Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/spatlyu/specula
Spatial Inference And Spatial Prediction With R
https://github.com/spatlyu/specula
geocomputation geoinformatics giscience r spatial-analysis spatial-predictions spatial-regression spatial-statistics
Last synced: about 1 month ago
JSON representation
Spatial Inference And Spatial Prediction With R
- Host: GitHub
- URL: https://github.com/spatlyu/specula
- Owner: SpatLyu
- Created: 2024-05-10T01:02:48.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2024-05-28T16:35:14.000Z (7 months ago)
- Last Synced: 2024-05-29T02:41:05.251Z (7 months ago)
- Topics: geocomputation, geoinformatics, giscience, r, spatial-analysis, spatial-predictions, spatial-regression, spatial-statistics
- Language: R
- Homepage:
- Size: 3.42 MB
- Stars: 4
- Watchers: 2
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.Rmd
Awesome Lists containing this project
README
---
output: github_document
---
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "##",
fig.path = "man/figures/README-",
out.width = "100%",
warning = FALSE,
message = FALSE
)
```
# spEcula 
[](https://CRAN.R-project.org/package=spEcula)
[](https://spatlyu.r-universe.dev/spEcula)
The goal of **spEcula** is to make it easier to use **R** for **spatial prediction** based on **various spatial relationships** (e.g. spatial dependence, spatial heterogeneity and geographical similarity).
## Overview
Full document of the most recent release of **spEcula** is online:
Current models and functions provided by **spEcula** are:
| **spatial prediction method** | **spEcula function** | **support status** |
|-----------------------------------|--------------------------|------------------------|
| Geographically Optimal Similarity | `gos()` | ✔️ |
| Sandwich Mapping Model | `sandwich()` | ✔️ |
## Installation
You can install the development version of *spEcula* from [*github*](https://github.com/SpatLyu/spEcula):
``` r
# install.packages("devtools")
devtools::install_github("SpatLyu/spEcula",
build_vignettes = T,
dep = T)
```
or install *spEcula* from [*r-universe*](https://spatlyu.r-universe.dev/spEcula):
```r
install.packages('spEcula',
repos = c("https://spatlyu.r-universe.dev",
"https://cran.rstudio.com/"),
dep = TRUE)
```
## Example
### Geographically Optimal Similarity (GOS) model
```{r example_gos}
library(spEcula)
data(zn)
data(grid)
zn$Zn = log(zn$Zn)
tictoc::tic()
g1 = gos(Zn ~ Slope + Water + NDVI + SOC + pH + Road + Mine,
data = zn, newdata = grid, kappa = 0.08,cores = 6)
tictoc::toc()
g1$pred = exp(g1$pred)
grid$pred = g1$pred
grid$uc99 = g1$`uncertainty99`
g1
```
```{r}
library(ggplot2)
library(cowplot)
library(viridis)
```
```{r gos_result,fig.width=9.5,fig.height=3.5}
f1 = ggplot(grid, aes(x = Lon, y = Lat, fill = pred)) +
geom_tile() +
scale_fill_viridis(option="magma", direction = -1) +
coord_equal() +
labs(fill='Prediction') +
theme_bw()
f2 = ggplot(grid, aes(x = Lon, y = Lat, fill = uc99)) +
geom_tile() +
scale_fill_viridis(option="mako", direction = -1) +
coord_equal() +
labs(fill=bquote(Uncertainty~(zeta==0.99))) +
theme_bw()
plot_grid(f1,f2,nrow = 1,label_fontfamily = 'serif',
labels = paste0('(',letters[1:2],')'),
label_fontface = 'plain',label_size = 10,
hjust = -1.5,align = 'hv') -> p
p
```
### Sandwich Mapping Model
```{r sandwich,fig.width=5.5,fig.height=3.5}
library(sf)
library(tidyverse)
library(spEcula)
simpath = system.file("extdata", "sim.gpkg", package="spEcula")
sampling = read_sf(simpath,layer = 'sim_sampling')
ssh = read_sf(simpath,layer = 'sim_ssh')
reporting = read_sf(simpath,layer = 'sim_reporting')
sampling_zone = sampling %>%
st_join(ssh['X']) %>%
st_drop_geometry()
library(ggpubr)
ggerrorplot(sampling_zone, x = "X", y = "Value",
desc_stat = "mean_sd", color = "black",
add = "violin", add.params = list(color = "darkgray")) +
geom_text(data = summarise(sampling_zone,vmean = mean(Value),.by = X),
aes(x = X, y = vmean, label = round(vmean,2)),
vjust = -0.5, hjust = -0.15, color = "black",size = 3) +
scale_x_discrete(labels = LETTERS[1:4]) +
theme(axis.title.x = element_blank())
sim_est = sandwich(sampling = sampling,stratification = ssh,reporting = reporting,
sampling_attr = 'Value',ssh_zone = 'X',reporting_id = 'Y',
weight_type = 'area')
sim_est
```
```{r sandwich_sim_est,fig.width=9.75, fig.height=3.5}
library(cowplot)
f1 = ggplot(data = sim_est, aes(fill = sandwichest_mean),
color = "darkgray") +
geom_sf() +
labs(fill='mean') +
scale_fill_gradient(low = "#f0bc9c", high = "red",
breaks = range(sim_est$sandwichest_mean)) +
theme_bw() +
theme(
axis.text = element_blank(),
axis.ticks = element_blank(),
axis.title = element_blank(),
panel.grid = element_blank(),
legend.position = 'right',
legend.background = element_rect(fill = 'transparent',color = NA)
)
f2 = ggplot(data = sim_est, aes(fill = sandwichest_standarderror),
color = "darkgray") +
geom_sf() +
labs(fill='se') +
scale_fill_gradient(low = "#b6edf0", high = "blue",
breaks = range(sim_est$sandwichest_standarderror)) +
theme_bw() +
theme(
axis.text = element_blank(),
axis.ticks = element_blank(),
axis.title = element_blank(),
panel.grid = element_blank(),
legend.position = 'right',
legend.background = element_rect(fill = 'transparent',color = NA)
)
plot_grid(f1, f2, nrow = 1,label_fontfamily = 'serif',
labels = paste0('(',letters[1:4],')'),
label_fontface = 'plain',label_size = 10,
hjust = 0.05,align = 'hv') -> p
p
```