---
title: "Filter the best classified occurrences within grid cells"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{naturaList_grid_filter_vignette}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

With `grid_filter()` you may filter occurrence records inside grid cells in order to keep only the record with the highest confidence level in each grid cell.

![Fig. 1 - The procedure of `grid_filter()` to select the record with the highest confidence level in each grid cell](../vignettes/grid_filter.png)

As example, we will filter the `occ.class` in square grid cells of 0.5 decimal degree of side, which is the default grid cell size of `grid_filter()`. You can provide the height and width of the grid cell in `grid.resolution` argument, or provide a raster layer in argument `r` from which the cell is used to filter occurrences.

```{r grid_filter, eval=F}
# Load package and data
library(naturaList)

data("A.setosa")
data("speciaLists")

# classification
occ.class <- classify_occ(A.setosa, speciaLists)

# grid filtering
occ.grid <- grid_filter(occ.class)

dim(occ.grid)
table(occ.grid$naturaList_levels)
```