version 0.1.1

DESCRIPTION

@@ -1,15 +1,16 @@
 Package: maxnet
 Type: Package
 Title: Fitting 'Maxent' Species Distribution Models with 'glmnet'
-Version: 0.1.0
-Date: 2016-11-12
+Version: 0.1.1
+Date: 2017-02-03
 Author: Steven Phillips
 Maintainer: Steven Phillips <mrmaxent@gmail.com>
 Imports: glmnet
 Description: Procedures to fit species distributions models from occurrence records and environmental variables, using 'glmnet' for model fitting. Model structure is the same as for the 'Maxent' Java package, version 3.4.0, with the same feature types and regularization options.  See the 'Maxent' website <http://biodiversityinformatics.amnh.org/open_source/maxent> for more details.
 License: MIT + file LICENSE
+URL: https://github.com/mrmaxent/maxnet
 RoxygenNote: 5.0.1
 NeedsCompilation: no
-Packaged: 2016-11-16 02:23:35 UTC; phillips
+Packaged: 2017-02-03 15:51:37 UTC; phillips
 Repository: CRAN
-Date/Publication: 2016-11-16 08:16:39
+Date/Publication: 2017-02-03 19:18:39

MD5

@@ -1,4 +1,4 @@
-0d2ea99c3ce6c0cfeaee90524801d6c2 *DESCRIPTION
+2b837427b32ae30ad903f6423588e7f2 *DESCRIPTION
 647c3da237fb33ee8bfa84337d7250e9 *LICENSE
 3e3a91a387434cae65abcff42cba04ac *NAMESPACE
 dc4580faf21895654550fce43c932296 *NEWS.md
@@ -7,7 +7,7 @@ e68f571f67610cd3314e700a32d8afac *R/categoricalval.R
 bef5dc4e35646907f004387a26f4a737 *R/data.R
 910564f66b072a8e09dbb0597c398055 *R/hinge.R
 2902c6ec58f78af1fd12ec83e6db080c *R/hingeval.R
-7fbf6fabe232baa1923bead33167694f *R/maxnet.R
+dc5666869a5fac5eca3bbc4cbb67e6fc *R/maxnet.R
 6eb6cbbb5689130f5e94010fa026cdd1 *R/maxnet.default.regularization.R
 05aa6f3586d566ccc842247e2d67513d *R/maxnet.formula.R
 3dca2d0753cf1b9e0486f7e8c17f1659 *R/plot.maxnet.R

R/maxnet.R

@@ -8,24 +8,25 @@ function(p, data, f=maxnet.formula(p, data), regmult=1.0,
    reg <- regfun(p,mm) * regmult
    weights <- p+(1-p)*100
    glmnet::glmnet.control(pmin=1.0e-8, fdev=0)  
-   model <- glmnet::glmnet(x=mm, y=as.factor(p), family="binomial", standardize=F, penalty.factor=reg, lambda=10^(seq(4,0,length.out=100))*sum(reg)/length(reg)*sum(p)/sum(weights), weights=weights, ...)
+   model <- glmnet::glmnet(x=mm, y=as.factor(p), family="binomial", standardize=F, penalty.factor=reg, lambda=10^(seq(4,0,length.out=200))*sum(reg)/length(reg)*sum(p)/sum(weights), weights=weights, ...)
    class(model) <- c("maxnet", class(model))
-   bb <- model$beta[,100]
+   if (length(model$beta) < 200) stop("Error: glmnet failed to complete regularization path")
+   bb <- model$beta[,200]
    model$betas <- bb[bb!=0]
    model$alpha <- 0
-   rr <- predict.maxnet(model, data[p==0,], type="exponent", clamp=F)
+   rr <- predict.maxnet(model, data[p==0, , drop = FALSE], type="exponent", clamp=F)
    raw <- rr / sum(rr)
    model$entropy <- -sum(raw * log(raw))
    model$alpha <- -log(sum(rr))
    model$penalty.factor <- reg
    model$featuremins <- apply(mm, 2, min)
    model$featuremaxs <- apply(mm, 2, max)
    vv <- (sapply(data, class)!="factor")
-   model$varmin <- apply(data[,vv], 2, min)
-   model$varmax <- apply(data[,vv], 2, max)
-   means <- apply(data[p==1,vv], 2, mean)
+   model$varmin <- apply(data[,vv, drop = FALSE], 2, min)
+   model$varmax <- apply(data[,vv, drop = FALSE], 2, max)
+   means <- apply(data[p==1,vv, drop = FALSE], 2, mean)
    majorities <- sapply(names(data)[!vv], 
-      function(n) which.max(table(data[p==1,n])))
+      function(n) which.max(table(data[p==1,n, drop = FALSE])))
    names(majorities) <- names(data)[!vv]
    model$samplemeans <- c(means, majorities)
    model$levels <- lapply(data, levels)