facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
theme_bw()
p
}
figure(new_lst$NED)
figure <- function(df){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
labs(title = "Number of common ID", fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = 32)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
theme_bw()
p
}
figure(new_lst$NED)
figure <- function(df){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
labs(title = "Number of common ID", fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = 35)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
theme_bw()
p
}
figure(new_lst$NED)
for (i in 1:28){
p <- figure(new_lst[[i]])
ggsave(plot = p,
filename = paste0(get.data.dir(),"/plots/",sheetsnames[i] ,"_plot.png"),
width = 12, height = 8)
}
figure <- function(df){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
labs(title = "Number of common ID", fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = 20)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
theme_bw()
p
}
new_lst[[1]]
names(new_lst[[1]])
figure <- function(df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
labs(title = "Number of common ID", fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
theme_bw()
p
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(new_lst[[i]], 35)
}
else{
p <- figure(new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0(get.data.dir(),"/plots/",sheetsnames[i] ,"_plot.png"),
width = 12, height = 8)
}
NED <- figure(new_lst$NED, 35)
NED <- figure(new_lst$NED, 35)
NED
ggsave(plot = NED,
filename = paste0(get.data.dir(),"/plots/","NED_" ,"_plot.png"),
width = 12, height = 8)
ggsave(plot = NED,
filename = paste0(get.data.dir(),"/plots/","NED_" ,"_plot.png"),
width = 15, height = 8)
ggsave(plot = NED,
filename = paste0(get.data.dir(),"/plots/","NED_" ,"_plot.png"),
width = 18, height = 8)
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(new_lst[[i]], 35)
}
else{
p <- figure(new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0(get.data.dir(),"/plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
labs(title = paste0("Plot of the Number of common ID for country:", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
theme_bw()
p
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0(get.data.dir(),"/plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
#labs(title = paste0("Plot of the Number of common ID for country: ", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
ggtitle(paste0("Plot of the number of common ID for country: ", country )) +
ggeasy::easy_center_title() +
theme_bw()
p
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0(get.data.dir(),"/plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
#labs(title = paste0("Plot of the Number of common ID for country: ", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
ggtitle(paste0("Plot of the number of common ID for country: ", country )) +
theme_update(plot.title = element_text(hjust = 0.5)) +
theme_bw()
p
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0(get.data.dir(),"/plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
#labs(title = paste0("Plot of the Number of common ID for country: ", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
ggtitle(paste0("Plot of the number of common ID for country: ", country )) +
theme(plot.title = element_text(hjust = 0.5)) +
theme_bw()
p
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0(get.data.dir(),"/plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
getwd()
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0("D:/public/data/fadn/lieferung_20210414/common_id_plots","/plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0("D:/public/data/fadn/lieferung_20210414/common_id_plots",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0("D:/public/data/fadn/lieferung_20210414/common_id_plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity",position = position_dodge(width = 0.8),width=0.5, fill = "limegreen" ) +
coord_flip()+
#labs(title = paste0("Plot of the Number of common ID for country: ", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
guides(fill=FALSE) +
ggtitle(paste0("Plot of the number of common ID for country: ", country )) +
theme(plot.title = element_text(hjust = 0.5)) +
theme_bw()
p
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0("D:/public/data/fadn/lieferung_20210414/common_id_plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
new_lst$BEL
figure("BEL",new_lst$BEL, 35)
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity", position = position_dodge(width = 0.8),width=0.5, fill = rainbow(n=length(df$Num_id)) ) +
coord_flip()+
#labs(title = paste0("Plot of the Number of common ID for country: ", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
ggtitle(paste0("Plot of the number of common ID for country: ", country )) +
theme(plot.title = element_text(hjust = 0.5)) +
theme_bw()
p
}
figure("BEL",new_lst$BEL, 35)
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity", position = position_dodge(width = 0.8),width=0.5, fill = rainbow(n=length(df$Num_id)) ) +
coord_flip()+
#labs(title = paste0("Plot of the Number of common ID for country: ", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
# other 20, NED = 35
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
ggtitle(paste0("Number of common ID for country: ", country )) +
theme(plot.title = element_text(hjust = 0.5)) +
theme_bw()
p
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0("D:/public/data/fadn/lieferung_20210414/common_id_plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity", position = position_dodge(width = 0.8),width=0.5, fill = rainbow(n=length(df$Num_id)) ) +
coord_flip()+
#labs(title = paste0("Plot of the Number of common ID for country: ", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
ggtitle(paste0("Number of common ID for country: ", country )) +
theme_bw() +
theme(plot.title = element_text(hjust = 0.5))
p
}
figure("BEL",new_lst$BEL, 35)
figure <- function(country, df, n){
p = df %>%
#ggplot( aes(x = reorder(Years, -Num_id) ,y=Num_id))
ggplot( aes(x = Years ,y=Num_id))+
geom_bar( stat="identity", position = position_dodge(width = 0.8),width=0.5, fill = rainbow(n=length(df$Num_id)) ) +
coord_flip()+
#labs(title = paste0("Plot of the Number of common ID for country: ", country ), fill = "Years") +
xlab("Years") +
ylab("Number of common ID") +
geom_text(aes(label=Num_id), vjust=0.5, colour="black", size=3.5)+
theme(axis.text.x=element_text(color = "black", size=6, angle=0, vjust=.8, hjust=0.8)) +
scale_x_discrete(labels = function(x) str_wrap(x, width = n)) +
facet_wrap( ~ group, scales="free",
labeller=names
#labeller(group = label_facet(df$group, "grouping"))
)+
ggtitle(paste0("Number of common ID for country: ", country )) +
theme_bw() +
theme(plot.title = element_text(hjust = 0.5))
p
}
for (i in 1:28){
if (sheetsnames[i] =="NED"){
p <- figure(sheetsnames[i],new_lst[[i]], 35)
}
else{
p <- figure(sheetsnames[i],new_lst[[i]], 20)
}
ggsave(plot = p,
filename = paste0("D:/public/data/fadn/lieferung_20210414/common_id_plots/",sheetsnames[i] ,"_plot.png"),
width = 18, height = 8)
}
knitr::opts_chunk$set(echo = TRUE)
summary(cars)
plot(pressure)
View(pressure)
knitr::opts_chunk$set(echo = TRUE)
file_names <- list.files("D:/public/data/fadn/lieferung_20210414/csv",include.dirs = TRUE)
file_names_gpg <- file_names[grepl(".gpg",file_names)]
length(file_names_gpg)
file_names <- file_names[!grepl(".gpg",file_names)]
library(tidyverse)
file_names <- file_names %>% tibble(name=.) %>% mutate(country=str_sub(name,1,3),year=str_sub(name,4,7))
file_names %>% print(n=30)
file_names %>% group_by(year) %>% count()
file_names %>% filter(country=="POL")
file_names <- file_names %>% mutate(provided="y")
data_requested_general <- read_xlsx(path = "D:/public/data/fadn/lieferung_20210414/FADN data request forms_March2021_MM210326.xlsx",sheet = 2,range = "A365:P393")
library(tidyverse)
library(readxl)
library(data.table)
data_requested_general <- read_xlsx(path = "D:/public/data/fadn/lieferung_20210414/FADN data request forms_March2021_MM210326.xlsx",sheet = 2,range = "A365:P393")
colnames(data_requested_general)[1] <- "country"
data_requested_general <- data_requested_general %>% mutate(country=str_sub(country,2,4))
data_requested_general_long <- data_requested_general %>% pivot_longer(-country)
data_requested_general_long <- data_requested_general_long %>% filter(value=="y") %>% rename(year=name,requested=value)
requested_and_provided <- data_requested_general_long %>% full_join(file_names %>% select(-name),by=c("country","year")) %>% filter(is.na(requested ) | is.na(provided))
## if empty, than everything is fine
requested_and_provided
# writexl::write_xlsx(x=requested_and_provided,path="D:/public/data/fadn/lieferung_20210414/missing_countries_years.xlsx")
files_list <- list()
folder_csv <- "D:/public/data/fadn/lieferung_20210414/csv/"
system.time({
files_list <- lapply(file_names %>% pull(name) %>% unique(),function(x)fread(input = paste0(folder_csv,x)))
}) ##   41.16 ~ 76 sec
files <- rbindlist(files_list)
files %>% select(YEAR,COUNTRY,countryyear)
## Number of sum of sample farms - observation - not unique
files %>% select(YEAR,COUNTRY) %>% group_by(COUNTRY) %>% count() %>% print(n=28)
## Combination of country and year in data set
files %>% select(YEAR,COUNTRY) %>% group_by(COUNTRY) %>% distinct()
library(fadnUtils)
library(fadnUtils)
library(fadnUtils)
library(fadnUtils)
library(fadnUtils)
library(fadnUtils)
devtools::load_all(".")
knitr::opts_chunk$set(echo = TRUE)
library(tidyverse)
library(readxl)
library(data.table)
library(skimr)
file_names <- list.files("D:/public/data/fadn/lieferung_20210414/csv",include.dirs = TRUE)
file_names_gpg <- file_names[grepl(".gpg",file_names)]
length(file_names_gpg)
file_names <- file_names[!grepl(".gpg",file_names)]
library(tidyverse)
file_names <- file_names %>% tibble(name=.) %>% mutate(country=str_sub(name,1,3),year=str_sub(name,4,7))
file_names %>% print(n=30)
file_names %>% group_by(year) %>% count()
file_names %>% filter(country=="POL")
file_names <- file_names %>% mutate(provided="y")
data_requested_general <- read_xlsx(path = "D:/public/data/fadn/lieferung_20210414/FADN data request forms_March2021_MM210326.xlsx",sheet = 2,range = "A365:P393")
colnames(data_requested_general)[1] <- "country"
data_requested_general <- data_requested_general %>% mutate(country=str_sub(country,2,4))
data_requested_general_long <- data_requested_general %>% pivot_longer(-country)
data_requested_general_long <- data_requested_general_long %>% filter(value=="y") %>% rename(year=name,requested=value)
requested_and_provided <- data_requested_general_long %>% full_join(file_names %>% select(-name),by=c("country","year")) %>% filter(is.na(requested ) | is.na(provided))
## if empty, than everything is fine
requested_and_provided
# writexl::write_xlsx(x=requested_and_provided,path="D:/public/data/fadn/lieferung_20210414/missing_countries_years.xlsx")