciais  para chegar perto dos
      coeficientes ideais (no exemplo abaixo sem considerar o subset):</p>
    <p><span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;"><span> </span>mod1<-nlsLM(e1,
        data = d,<span> </span></span><br style="color: rgb(0, 0, 0);
        font-family: Verdana, Geneva, Helvetica, Arial, sans-serif;
        font-size: 13.44px; font-style: normal; font-variant-ligatures:
        normal; font-variant-caps: normal; font-weight: 400;
        letter-spacing: normal; orphans: 2; text-align: start;
        text-indent: 0px; text-transform: none; white-space: normal;
        widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px;
        background-color: rgb(255, 255, 255); text-decoration-style:
        initial; text-decoration-color: initial;">
      <span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;">      start =
        list(a1 = 0.1, a2 = 10),<span> </span></span><br style="color:
        rgb(0, 0, 0); font-family: Verdana, Geneva, Helvetica, Arial,
        sans-serif; font-size: 13.44px; font-style: normal;
        font-variant-ligatures: normal; font-variant-caps: normal;
        font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial;">
      <span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;">      control
        = nls.control(maxiter = 1000))</span></p>
    <pre tabindex="0" class="GNKRCKGCGSB" id="rstudio_console_output" style="font-family: 'Lucida Console'; font-size: 10pt !important; outline: none; border: none; word-break: break-all; margin: 0px; -webkit-user-select: text; white-space: pre-wrap !important; line-height: 15px; color: rgb(0, 0, 0); font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: -webkit-left; text-indent: 0px; text-transform: none; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; background-color: rgb(255, 255, 255);"><span><span class="GNKRCKGCGSB" style="outline: none; border: none; word-break: break-all; margin: 0px; -webkit-user-select: text; white-space: pre-wrap !important;">Nonlinear regression model
  model: Diameter ~ a1 * Age^a2
   data: d
       a1        a2 
5.845e-08 4.283e+00 
 residual sum-of-squares: 1913

Number of iterations till stop: 100 
Achieved convergence tolerance: 1.49e-08
Reason stopped: Number of calls to `fcn' has reached or exceeded `maxfev' == 300</span></span></pre>
    <p>2) Depois utilizo os coeficientes do ajuste anterior novamente:</p>
    <p><span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;"><span> </span>mod2<-nlsLM(e1,
        data = d,<span> </span></span><br style="color: rgb(0, 0, 0);
        font-family: Verdana, Geneva, Helvetica, Arial, sans-serif;
        font-size: 13.44px; font-style: normal; font-variant-ligatures:
        normal; font-variant-caps: normal; font-weight: 400;
        letter-spacing: normal; orphans: 2; text-align: start;
        text-indent: 0px; text-transform: none; white-space: normal;
        widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px;
        background-color: rgb(255, 255, 255); text-decoration-style:
        initial; text-decoration-color: initial;">
      <span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;">      start =
        list(a1 = 5.845e-08, a2 = 4.283e+000),<span> </span></span><br
        style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial;">
      <span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;">      control
        = nls.control(maxiter = 1000))</span></p>
    <p><br>
    </p>
    <pre tabindex="0" class="GNKRCKGCGSB" id="rstudio_console_output" style="font-family: 'Lucida Console'; font-size: 10pt !important; outline: none; border: none; word-break: break-all; margin: 0px; -webkit-user-select: text; white-space: pre-wrap !important; line-height: 15px; color: rgb(0, 0, 0); font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: -webkit-left; text-indent: 0px; text-transform: none; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; background-color: rgb(255, 255, 255);"><span><span class="GNKRCKGCGSB" style="outline: none; border: none; word-break: break-all; margin: 0px; -webkit-user-select: text; white-space: pre-wrap !important;">Nonlinear regression model
  model: Diameter ~ a1 * Age^a2
   data: d
    a1     a2 
0.3357 0.8376 
 residual sum-of-squares: 271

Number of iterations to convergence: 54 
Achieved convergence tolerance: 1.49e-08</span></span></pre>
    <p><span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;"></span></p>
    <p>3) Novamente:</p>
    <p><span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;"><span> </span>mod3<-nlsLM(e1,
        data = d,<span> </span></span><br style="color: rgb(0, 0, 0);
        font-family: Verdana, Geneva, Helvetica, Arial, sans-serif;
        font-size: 13.44px; font-style: normal; font-variant-ligatures:
        normal; font-variant-caps: normal; font-weight: 400;
        letter-spacing: normal; orphans: 2; text-align: start;
        text-indent: 0px; text-transform: none; white-space: normal;
        widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px;
        background-color: rgb(255, 255, 255); text-decoration-style:
        initial; text-decoration-color: initial;">
      <span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;">      start =
        list(a1 = 0.3357, a2 =0.8376),<span> </span></span><br
        style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial;">
      <span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;">      control
        = nls.control(maxiter = 1000))</span></p>
    <pre tabindex="0" class="GNKRCKGCGSB" id="rstudio_console_output" style="font-family: 'Lucida Console'; font-size: 10pt !important; outline: none; border: none; word-break: break-all; margin: 0px; -webkit-user-select: text; white-space: pre-wrap !important; line-height: 15px; color: rgb(0, 0, 0); font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: -webkit-left; text-indent: 0px; text-transform: none; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; background-color: rgb(255, 255, 255);"><span><span class="GNKRCKGCGSB" style="outline: none; border: none; word-break: break-all; margin: 0px; -webkit-user-select: text; white-space: pre-wrap !important;">Nonlinear regression model
  model: Diameter ~ a1 * Age^a2
   data: d
    a1     a2 
0.3357 0.8376 
 residual sum-of-squares: 271

Number of iterations to convergence: 2 
Achieved convergence tolerance: 1.49e-08</span></span></pre>
    <p><span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;"></span></p>
    <p>4) Deu no exemplo acima 2 iteraçõe. Quero repetir até 1 iteração
      para ter uma ideia da estabilidade dos coeficientes!!!</p>
    <p><br>
    </p>
    <p><br>
    </p>
    <p><br>
    </p>
    <p><br>
    </p>
    <p><span style="color: rgb(0, 0, 0); font-family: Verdana, Geneva,
        Helvetica, Arial, sans-serif; font-size: 13.44px; font-style:
        normal; font-variant-ligatures: normal; font-variant-caps:
        normal; font-weight: 400; letter-spacing: normal; orphans: 2;
        text-align: start; text-indent: 0px; text-transform: none;
        white-space: normal; widows: 2; word-spacing: 0px;
        -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
        255); text-decoration-style: initial; text-decoration-color:
        initial; display: inline !important; float: none;"><span></span><br>
        <span></span></span><span style="color: rgb(0, 0, 0);
        font-family: Verdana, Geneva, Helvetica, Arial, sans-serif;
        font-size: 13.44px; font-style: normal; font-variant-ligatures:
        normal; font-variant-caps: normal; font-weight: 400;
        letter-spacing: normal; orphans: 2; text-align: start;
        text-indent: 0px; text-transform: none; white-space: normal;
        widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px;
        background-color: rgb(255, 255, 255); text-decoration-style:
        initial; text-decoration-color: initial; display: inline
        !important; float: none;"></span></p>
    <p><br>
    </p>
    <pre class="moz-signature" cols="72">-- 
======================================================================
Alexandre dos Santos
Proteção Florestal 
IFMT - Instituto Federal de Educação, Ciência e Tecnologia de Mato Grosso
Campus Cáceres
Caixa Postal 244
Avenida dos Ramires, s/n
Bairro: Distrito Industrial 
Cáceres - MT                      CEP: 78.200-000
Fone: (+55) 65 99686-6970 (VIVO) (+55) 65 3221-2674 (FIXO)
<a class="moz-txt-link-abbreviated" href="mailto:e-mails:alexandresantosbr@yahoo.com.br">e-mails:alexandresantosbr@yahoo.com.br</a> 
        <a class="moz-txt-link-abbreviated" href="mailto:alexandre.santos@cas.ifmt.edu.br">alexandre.santos@cas.ifmt.edu.br</a> 
Lattes: <a class="moz-txt-link-freetext" href="http://lattes.cnpq.br/1360403201088680">http://lattes.cnpq.br/1360403201088680</a> 
OrcID: orcid.org/0000-0001-8232-6722   -   ResearcherID: A-5790-2016
Researchgate: <a class="moz-txt-link-abbreviated" href="http://www.researchgate.net/profile/Alexandre_Santos10">www.researchgate.net/profile/Alexandre_Santos10</a>                       
LinkedIn: br.linkedin.com/in/alexandre-dos-santos-87961635
Mendeley:www.mendeley.com/profiles/alexandre-dos-santos6/
======================================================================</pre>
    <div class="moz-cite-prefix">Em 25/04/2019 13:22, Fernando Souza
      escreveu:<br>
    </div>
    <blockquote type="cite"
      cite="mid:522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div>Veja se entendi bem.</div>
      <div>Mas se eu rodo um modelo para o subgrupo Feature ==A  e um
        modelo para o subgrupo Featura==B e obtenho os parametros do
        modelo a1 e a2, e b1 e b2 respectivamente. Eu vou utilizar esses
        parâmetros para ajustar esses modelo novamente sobre os mesmos
        dados 999? Faz sentido isso? Acredito que você queira utilizar
        esses parâmetros como valores iniciais  para outro conjunto de
        dados.</div>
      <div>Os parametros a1 e a2 do modelo Feature==A  e b1 e b2 do
        modelo Feature ==B são os melhores parametros do ajuste do
        modelos , ficar reajustando o modelo com esses parâmetros  como
        valores iniciais, para o mesmo banco de dados, não alterarar o
        modelo em nada.</div>
      <br>
      <div>"<span style="color:rgb(35, 31, 32)"><font
            style="font-size:14.5px"><font style="font-family:Nylas-Pro,
              Helvetica, "Lucidia Grande", sans-serif">Sim
              isso não consegui colocar como condição, mas eu queria
              para quando a iteração para os dois modelos fosse igual a
              1, ou n modelos se houverem."</font></font></span></div>
      <div><span style="color:rgb(35, 31, 32)"><font
            style="font-size:14.5px"><font style="font-family:Nylas-Pro,
              Helvetica, "Lucidia Grande", sans-serif">Acredito
              que  afunção nlsLM conte o número de interação para obter
              o melhor ajuste e limita um valor máximo para nao ficar
              rodando indefinidamente. Esta contagem é crescente não? </font></font></span></div>
      <div><span style="color:rgb(35, 31, 32)"><font
            style="font-size:14.5px"><font style="font-family:Nylas-Pro,
              Helvetica, "Lucidia Grande", sans-serif">Desculpe
              a pergunta mas estou tentando enteder a questão para
              propor um solução</font></font></span></div>
      <br>
      <div class="gmail_quote_attribution">On Apr 25 2019, at 12:46 pm,
        ASANTOS <a class="moz-txt-link-rfc2396E" href="mailto:alexandre.santos@cas.ifmt.edu.br"><alexandre.santos@cas.ifmt.edu.br></a> wrote:</div>
      <blockquote>
        <div><br>
          <div>Obrigado pelo feedback Fernando,</div>
          <br>
          <div>1 -  O loop de 999 vezes é para ajustar 999 modelos
            diferentes? Que parâmetro você altera em cada loop.</div>
          <br>
          <div>O loop em 999 é para a cada vez que é realizado um
            ajuste, extrair os novos coeficientes coef(mod_ND[[i]])[1] e
            coef(mod_ND[[i]])[2] e os coloca no novo ajuste e repete
            isso n vezes até que a iteração de algum ajuste seja =1.
            Então necessariamente não precisa rodar 999 vezes, coloquei
            um número alto, mas imagino que a iteração igual a 1 venha a
            surgir antes.</div>
          <br>
          <div>2- "porém a cada vez que realiza o loop eu gostaria de
            reciclar os start values através da modificação em
            coef(mod_ND[[i]])[1]  e  coef(mod_ND[[i]])[2]." .  O que
            você quer dizer com  "Reciclar" --> modificar para qual
            valores?</div>
          <br>
          <div>Reciclar é a cada novo ajuste utilizar como start values
            os coef(mod_ND[[i]])[1] e coef(mod_ND[[i]])[2] do ajuste
            anterior, a menos que a iteração seja igual a 1.</div>
          <br>
          <div>4-  Explique melhor as interações: A condição para o loop
            parar é quando a iteração for igual a 1 para os dois modelos
            (  if (mod_ND[[z,c(finIter")]] <= ){ break } ## ).</div>
          <br>
          <div>Sim isso não consegui colocar como condição, mas eu
            queria para quando a iteração para os dois modelos fosse
            igual a 1, ou n modelos se houverem.</div>
          <br>
          <div>Baseada na sua resposta, que também eu não consegui ainda
            fazer funcionar, seria:</div>
          <br>
          <div># #Realizando os ajustes</div>
          <div>
            <div>for(z in 1:999){</div>
          </div>
          <br>
          <div>modelo<-function(dados){</div>
          <div>
            <div>  </div>
            <div>    mod_ND[[z]] < -summary(nlsLM(Diameter ~ a1 *
              Age^a2,start = list(a1 = 0.1, a2 = 10), control =
              nls.control(maxiter = 1000), data = dados))</div>
            <div>  </div>
            <div>
              <div>   if (mod_ND[[z,c("finIter")]] <= 1){ break }</div>
            </div>
            <div>}</div>
          </div>
          <div>}</div>
          <div>list_modelos <- dlply(d,.(Feature),modelo)</div>
          <br>
          <div>list_modelos</div>
          <br>
          <br>
          <div><br>
          </div>
          <br>
          <div><br>
          </div>
          <br>
          <div><br>
          </div>
          <br>
          <code>
            <pre style="background-color:rgba(0,0,0,0.05);padding:0.2em 1em">-- 
======================================================================
Alexandre dos Santos
Proteção Florestal 
IFMT - Instituto Federal de Educação, Ciência e Tecnologia de Mato Grosso
Campus Cáceres
Caixa Postal 244
Avenida dos Ramires, s/n
Bairro: Distrito Industrial 
Cáceres - MT                      CEP: 78.200-000
Fone: (+55) 65 99686-6970 (VIVO) (+55) 65 3221-2674 (FIXO)
<a href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/0?redirect=mailto%3Ae-mails%3Aalexandresantosbr%40yahoo.com.br&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D" title="mailto:e-mails:alexandresantosbr@yahoo.com.br" moz-do-not-send="true">e-mails:alexandresantosbr@yahoo.com.br</a> 
        <a href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/1?redirect=mailto%3Aalexandre.santos%40cas.ifmt.edu.br&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D" title="mailto:alexandre.santos@cas.ifmt.edu.br" moz-do-not-send="true">alexandre.santos@cas.ifmt.edu.br</a> 
Lattes: <a href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/2?redirect=http%3A%2F%2Flattes.cnpq.br%2F1360403201088680&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D" title="http://lattes.cnpq.br/1360403201088680" moz-do-not-send="true">http://lattes.cnpq.br/1360403201088680</a> 
OrcID: orcid.org/0000-0001-8232-6722   -   ResearcherID: A-5790-2016
Researchgate: <a href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/3?redirect=http%3A%2F%2Fwww.researchgate.net%2Fprofile%2FAlexandre_Santos10&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D" title="http://www.researchgate.net/profile/Alexandre_Santos10" moz-do-not-send="true">www.researchgate.net/profile/Alexandre_Santos10</a>                       
LinkedIn: br.linkedin.com/in/alexandre-dos-santos-87961635
Mendeley:www.mendeley.com/profiles/alexandre-dos-santos6/
======================================================================</pre>
          </code>
          <div class="moz-cite-prefix">
            <div>Em 25/04/2019 11:13, Fernando Souza escreveu:</div>
          </div>
          <blockquote><br>
            <div>Alexandre , Algumas de suas explicações não estão muito
              claras para mim.</div>
            <br>
            <div>1 -  O loop de 999 vezes é para ajustar 999 modelos
              diferentes? Que parametro você altera em cada loop.</div>
            <br>
            <div>2- "<span style="color:rgb(35, 31, 32)"><font
                  style="font-size:14.5px"><font
                    style="font-family:Nylas-Pro, Helvetica,
                    "Lucidia Grande", sans-serif">porém a cada
                    vez que realiza o loop eu gostaria de reciclar os
                    start values através da modificação em
                    coef(mod_ND[[i]])[1]  e  coef(mod_ND[[i]])[2]." .  O
                    que você quer dizer com  "Reciclar" --> modificar
                    para qual valores?</font></font></span></div>
            <br>
            <div>4-  Explique melhor as interações: <span
                style="color:rgb(35, 31, 32)"><font
                  style="font-size:14.5px">A condição para o loop parar
                  é quando a iteração for igual a 1 para os dois modelos
                  (  if (mod_ND[[z,c(finIter")]] <= ){ break } ## ).</font></span></div>
            <br>
            <br>
            <div>Acredito que a solução final para seu problema passe
              pela seguinte abordagem. No entanto preciso de mais
              esclarecimentos para melhorar.</div>
            <br>
            <div>library(minpack.lm)</div>
            <div>library(plyr)</div>
            <br>
            <div>##Banco de dados</div>
            <br>
            <div>Feature<-sort(rep(c("A","B"),22))</div>
            <div>Age<-c(60,72,88,96,27,36,48,60,72,88,96,27,36,48,60,72,88,96,27,36,48,60,27,27,36,48,60,72,88,96,27,36,48,60,72,88,96,27,36,48,60,72,88,96)</div>
            <div>Diameter<-c(13.9,16.2,19.1,19.3,4.7,6.7,9.6,11.2,13.1,15.3,15.4,5.4,7,9.9,11.7,13.4,16.1,16.2,5.9,8.3,12.3,14.5,2.3,5.2,6.2,8.6,9.3,11.3,15.1,15.5,5,7,7.9,8.4,10.5,14,14,4.1,4.9,6,6.7,7.7,8,8.2)</div>
            <div>d<-dados <- data.frame(Feature,Age,Diameter)</div>
            <div>str(d)</div>
            <br>
            <div># #Realizando os ajustes</div>
            <br>
            <div>modelo<-function(dados){</div>
            <br>
            <br>
            <div>}</div>
            <br>
            <div>list_modelos <- dlply(d,.(Feature),modelo)</div>
            <br>
            <div>list_modelos</div>
            <br>
            <div class="gmail_quote_attribution">On Apr 25 2019, at
              10:53 am, ASANTOS por (R-br) <a
href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/4?redirect=mailto%3Ar-br%40listas.c3sl.ufpr.br&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
                title="mailto:r-br@listas.c3sl.ufpr.br"
                moz-do-not-send="true"><r-br@listas.c3sl.ufpr.br></a>
              wrote:</div>
            <blockquote>
              <div>
                <div>Prezados Membros,</div>
                <br>
                <div>    Gostaria de fazer um loop (999 vezes) para
                  ajustar dois modelos nls</div>
                <div>(Diameter ~ a1 * Age^a2) à partir de um subset em
                  função da minha</div>
                <div>variável Feature, porém a cada vez que realiza o
                  loop eu gostaria de</div>
                <div>reciclar os start values através da modificação em
                  coef(mod_ND[[i]])[1]</div>
                <div>e  coef(mod_ND[[i]])[2]. A condição para o loop
                  parar é quando a</div>
                <div>iteração for igual a 1 para os dois modelos (  if</div>
                <div>(mod_ND[[z,c(finIter")]] <= 1){ break } ## ).
                  Bom, mas infelizmente não</div>
                <div>esta funcionando não, alguma ideia? Segue CRM:</div>
                <br>
                <div>#Pacote</div>
                <div>library(minpack.lm)</div>
                <br>
                <div># Meu banco de dados onde tenho o Diameter das
                  plantas em função de</div>
                <div>Feature e Age.</div>
                <div>Feature<-sort(rep(c("A","B"),22))</div>
                <div>Age<-c(60,72,88,96,27,</div>
                <div>36,48,60,72,88,96,27,36,48,60,72,</div>
                <div>88,96,27,36,48,60,27,27,36,48,60,</div>
                <div>72,88,96,27,36,48,60,72,88,96,27,</div>
                <div>36,48,60,72,88,96)</div>
                <div>Diameter<-c(13.9,16.2,</div>
                <div>19.1,19.3,4.7,6.7,9.6,11.2,13.1,15.3,</div>
                <div>15.4,5.4,7,9.9,11.7,13.4,16.1,16.2,</div>
                <div>5.9,8.3,12.3,14.5,2.3,5.2,6.2,8.6,9.3,</div>
                <div>11.3,15.1,15.5,5,7,7.9,8.4,10.5,14,14,</div>
                <div>4.1,4.9,6,6.7,7.7,8,8.2)</div>
                <div>d<-dados <- data.frame(Feature,Age,Diameter)</div>
                <div>str(d)</div>
                <br>
                <div>#Realizando os ajustes</div>
                <br>
                <div>e1<- Diameter ~ a1 * Age^a2</div>
                <div>Fecture_vec<-unique(d$Feature)</div>
                <div>mod_ND <- list() #List para salvar cada modelo</div>
                <div>for(i in 1:length(Fecture_vec)){</div>
                <div>     d2 <- subset(d, d$Feature ==
                  Fecture_vec[i])</div>
                <div>     mod_ND[[i]] <-  nlsLM(e1, data = d2,</div>
                <div>     start = list(a1 = 0.1, a2 = 10),</div>
                <div>     control = nls.control(maxiter = 1000))</div>
                <div>Xs<-data.frame()</div>
                <div>for(z in 1:999){</div>
                <div>     d2 <- subset(d, d$Feature ==
                  Fecture_vec[i])</div>
                <div>     mod_ND[[z]] <-  nlsLM(e1, data = d2,</div>
                <div>     start = list(a1 = coef(mod_ND[[i]])[1], a2 =
                  mod_ND[[i]])[2]),</div>
                <div>     control = nls.control(maxiter = 1000))</div>
                <div>   if (mod_ND[[z,c(finIter")]] <= 1){ break } ##
                  Só para quando</div>
                <div>iteração = 1</div>
                <div>print(summary(mod_ND[[z]]))</div>
                <div>}</div>
                <div>}</div>
                <div>#</div>
                <br>
                <div>Obrigado</div>
                <br>
                <br>
                <br>
                <div>--</div>
                <div>======================================================================</div>
                <div>Alexandre dos Santos</div>
                <div>Proteção Florestal</div>
                <div>IFMT - Instituto Federal de Educação, Ciência e
                  Tecnologia de Mato Grosso</div>
                <div>Campus Cáceres</div>
                <div>Caixa Postal 244</div>
                <div>Avenida dos Ramires, s/n</div>
                <div>Bairro: Distrito Industrial</div>
                <div>Cáceres - MT CEP: 78.200-000</div>
                <div>Fone: (+55) 65 99686-6970 (VIVO) (+55) 65 3221-2674
                  (FIXO)</div>
                <div><a
href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/5?redirect=mailto%3Ae-mails%3Aalexandresantosbr%40yahoo.com.br&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
title="mailto:e-mails:alexandresantosbr@yahoo.com.br"
                    moz-do-not-send="true">e-mails:alexandresantosbr@yahoo.com.br</a></div>
                <div><a
href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/6?redirect=mailto%3Aalexandre.santos%40cas.ifmt.edu.br&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
                    title="mailto:alexandre.santos@cas.ifmt.edu.br"
                    moz-do-not-send="true">alexandre.santos@cas.ifmt.edu.br</a></div>
                <div>Lattes: <a
href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/7?redirect=http%3A%2F%2Flattes.cnpq.br%2F1360403201088680&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
                    title="http://lattes.cnpq.br/1360403201088680"
                    moz-do-not-send="true">http://lattes.cnpq.br/1360403201088680</a></div>
                <div>OrcID: orcid.org/0000-0001-8232-6722 -
                  ResearcherID: A-5790-2016</div>
                <div>Researchgate: <a
href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/8?redirect=http%3A%2F%2Fwww.researchgate.net%2Fprofile%2FAlexandre_Santos10&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
title="http://www.researchgate.net/profile/Alexandre_Santos10"
                    moz-do-not-send="true">www.researchgate.net/profile/Alexandre_Santos10</a></div>
                <div>LinkedIn:
                  br.linkedin.com/in/alexandre-dos-santos-87961635</div>
                <div>Mendeley:www.mendeley.com/profiles/alexandre-dos-santos6/</div>
                <div>======================================================================</div>
                <br>
                <div>_______________________________________________</div>
                <div>R-br mailing list</div>
                <div><a
href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/9?redirect=mailto%3AR-br%40listas.c3sl.ufpr.br&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
                    title="mailto:R-br@listas.c3sl.ufpr.br"
                    moz-do-not-send="true">R-br@listas.c3sl.ufpr.br</a></div>
                <div><a
href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/10?redirect=https%3A%2F%2Flistas.inf.ufpr.br%2Fcgi-bin%2Fmailman%2Flistinfo%2Fr-br&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
title="https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br"
                    moz-do-not-send="true">https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br</a></div>
                <div>Leia o guia de postagem (<a
href="https://link.getmailspring.com/link/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com/11?redirect=http%3A%2F%2Fwww.leg.ufpr.br%2Fr-br-guia&recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
                    title="http://www.leg.ufpr.br/r-br-guia"
                    moz-do-not-send="true">http://www.leg.ufpr.br/r-br-guia</a>)
                  e fornea cdigo mnimo reproduzvel.</div>
              </div>
            </blockquote>
            <div><img class="mailspring-open" alt="Sent from Mailspring"
                style="border:0; width:0; height:0;"
src="https://link.getmailspring.com/open/236E0160-0BAC-45B9-8669-BBB5A4D1EFEB@getmailspring.com?recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
                moz-do-not-send="true" width="0" height="0"></div>
          </blockquote>
        </div>
      </blockquote>
      <img class="mailspring-open" alt="Sent from Mailspring"
        style="border:0; width:0; height:0;"
src="https://link.getmailspring.com/open/522C999C-97DC-4281-AAE7-236E9F648D2E@getmailspring.com?recipient=YWxleGFuZHJlLnNhbnRvc0BjYXMuaWZtdC5lZHUuYnI%3D"
        moz-do-not-send="true" width="0" height="0">
    </blockquote>
  </body>
</html>