diff --git a/.Rhistory b/.Rhistory
index e97b463..0fbff4c 100755
--- a/.Rhistory
+++ b/.Rhistory
@@ -272,167 +272,520 @@ m1 <- tar_read(plot_covariate_maps_97adc691)
 m1
 1259*24
 1259*24/2
-getwd()
-tar_meta(sf_pesticide_partition_cleaned)
-tm <- tar_meta(sf_pesticide_partition_cleaned)
-tm
-tm$data
-tm$size
-tm$path
-unlist(tm)
-tm[[1]]
-tm[[2]]
-tm[[3]]
-tm[[4]]
-tm[[5]]
-tm[[6]]
-colnames(tm)
-tm$time
-tm$size
-tm$format
-tm$repository
-tm$iteration
-tl <- tar_load(sf_pesticide_partition_cleaned)
-tl
-tar_load(sf_pesticide_partition_cleaned)
-tm <- tar_manifest(sf_pesticide_partition_cleaned)
-tm
-tm$name
-tm$command
-tm <- tar_meta_read(sf_pesticide_partition_cleaned)
-tm
-tm <- tar_meta(sf_pesticide_partition_cleaned)
-tm
-tm$size
-typeof(tm)
-tm
-tm[[1]]
-tm[[18]]
-tm[[17]]
-tm[[16]]
-tn
-tm
-tm$bytes
-tm$command
-tm$iteration
-to <- tar_outdated(sf_pesticide_partition_cleaned)
-to
-to <- tar_outdated()
-to
-tar_outdated()
-getwD()
-getwd()
-tar_exist_objects(sf_pesticide_partition_cleaned)
-tar_exist_objects("sf_pesticide_partition_cleaned")
-tar_process(sf_pesticide_partition_cleaned)
-tar_make()
-?tar_file_read
-if (identical(Sys.getenv("TAR_LONG_EXAMPLES"), "true")) {
-targets::tar_dir({ # tar_dir() runs code from a temporary directory.
-targets::tar_script({
-tar_file_read(data, get_path(), read_csv(file = !!.x, col_types = cols()))
+=======
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ ifelse(.x$lft_cns == 0, .x$cncntrt, 1e-9))) |>
+pull(data)
+# 2) A list of LOD, each element is a vector of the limit of detection
+# use dplyr to create a separate list by each ChmlNm and create an LOD. The
+# LOD should be the cnctrnt
+pesticide_lod <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x$cnctrnt)) |>
+pull(data)
+# 2) A list of LOD, each element is a vector of the limit of detection
+# use dplyr to create a separate list by each ChmlNm and create an LOD. The
+# LOD should be the cnctrnt
+pesticide_lod <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x$cncntrt)) |>
+pull(data)
+pesticide_outcomes[[1]][1:10]
+pesticide_lod[[1]][1:10]
+#2a) Create a list of the of the lft_cns variable by ChmclNm
+pesticide_lft_cns <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x$lft_cns)) |>
+pull(data)
+pesticide_lft_cns[[1]][1:10]
+cbind(pesticide_lft_cns[[1]][1:10],pesticide_outcomes[[1]][1:10],pesticide_lod[[1]][1:10])
+# 4) A list of X's, each element is a matrix of the covariates
+# For now, we will use the covariates in the data_analysis
+pesticide_covariates <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(-cncntrt, -lft_cns))) |>
+pull(data)
+pesticide_covariates[[1]]
+pesticide_covariates[[2]]
+pesticide_covariates[[3]]
+pesticide_covariates[[4]]
+ls
+pesticide_covariates[[3]]
+data_analysis
+data_analysis[[1]][,1:20]
+data_analysis[[1]][,1:10]
+data_analysis[,1:10]
+# 5) A list of locs, each element is a matrix of the locations
+# For now, we will use the locations in the data_analysis - it includes x, y, and time
+# Because the locations are from the sf geometry object we need to use the original data
+# We get the geometry from the sf object and then add the time (column name is Year)
+pesticide_locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ st_coordinates(.x |> select(geometry)))) |>
+pull(data)
+# 5) A list of locs, each element is a matrix of the locations
+# For now, we will use the locations in the data_analysis - it includes x, y, and time
+# Because the locations are from the sf geometry object we need to use the original data
+# We get the geometry from the sf object and then add the time (column name is Year)
+data_analysis <- splits |>
+analysis() |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(geometry = st_as_text(geometry)) |>
+sf::st_drop_geometry()
+# 5) A list of locs, each element is a matrix of the locations
+# For now, we will use the locations in the data_analysis - it includes x, y, and time
+# Because the locations are from the sf geometry object we need to use the original data
+# We get the geometry from the sf object and then add the time (column name is Year)
+data_analysis <- splits |>
+analysis() |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, st_coordinates())) |>
+sf::st_drop_geometry()
+data_analysis <- splits |>
+analysis() |>
+group_by(ChmclNm)
+data_analysis <- splits |>
+analysis()
+data_analysis
+# 5) A list of locs, each element is a matrix of the locations
+# We need to get the locations from the original data which is an sf object
+# Then we need to get the time variable - Year
+# Then we need to group by chemical name and nest the data
+# Then we need to create a list of the locations
+pesticide_locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+pesticide_locs
+# 5) A list of locs, each element is a matrix of the locations
+# We need to get the locations from the original data which is an sf object
+# Then we need to get the time variable - Year
+# Then we need to group by chemical name and nest the data
+# Then we need to create a list of the locations
+# Then we need to conver the geometry to a matrix since PrestoGP doesn't take sf objects
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ st_coordinates(.x))) |>
+pull(data)
+locs
+locs[[1]]
+locs[[2]]
+# 5) A list of locs, each element is a matrix of the locations
+# We need to get the locations from the original data which is an sf object
+# Then we need to get the time variable - Year
+# Then we need to group by chemical name and nest the data
+# Then we need to create a list of the locations
+# Then we need to conver the geometry to a matrix since PrestoGP doesn't take sf objects
+# Make sure the Time variable is included in the data
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+locs[[1]]
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+locs <- locs |>
+map(~ st_coordinates(.x)) |>
+map(~ as.matrix(.x)) |>
+map(~ .x[,-1])
+locs[[1]]
+length(locs)
+head(locs[[1]])
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+head(locs[[1]])
+locs <- locs |>
+map(~ st_coordinates(.x))
+locs[[1]]
+locs[[1]] |> head()
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+locs[[1]]
+locs <- locs |>
+map(~ st_coordinates(.x))
+locs[[1]]
+locs[[1]] |> head()
+locs <- locs |>
+map(~ st_coordinates(.x) + Year)
+locs <- locs |> # Retain Year
+map(~ .x %>%
+st_coordinates() |>
+as.data.frame() |>
+as.matrix())
+?map
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+locs
+locs <- locs |>
+map(~ st_coordinates(.x)) |
+map(~ as.matrix(.x)) |>
+map(~ .x[,-1])
+locs <- locs |> # Retain Year
+map(~ st_coordinates(.x)) |>
+map(~ as.matrix(.x)) |>
+map(~ .x[,-1])
+locs[[1]]
+locs[[1]] |> head()
+locs <- locs |> # Retain Year
+map(~ st_coordinates(.x)) |>
+map(~ .x[,-1])
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+locs <- locs |> # Retain Year
+map(~ st_coordinates(.x)) |>
+map(~ .x[,-1])
+locs[[1]] |> head()
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+locs <- locs |> # Retain Year
+map(~ st_coordinates(.x))
+locs[[1]]
+locs[[1]]|> head()
+typeof(locs[[1]])
+locs[[1]]$X
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+locs[[1]]
+locs <- locs |> # Retain Year
+map(Year ~ st_coordinates(.x))
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+# Update the locs list
+locs <- locs |>
+map(~ {
+coords <- st_coordinates(.x)
+data.frame(
+X = coords[, "X"],
+Y = coords[, "Y"],
+Year = .x$Year
+)
 })
-targets::tar_manifest()
+locs[[1]]
+locs[[1]] |> head()
+# 5) A list of locs, each element is a matrix of the locations
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+# Update the locs list
+locs <- locs |>
+map(~ {
+coords <- st_coordinates(.x)
+data.frame(
+X = coords[, "X"],
+Y = coords[, "Y"],
+Year = .x$Year
+)
 })
-}
-targets::tar_dir({ # tar_dir() runs code from a temporary directory.
-+         targets::tar_script({
-+             tar_file_read(data, get_path(), read_csv(file = !!.x, col_types = cols()))
-+         })
-?tar_read
-tm
-tm$error
-tm$data
-?tar_target
-?tar_option_set
-(1.1-1)
-(1.1-1)/1
-(1.1-1)/1.1
-getwd()
-tar_make
+# 6) Apanasovich is TRUE
+Apanasovich = TRUE
+# 7) scaling = c(1,1,2) -> maybe needs to be list? could also do c(1,2,3)
+scaling = c(1,1,2)
+pesticide_mvm <-  new("MultivariateVecchiaModel", n_neighbors = 10)
+# 3) Impute.y = TRUE, impute the missing values in Y
+Impute.y = TRUE
+?prestogp_fit
+soil.mvm <- prestogp_fit(pesticide_mvm, Y = pesticide_outcomes, lod = pesticide_lod,
+impute.y = Impute.y, X = pesticide_covariates, locs = locs,
+apanasovich = Apanasovich, scaling = scaling)
+locs[[1]]
+is.matrix(locs[[1]])
+as.matrix(locs[[1]])
+is.matrix(as.matrix(locs[[1]]))
+locs <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(id, Year))) |>
+pull(data)
+# Update the locs list
+locs <- locs |>
+map(~ {
+coords <- st_coordinates(.x)
+mat <- as.matrix(data.frame(
+X = coords[, "X"],
+Y = coords[, "Y"],
+Year = .x$Year
+))
+return(mat)
+})
+locs[[1]] |> head()
+soil.mvm <- prestogp_fit(pesticide_mvm, Y = pesticide_outcomes, lod = pesticide_lod,
+impute.y = Impute.y, X = pesticide_covariates, locs = locs,
+apanasovich = Apanasovich, scaling = scaling)
+pesticide_covariates[[1]]
+# 4) A list of X's, each element is a matrix of the covariates
+# For now, we will use the covariates in the data_analysis
+pesticide_covariates <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(-cncntrt, -lft_cns, -id, -Year))) |>
+pull(data)
+pesticide_covariates[[1]]
+# 4) A list of X's, each element is a matrix of the covariates
+# For now, we will use the covariates in the data_analysis
+pesticide_covariates <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ {
+selected_data <- .x %>% select(-cncntrt, -lft_cns, -id, -Year)
+as.matrix(selected_data)
+})) |>
+pull(data)
+# 4) A list of X's, each element is a matrix of the covariates
+# For now, we will use the covariates in the data_analysis
+# Convert each list element to a matrix (i.e. is.matrix = TRUE)
+pesticide_covariates <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ as.matrix(.x %>% select(-cncntrt, -lft_cns)))) |>
+pull(data)
+# 4) A list of X's, each element is a matrix of the covariates
+# For now, we will use the covariates in the data_analysis
+# Convert each list element to a matrix (i.e. is.matrix = TRUE)
+pesticide_covariates <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(-cncntrt, -lft_cns, -id, -Year))) |>
+pull(data)
+pesticide_covariates
+pesticide_covariates[[1]]
+pesticide_covariates <- lapply(pesticide_covariates, as.matrix)
+pesticide_covariates[[1]]
+lapply(pesticide_covariates,length)
+# convert to matrix, but retain dimensions
+pesticide_covariates <- pesticide_covariates |>
+map(~ as.matrix(.x, is.matrix = TRUE))
+# 4) A list of X's, each element is a matrix of the covariates
+# For now, we will use the covariates in the data_analysis
+# Convert each list element to a matrix (i.e. is.matrix = TRUE)
+pesticide_covariates <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(-cncntrt, -lft_cns, -id, -Year))) |>
+pull(data)
+# convert to matrix, but retain dimensions
+pesticide_covariates <- pesticide_covariates |>
+map(~ as.matrix(.x, is.matrix = TRUE))
+pesticide_covariates[[1]] |> head()
+pesticide_covariates[[1]][,1:20] |> head()
+# 4) A list of X's, each element is a matrix of the covariates
+# For now, we will use the covariates in the data_analysis
+# Convert each list element to a matrix (i.e. is.matrix = TRUE)
+pesticide_covariates <- data_analysis |>
+group_by(ChmclNm) |>
+nest() |>
+mutate(data = map(data, ~ .x %>% select(-cncntrt, -lft_cns, -id, -id.1, -Year))) |>
+pull(data)
+# convert to matrix, but retain dimensions
+pesticide_covariates <- pesticide_covariates |>
+map(~ as.matrix(.x, is.matrix = TRUE))
+soil.mvm <- prestogp_fit(pesticide_mvm, Y = pesticide_outcomes, lod = pesticide_lod,
+impute.y = Impute.y, X = pesticide_covariates, locs = locs,
+apanasovich = Apanasovich, scaling = scaling)
+pesticide_lod[[1]]
+soil.mvm <- prestogp_fit(pesticide_mvm, Y = pesticide_outcomes, lod = pesticide_lod,
+impute.y = Impute.y, X = pesticide_covariates, locs = locs,
+apanasovich = Apanasovich, scaling = scaling)
+pesticide_lod[[1]]
+pesticide_lod[[1]] |> dim()
+pesticide_lod[[2]]
+pesticide_lod[[3]]
+pesticide_lod[[4]]
+pesticide_lod[[5]]
+pesticide_lod[[6]]
+lapply(pesticide_lod,length)
+lapply(pesticide_outcomes,length)
+soil.mvm <- prestogp_fit(pesticide_mvm, Y = pesticide_outcomes, lod = pesticide_lod,
+impute.y = Impute.y, X = pesticide_covariates, locs = locs,
+apanasovich = Apanasovich, scaling = scaling)
+pesticide_mvm <- prestogp_fit(pesticide_mvm, Y = pesticide_outcomes, lod = pesticide_lod,
+impute.y = Impute.y, X = pesticide_covariates, locs = locs,
+apanasovich = Apanasovich, scaling = scaling)
+debug(prestogp_fit)
+pesticide_mvm <- prestogp_fit(pesticide_mvm, Y = pesticide_outcomes, lod = pesticide_lod,
+impute.y = Impute.y, X = pesticide_covariates, locs = locs,
+apanasovich = Apanasovich, scaling = scaling)
+undebug(prestogp_fit)
+debug(check_input)
+debug(PrestoGP::check_input)
 tar_make()
-tar_visnetwork()
-tar_visnetwork(targets_only = T)
-p3
-p10
+tar_visnetwork(targets_only = T, exclude = c("kfolds_iter","kfolds_iter3", "set_path","readQS"))
+sf_plot_outcome_maps <- tar_read(sf_plot_outcome_maps)
+sf_plot_outcome_maps
+plot_pesticide_maps <- tar_read(plot_pesticide_maps)
+plot_pesticide_maps <- tar_read(plot_pesticide_maps, branches = 1)
+plot_covariate_maps1 <- tar_read(plot_covariate_maps, branches = 1)
+plot_covariate_maps1
+plot_pesticide_maps <- tar_read(plot_pesticide_maps, branches = 1)
+plot_pesticide_maps
+plot_pesticide_maps <- tar_read(plot_pesticide_maps, branches = 3)
+plot_pesticide_maps
+explore_skim
+library(terra)
+r <- terra::rast("/Volumes/set/OpenLandMapData/Bulk_Density/OpenLand_SoilBulkDensity_010cm.tif")
+plot(r)
+#clip to the US extent
+r <- terra::rast("/Volumes/set/OpenLandMapData/OLM_Combined/OLM_US_Crop_Clay_Content.tif")
+r
+plot(r[[1]])
+plot(r[[6]])
+r <- terra::rast("/Volumes/set/OpenLandMapData/OLM_Combined/OLM_US_Crop_Soil_Order.tif")
+plot(r[[1]])
+r
+names(r)
+plot(r[6])
+plot(r[[6]])
+colnames(sf_pesticide_partition_cleaned[[1]])
+# Create and plot using ggplot an example covariance matrix of 1000 variables
+library(MASS)
+library(ggplot2)
+set.seed(123)
+n <- 1000
+Sigma <- matrix(0, n, n)
+for (i in 1:n) {
+for (j in 1:n) {
+Sigma[i, j] <- 0.5^abs(i - j)
+}
+}
+Sigma <- Sigma + diag(n) * 0.1
+X <- mvrnorm(n = 1000, rep(0, n), Sigma)
+ggplot(data = as.data.frame(Sigma), aes(x = Var1, y = Var2, fill = Freq)) +
+geom_tile() +
+theme_minimal() +
+scale_fill_gradient(low = "white", high = "blue") +
+theme(axis.text.x = element_blank(),
+axis.text.y = element_blank(),
+axis.ticks = element_blank(),
+axis.title.x = element_blank(),
+axis.title.y = element_blank())
+head(Sigma)
+Sigma <- matrix(0, n, n)
+for (i in 1:n) {
+for (j in 1:n) {
+Sigma[i, j] <- 0.5^abs(i - j)
+}
+}
+Sigma <- Sigma + diag(n) * 0.1
+X <- mvrnorm(n = 1000, rep(0, n), Sigma)
+Sigma <- cov(X)
+Sigma <- as.data.frame(as.table(Sigma))
+names(Sigma) <- c("Var1", "Var2", "Freq")
+ggplot(data = Sigma, aes(x = Var1, y = Var2, fill = Freq)) +
+geom_tile() +
+theme_minimal() +
+scale_fill_gradient(low = "white", high = "blue") +
+theme(axis.text.x = element_blank(),
+axis.text.y = element_blank(),
+axis.ticks = element_blank(),
+axis.title.x = element_blank(),
+axis.title.y = element_blank())
+n <- 100
+Sigma <- matrix(0, n, n)
+for (i in 1:n) {
+for (j in 1:n) {
+Sigma[i, j] <- 0.5^abs(i - j)
+}
+}
+Sigma <- Sigma + diag(n) * 0.1
+X <- mvrnorm(n = 1000, rep(0, n), Sigma)
+Sigma <- cov(X)
+Sigma <- as.data.frame(as.table(Sigma))
+names(Sigma) <- c("Var1", "Var2", "Freq")
+ggplot(data = Sigma, aes(x = Var1, y = Var2, fill = Freq)) +
+geom_tile() +
+theme_minimal() +
+scale_fill_gradient(low = "white", high = "blue") +
+theme(axis.text.x = element_blank(),
+axis.text.y = element_blank(),
+axis.ticks = element_blank(),
+axis.title.x = element_blank(),
+axis.title.y = element_blank())
+colnames(sf_pesticide_partition_cleaned[[1]])
+View(sf_pesticide_partition_cleaned)
+plot(r[6])
+plot_pesticide_maps <- tar_read(plot_pesticide_maps, branches = 6)
+plot_pesticide_maps
+data <- sf_pesticide_partition_cleaned[[1]]
+states <- st_as_sf(maps::map("state", plot = FALSE, fill = TRUE))
+df_obs <- data |> filter(lft_cns == 0)
+df_cens <- data |> filter(lft_cns == 1)
+p <- ggplot() +
+geom_sf(data = df_cens, size = 0.5) +
+geom_sf(data = df_obs, aes(color = cncntrt)) +
+# facet_wrap(~ Year) +
+scale_color_viridis_c(option = "A", trans = scales::pseudo_log_trans(sigma = 0.005)) +
+geom_sf(data = states, fill = NA, size=0.15) +
+theme_minimal() +
+theme(legend.position = "right") +
+ggtitle(data$ChmclNm)
+p
 getwd()
-# install.packages("devtools")
-devtools::install_github("NIEHS/PrestoGP")
-library(PrestoGP)
-tar_make()
-library(targets)
-tar_make()
-tar_visnetwork()
 tar_visnetwork(targets_only = T)
-data(soil)
-soil <- soil[!is.na(soil[,5]),] # remove rows with NA's
-y <- soil[,4]                   # predict moisture content
-X <- as.matrix(soil[,5:9])
-locs <- as.matrix(soil[,1:2])
-# Vecchia model
-soil.vm <- new("VecchiaModel", n_neighbors = 10)
-soil.vm <- prestogp_fit(soil.vm, y, X, locs)
-# Full model
-soil.fm <- new("FullModel")
-soil.fm <- prestogp_fit(soil.fm, y, X, locs)
-# Multivariate model
-ym <- list()
-ym[[1]] <- soil[,5]             # predict two nitrogen concentration levels
-ym[[2]] <- soil[,7]
-Xm <- list()
-Xm[[1]] <- Xm[[2]] <- as.matrix(soil[,c(4,6,8,9)])
-locsm <- list()
-locsm[[1]] <- locsm[[2]] <- locs
-soil.mvm <-  new("MultivariateVecchiaModel", n_neighbors = 10)
-?new
-soil.mvm <- prestogp_fit(soil.mvm, ym, Xm, locsm)
-# Space/elevation model
-data(soil250, package="geoR")
-y2 <- soil250[,7]               # predict pH level
-X2 <- as.matrix(soil250[,c(4:6,8:22)])
-# columns 1+2 are location coordinates; column 3 is elevation
-locs2 <- as.matrix(soil250[,1:3])
-soil.vm2 <- new("VecchiaModel", n_neighbors = 10)
-# fit separate scale parameters for location and elevation
-soil.vm2 <- prestogp_fit(soil.vm2, y2, X2, locs2, scaling = c(1, 1, 2))
-soil.mvm
-ym
-soil
-View(soil)
-ym
-dim(ym)
-length*ym
-length(ym)
-length(xm)
-length(Xm)
-dim(ym[[1]])
-length(ym[[1]])
-dim(ym[[2]])
-length(ym[[2]])
-dim(Xm[[1]])
-?prestogp_fit
-ym[[1]] <- soil[1:100,5]             # predict two nitrogen concentration levels
-ym[[2]] <- soil[,7]
-Xm <- list()
-Xm[[1]] <- Xm[[2]] <- as.matrix(soil[,c(4,6,8,9)])
-Xm[[1]] <- Xm[[1]][1:100,]
-locsm <- list()
-locsm[[1]] <- locsm[[2]] <- locs
-locsm[[1]] <- locsm[[1]][1:100,]
-soil.mvm <-  new("MultivariateVecchiaModel", n_neighbors = 10)
-soil.mvm <- prestogp_fit(soil.mvm, ym, Xm, locsm)
-soil.mvm
-soil.mvm[[1]]
-soil.mvm
-coef(soil.mvm)
-summary(soil.mvm)
-sf_pesticide_for_fit <- tar_read(sf_pesticide_for_fit)
-sf_pesticide_for_fit
-View(sf_pesticide_for_fit)
-sf_pesticide_partition_cleaned <- tar_read(sf_pesticide_partition_cleaned)
-sf_pesticide_partition_cleaned
-sf_pesticide_partition_cleaned[[1]]
+tar_visnetwork(targets_only = T, exclude = c("kfolds_iter","kfolds_iter3"))
+COMPUTE_MODE <- 1
+path_base <-
+ifelse(COMPUTE_MODE == 1,
+"/Volumes/SET/Projects/PrestoGP_Pesticides/input/",
+ifelse(COMPUTE_MODE == 2,
+"/ddn/gs1/group/set/Projects/PrestoGP_Pesticides/input/",
+ifelse(COMPUTE_MODE == 3,
+"/opt/",
+ifelse(COMPUTE_MODE == 4,
+"~/Downloads/",
+stop("COMPUTE_MODE should be one of 1, 2, 3, or 4.\n")
+)
+)
+)
+)
+path_base
+## TWI
+twi <- paste0(path_base, "TWI/CONUS_TWI_epsg5072_30m_unmasked.tif")
+twi_ras <- terra::rast(twi)
+twi_ras
+plot(twi_ras)
 View(sf_pesticide_partition_cleaned)
+<<<<<<< HEAD
 sf_pesticide_partition_cleaned[[1]]
 sf_pesticide_partition_cleaned[[2]]
 sf_pesticide_partition_cleaned <- tar_read(sf_pesticide_partition_cleaned)