Skip to content

Commit

Permalink
modifica a estrutura de pastas do repositorio
Browse files Browse the repository at this point in the history
  • Loading branch information
@AndreaSanchezTapia
AndreaSanchezTapia committed Aug 17, 2019
1 parent 3c740ea commit baebaf2
Show file tree
Hide file tree
Showing 6 changed files with 44 additions and 374 deletions.
3 changes: 3 additions & 0 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -14,3 +14,6 @@ buffer_and_randomPoints.Rmd
Euclidean_distance.Rmd
small_dataset.Rmd
- still not updated to work in the repo

?dist

82 changes: 39 additions & 43 deletions buffer_and_randomPoints/buffer_and_randomPoints.Rmd
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,27 +13,24 @@ output:
knitr::opts_chunk$set(echo = TRUE, message = F, warning = T)
```

This workflow tests background point generation in `Model-R`. We perform tests with different types of buffer and different code options to sample pseudoabsences inside a geographic buffer. Later, we explore how different methods for sampling pseudoabsences results on different model predictions.
This workflow tests background point generation in `modleR`. We perform tests with different types of buffer and different code options to sample pseudoabsences inside a geographic buffer. Later, we explore how different methods for sampling pseudoabsences results on different model predictions.

To run this example you will need `modleR`and the additional packages `rJava`, `raster`, and `dplyr`. To check if they are already installed and install eventually missing packages run the code bellow.
To run this example you will need `modleR`and the additional packages `rJava`, `raster`, and `dplyr`. To check if they are already installed and install eventually missing packages run the code below.

```{r, message=FALSE, eval=FALSE}
packages <- c("rJava", "raster", "dplyr", "devtools")
instpack <- packages[!packages %in% installed.packages()]
if(length(instpack)>0)
{
if(length(instpack)>0) {
install.packages(packages[!packages %in% installed.packages()])
}
```

If you don't have `modleR` installed, run:

```{r, eval=F}
devtools::install_github("Model-R/modleR", ref="master")
devtools::install_github("Model-R/modleR", ref = "master")
```



Then, load all required packages.

```{r load}
Expand All @@ -48,7 +45,7 @@ devtools::load_all("../../modleR")

# The example data set

We use a standard dataset inside the package `modleR`. First, from `coordenadas` object we select only data from one species *Eugenia florida* and create one training set (70% of the data) and one test set (30% of the data) for the data.
We use a standard dataset inside the package `modleR`. First, from `coordenadas` object we select only data from one species *Abarema langsdorffii* and create one training set (70% of the data) and one test set (30% of the data) for the data.

````{r dataset}
## Creating an object with species names
Expand Down Expand Up @@ -93,13 +90,13 @@ First, directly from `randomPoints`.
rand <- dismo::randomPoints(predictor,
n = 500,
excludep = T,
p = coord1sp)
p = pts)
# ploting environmental layer with background values
plot(predictor, legend = F, main = "nobuffer randomPoints")
points(rand, col = "red")
```
And creating an object with `create_buffer` to be used in `randomPoints` function. We will use this code structure when generating background values with different types of buffer.
And creating an object with `create_buffer()` to be used in `randomPoints()` function. We will use this code structure when generating background values with different types of buffer.
```{r nobuffer modleR, echo = F, eval = T}
# creating an object with coordinates values
Expand All @@ -109,7 +106,7 @@ buf <- create_buffer(occurrences = coord1sp[,c(2,3)],
buf.p <- dismo::randomPoints(buf,
n = 500,
excludep = T,
p = coord1sp)
p = pts)
# plotting environmental layer with background values
## environmental layer
plot(predictor, legend = F, main = "nobuffer createbuffer")
Expand All @@ -121,7 +118,7 @@ points(buf.p, col = "blue")
# Pseudoabsence selection with a buffer
Here, for all types of buffer, first we create an object with `create_buffer` using different types of buffer and then generate the background values with `randomPoints` from `dismo` package. In `create_buffer` function we implemented four types of buffer:
Here, for all types of buffer, first we create an object with `create_buffer()` using different types of buffer and then generate the background values with `randomPoints()` from `dismo` package. In `create_buffer()` function we implemented four types of buffer:
- `max`: maximum distance between occurrence points
- `mean`: mean distance between occurrence points
Expand All @@ -130,19 +127,19 @@ Here, for all types of buffer, first we create an object with `create_buffer` us
## Buffer with maximum distance `max`
In this example, we use `buffer_type="max"` to generate our first object.
In this example, we use `buffer_type = "max"` to generate our first object.
```{r buffermax, echo = F, eval = T}
#par(mfrow = c(1,1))
# creating an object with create_buffer with maximum distance buffer type
buf.max <- create_buffer(occurrences = coord1sp[,c(2,3)],
predictors = predictor,
buffer_type = "max")
predictors = predictor,
buffer_type = "max")
# creating 500 background values from buf.max
buf.max.p <- dismo::randomPoints(buf.max,
n = 500,
excludep = T,
p = coord1sp)
p = pts)
# plotting environmental layer with background values
## environmental layer
plot(predictor,
Expand All @@ -156,7 +153,7 @@ points(buf.max.p, col = "blue")
## Buffer with `mean` distance
In this example, we use `buffer_type="mean"` to generate our first object.
In this example, we use `buffer_type = "mean"` to generate our first object.
```{r buffermean, echo = F, eval = T}
#par(mfrow = c(1,1))
Expand All @@ -168,7 +165,7 @@ buf.mean <- create_buffer(occurrences = coord1sp[,c(2,3)],
buf.mean.p <- dismo::randomPoints(buf.mean,
n = 500,
excludep = T,
p = coord1sp)
p = pts)
# plotting environmental layer with background values
## environmental layer
plot(predictor,
Expand All @@ -194,7 +191,7 @@ buf.med <- create_buffer(occurrences = coord1sp[,c(2,3)],
buf.med.p <- dismo::randomPoints(buf.med,
n = 500,
excludep = T,
p = coord1sp)
p = pts)
# plotting environmental layer with background values
## environmental layer
plot(predictor,
Expand All @@ -220,7 +217,7 @@ buf.dist <- create_buffer(occurrences = coord1sp[,c(2,3)],
buf.dist.p <- dismo::randomPoints(buf.dist,
n = 500,
excludep = T,
p = coord1sp)
p = pts)
# plotting environmental layer with background values
## environmental layer
plot(predictor,
Expand All @@ -234,7 +231,7 @@ points(buf.dist.p, col = "blue")
# Within `setup_sdmdata`
Another option for implementing no buffer or any type of buffer is using function `setup_sdmdata` from `modleR`. In this case there's no need to call for `randomPoints` from `dismo` package since it is implemented inside `setup_sdmdata`.
Another option for implementing no buffer or any type of buffer is using function `setup_sdmdata()` from `modleR`. In this case there's no need to call for `randomPoints()` from `dismo` package or `create_buffer()` from `modleR` since it is implemented inside `setup_sdmdata()`.
## `setup_sdmdata` without buffer
Expand All @@ -254,7 +251,7 @@ a <- setup_sdmdata(species_name = especies[1],
# checking the output
head(a)
# plotting background point generation
knitr::include_graphics("./setupsdmdatanul/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
knitr::include_graphics("./setupsdmdatanul/Abarema_langsdorffii/present/data_setup/sdmdata_Abarema_langsdorffii.png")
```
## `setup_sdmdata` with maximum distance
Expand All @@ -275,7 +272,7 @@ b <- setup_sdmdata(species_name = especies[1],
# chacking the output
head(b)
# plotting backgrounf point generation
knitr::include_graphics("./setupsdmdatamax/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
knitr::include_graphics("./setupsdmdatamax/Abarema_langsdorffii/present/data_setup/sdmdata_Abarema_langsdorffii.png")
```
## `setup_sdmdata` with mean distance
Expand All @@ -296,7 +293,7 @@ c <- setup_sdmdata(species_name = especies[1],
# checking the output
head(c)
# plotting background point generation
knitr::include_graphics("./setupsdmdatamean/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
knitr::include_graphics("./setupsdmdatamean/Abarema_langsdorffii/present/data_setup/sdmdata_Abarema_langsdorffii.png")
```
## `setup_sdmdata` with median of distance between occurrence points
Expand All @@ -317,7 +314,7 @@ d <- setup_sdmdata(species_name = especies[1],
# checking the output
head(d)
# plotting background point generation
knitr::include_graphics("./setupsdmdatamed/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
knitr::include_graphics("./setupsdmdatamed/Abarema_langsdorffii/present/data_setup/sdmdata_Abarema_langsdorffii.png")
```
## `setup_sdmdata` with a specific distance between occurrence points
Expand All @@ -339,7 +336,7 @@ e <- setup_sdmdata(species_name = especies[1],
# checking the output
head(e)
# plotting the background point generation
knitr::include_graphics("./setupsdmdatadist/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
knitr::include_graphics("./setupsdmdatadist/Abarema_langsdorffii/present/data_setup/sdmdata_Abarema_langsdorffii.png")
```
# Effect of different types of buffer on model projection and evaluation
Expand All @@ -348,21 +345,20 @@ Now we want to inspect how different types of buffer affect the prediction of th
## Using no buffer on model predictions
Here, we use the algorithm `maxent` without any buffer to generate a model for the species *Eugenia florida*.
Here, we use the algorithm `maxent` without any buffer to generate a model for the species *Abarema langsdorfii*.
A simple model with no buffer:
```{r enm1, message = F, eval = T}
# running a maxent model for Eugenia florida with no buffer
no_buf <- do_any(species_name = especies[1], # sp 1 Eugenia florida
# running a maxent model for Abarema langsdorfii with no buffer
no_buf <- setup_sdmdata(species_name = especies[1], # sp 1 Abarema langsdorfii
occurrences = coord1sp[, -1], # coordinates for the species
predictors = example_vars, # example environmental data
models_dir = "./buftest1",
algo = "maxent", # maxent algorithm
clean_dupl = T,
clean_na = T)
# creating an raster object with the output of the model
no_buf_m <- raster("./buftest1/Eugenia florida DC./present/partitions/maxent_cut_Eugenia florida DC._1_1.tif")
# creating a raster object with the output of the model
no_buf_m <- raster("./buftest1/Abarema_langsdorffii/present/partitions/")
# plotting the model
plot(no_buf_m)
# adding political limts to the map
Expand All @@ -371,13 +367,13 @@ maps::map(,,add = T)
## Using `median` buffer on model predictions
Again, we use the algorithm `algo=maxent` but now we set `buffer_type=median` to generate a model for the species *Eugenia florida*.
Again, we use the algorithm `algo=maxent` but now we set `buffer_type=median` to generate a model for the species *Abarema langsdorfii*.
With a median buffer:
```{r enm2, message = F, eval = F}
# Running a maxent model for E. florida with a median buffer
mn_buf <- do_any(species_name = especies[1], # sp 1 Eugenia florida
mn_buf <- do_any(species_name = especies[1], # sp 1 Abarema langsdorfii
occurrences = coord1sp[, -1], # coodinates for the species
predictors = example_vars, # environmental layers
models_dir = "./buftest2",
Expand All @@ -386,7 +382,7 @@ mn_buf <- do_any(species_name = especies[1], # sp 1 Eugenia florida
clean_na = T,
buffer_type = "median") # setting median buffer
# creating a raster object with the output of the model
mn_buf_m <- raster("./buftest2/Eugenia florida DC./present/partitions/maxent_cut_Eugenia florida DC._1_1.tif")
mn_buf_m <- raster("./buftest2/Abarema langsdorfii DC./present/partitions/maxent_cut_Abarema langsdorfii DC._1_1.tif")
# plotting the model
plot(mn_buf_m)
# adding political limits to map
Expand All @@ -399,8 +395,8 @@ Fitting and evaluating the models with different background samples has a differ
```{r remedy001, fig.cap="Comparing the sdmdata datasets", eval = T, echo=FALSE, fig.width=6, fig.height=3}
knitr::include_graphics("./buftest1/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png") ;
knitr::include_graphics("./buftest2/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
knitr::include_graphics("./buftest1/Abarema langsdorfii DC./present/partitions/sdmdata_Abarema langsdorfii DC..png") ;
knitr::include_graphics("./buftest2/Abarema langsdorfii DC./present/partitions/sdmdata_Abarema langsdorfii DC..png")
```
# Effect of buffer on the background data in MaxEnt
Expand Down Expand Up @@ -538,10 +534,10 @@ no_buf <- do_any(species_name = especies[1],
clean_na = T,
buffer_type = NULL)
no_buf_m <- raster("./maxenttest//Eugenia florida DC./present/partitions/maxent_cut_Eugenia florida DC._1_1.tif")
no_buf_m <- raster("./maxenttest//Abarema langsdorfii DC./present/partitions/maxent_cut_Abarema langsdorfii DC._1_1.tif")
plot(no_buf_m)
maps::map(,,add = T)
knitr::include_graphics("./maxenttest/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
knitr::include_graphics("./maxenttest/Abarema langsdorfii DC./present/partitions/sdmdata_Abarema langsdorfii DC..png")
no_buf$TSS
```
Expand All @@ -557,10 +553,10 @@ mx_buf <- do_any(species_name = especies[1],
clean_na = T,
buffer_type = "median")
no_buf_m <- raster("./maxentmax/Eugenia florida DC./present/partitions/maxent_cut_Eugenia florida DC._1_1.tif")
no_buf_m <- raster("./maxentmax/Abarema langsdorfii DC./present/partitions/maxent_cut_Abarema langsdorfii DC._1_1.tif")
plot(no_buf_m)
maps::map(,,add = T)
knitr::include_graphics("./maxentmax/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
knitr::include_graphics("./maxentmax/Abarema langsdorfii DC./present/partitions/sdmdata_Abarema langsdorfii DC..png")
mx_buf$TSS
```
Expand All @@ -577,10 +573,10 @@ mx_dbuf <- do_any(species_name = especies[1],
buffer_type = "distance",
dist_buf = 4)
#dbuf_mx <- raster("./maxentdist/Eugenia florida DC./present/partitions/maxent_cut_Eugenia florida DC._1_1.tif")
#dbuf_mx <- raster("./maxentdist/Abarema langsdorfii DC./present/partitions/maxent_cut_Abarema langsdorfii DC._1_1.tif")
#plot(dbuf_mx)
#maps::map(,,add = T)
#knitr::include_graphics("./maxentdist/Eugenia florida DC./present/partitions/sdmdata_Eugenia florida DC..png")
#knitr::include_graphics("./maxentdist/Abarema langsdorfii DC./present/partitions/sdmdata_Abarema langsdorfii DC..png")
#dbuf_mx$TSS
```
Expand Down
3 changes: 2 additions & 1 deletion euclidean_distance_tests/Euclidean_distance.Rmd
View file
Original file line number Diff line number Diff line change
Expand Up @@ -33,9 +33,10 @@ dim(coord1sp)
ceiling(0.7 * nrow(coord1sp))
# Making a sample of 70% of species' records
set <- sample(1:nrow(coord1sp), size = ceiling(0.7 * nrow(coord1sp)))
# Creating training data set (70% of species' records)
train_set <- coord1sp[set,]
# Creating test data set (other 30%)
test_set <- coord1sp[setdiff(1:nrow(coord1sp),set),]
```

# Fitting and projection datasets
Expand Down
Loading

0 comments on commit baebaf2

Please sign in to comment.