dataviz_adh_ternum

---
title: "Valorisation de la donnée : Météo des adhérents"
author: "POC' Markdown JB"
date: "11/04/2018"
output:
  ioslides_presentation:
    smaller: yes
  slidy_presentation: default
  beamer_presentation: default
---

```{r setup, include=FALSE}
# knitr::opts_chunk$set(echo = TRUE)
#  https://stackoverflow.com/questions/25646333/code-chunk-font-size-in-rmarkdown-with-knitr-and-latex
library(ggplot2)
library(dplyr)
library(knitr)
library(dplyr)
library(XML)
library(plyr)
library(data.table)
library(dtplyr)
library(ggplot2)
library(xlsimple)
library(sqldf)
library(stringr)
library(tidyr)
library(readODS)
library(dismo)
library(XML)
library(stringi)
library(httr)
library(doBy)
library(scales)
library(banR)
library(readr)
library(readxl)
library(gridExtra)
library(treemapify)
library(forcats)
```

# R Markdown
```{r readin,  echo=FALSE}
df0 <- read_excel("C:/COPY_data_local/ternum/Liste adhérents pour météo_04_2018.xlsx")

df0[c("nature_ad")] <- NA
setnames(df0, old = 2, new = c('college'))
# df1 <- transform(df1,"samedi"= ifelse(jour4_f=="samedi" & jour4_o=="lundi" , lundi, samedi))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==1, 'Membres fondateurs', NA))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' >=2 &  df0$'college' <=5 | df0$'college' >=14 & df0$'college' <=17, 'Communes de moins de 3500 hab' , nature_ad))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==6, 'Communes de 3500 a 20000 hab', nature_ad))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==7, 'Communes plus de 20000 hab', nature_ad))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==8, 'Groupements de collectivités', nature_ad))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==9, 'Syndicat intercommunaux', nature_ad))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==10, 'Etablissement Recherche', nature_ad))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==11, 'Etablissement Santé', nature_ad))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==12, 'Sanitaires et Protection civile', nature_ad))
df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==13, 'Divers', nature_ad))

# df0[c("col_lab")] <- NA
# df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' ==1, 'Membres fondateurs', NA))
# df0 <- transform(df0, "nature_ad" = ifelse(df0$'college' >=2, 'Communes de moins de 3500 hab de Cote d''Or', NA))

df_ad <- subset(df0, df0$'college' != 1)
cat <- sapply(df_ad, is.factor)
df_ad[cat] <- lapply(df_ad[cat], factor)

```
Exemple de présentation utilisant R Markdown. Manière simple de créer des rapports automatisés en HTML, PDF, and MS Word documents. Plus de détails sur RMarkdown, regarder <http://rmarkdown.rstudio.com>.

# **Objectif** 
## Faire parler le tableur de Laetita **liste des adhérents** avec des quelques graphes et proposer quelques indicateurs
## Il sera présenter plusieurs graphes selon des **territoires** avec differents regroupements :  

>- collège (définition GIP)  
>- Département : on regroupe les adhérents selon leur appartenance à un département (Code Département)  
>- Nature : Un autre découpage simplifiant les collèges  

## Quelques chiffres clés
```{r chiffre_cle, echo=FALSE, warning=FALSE, collapse = TRUE, message = FALSE}
resume <- df0 %>%
  group_by(nature_ad) %>%
  summarize_each(funs(length, min, median, mean, max, sum), cotisation_annuelle)
names(resume) <- c("college","nombre","min","medianne","moyen","max", "somme")
resume$moyen <- round(resume$moyen,0)
resume$somme <- round(resume$somme,0)
resume2 <- resume[order(-resume$nombre),]
knitr::kable(resume2)
```


# graphes Nombres d'adhérents

## Nombre d'adhérents par college
```{r nb_adh_college, echo=FALSE}
h0 <- ggplot(data=df0, aes(x = factor(df0$'college'))) +geom_bar()
h0 + labs(fill= "Type local", x="college", y = "nombre") + geom_text(color = "white", stat='count', aes(label=..count..), vjust=1.25)
```

## En classant  
```{r nb_adh_college_horizon, echo=FALSE}
h0 <- ggplot(data=df0, aes(x = fct_infreq(as.factor(df0$college)))) +geom_bar()
# h0 <- ggplot(data=df0, aes(x = factor(df0$'college'))) +geom_bar()
h0 + labs(fill= "Type local", x="college", y = "nombre") + geom_text(color = "white", stat='count', aes(label=..count..), vjust=1.25)
```

## En classant WIP
```{r nb_adh_college_o, echo=FALSE}
h0 <- ggplot(data=df0, aes(x = fct_infreq(as.factor(df0$college)))) +geom_bar()
# h0 <- ggplot(data=df0, aes(x = factor(df0$'college'))) +geom_bar()
h1 <- h0 + labs(fill= "Type local", x="college", y = "nombre") + geom_text(color = "white", stat='count', aes(label=..count..), hjust=2) + coord_flip()
h1
```




## En couleur par département
```{r nb_adh_collge_dep, echo=FALSE, fig.cap = "..."}
h1 <- ggplot(data=df0, aes(x = factor(fct_infreq(as.factor(df0$college))), fill=factor(df0$CodeDepartement))) +geom_bar() 
# + facet_wrap( ~ df0$'CodeDepartement')
h2 <- h1 + theme(legend.position = c(0.8, 0.7), legend.box = "horizontal") + labs(fill= "Département", x="college", y = "nombre")
h2 + geom_text(size = 2.5, color = "white", stat='count', aes(label=..count..), position = position_stack(vjust = 0.5))
```


## Par nature d'adherents
```{r nb_adh_newtypo_horizontal, echo=FALSE, warning=FALSE}
f1 <- ggplot(data=df0, aes(x = factor(fct_infreq(as.factor(df0$'nature_ad'))))) +geom_bar()
f2 <- f1 + labs(fill= "Type local", x="Nature", y = "nombre") + geom_text(color = "white", stat='count', aes(label=..count..), hjust=1.25, size = 4) 
f2 + scale_x_discrete()  + coord_flip()
# https://stackoverflow.com/questions/30614105/create-abbreviated-legends-manually-for-long-x-labels-in-ggplot2
```

## Par nature d'adherents
```{r nb_adh_newtypo_0, echo=FALSE, warning=FALSE}
f1 <- ggplot(data=df0, aes(x = factor(fct_infreq(as.factor(df0$'nature_ad'))))) +geom_bar()
f2 <- f1 + labs(fill= "Type local", x="Nature", y = "nombre") + geom_text(color = "white", stat='count', aes(label=..count..), vjust=1.5) 
f2 + scale_x_discrete(labels = abbreviate) +  theme(legend.position = c(0.8, 0.7))
# https://stackoverflow.com/questions/30614105/create-abbreviated-legends-manually-for-long-x-labels-in-ggplot2
```

## Nouveau par nature d'adherents
```{r nb_adh_newtypo_1, echo=FALSE, warning=FALSE}
h3 <- ggplot(data=df0, aes(x = factor(fct_infreq(as.factor(df0$'nature_ad'))), fill=factor(df0$CodeDepartement))) +geom_bar() 
# + facet_wrap( ~ df0$'CodeDepartement')
h4 <- h3 + theme(legend.position = c(0.8, 0.7), legend.box = "horizontal") + labs(fill= "Département", x="Nature", y = "nombre")
h4 + guides(col = guide_legend(nrow = 2)) + geom_text(size = 2.5, color = "white", stat='count', aes(label=..count..), position = position_stack(vjust = 0.5)) + scale_x_discrete(labels = abbreviate)
```


## Graphique circulaire dép
```{r pie_dep, echo=FALSE, warning=FALSE}
dfp <- as.data.frame(table(df0$'CodeDepartement'))
dfp <- subset(dfp, dfp$Freq != 0)
# https://stackoverflow.com/questions/47752037/pie-chart-with-ggplot2-with-specific-order-and-percentage-annotations
# changement ordre
dfp$Var1 <- factor(dfp$Var1, levels = rev(as.character(dfp$Var1)))
# https://stackoverflow.com/questions/18209061/how-to-calculate-percentage-from-a-vector-of-counts
dfp$percent = round((dfp$Freq / sum(dfp$Freq)) * 100 , 0)
p0 <- ggplot(data=dfp, aes(x="", y = Freq, fill= factor(Var1))) +geom_bar(width=1, stat="identity")
# https://stackoverflow.com/questions/28912059/labels-on-the-pie-chart-for-small-pieces-ggplot
p0 +labs(fill="") + coord_polar("y", start=0) + labs(fill= "Dép", x="", y = "") + geom_text(color = "white", aes(label=paste(Freq, paste(percent, '%', sep=' '), sep='\n') , y=Freq / 2 + c(0, cumsum(Freq)[-length(Freq)]))) + theme(legend.position = "right",   ### Solution to part 1, no legend
        axis.ticks = element_blank(), 
        panel.grid = element_blank(), 
        axis.text  = element_blank())

```

## Graphique circulaire nature
```{r pie_nat, echo=FALSE, warning=FALSE}
dfp <- as.data.frame(table(df0$nature_ad))
dfp <- subset(dfp, dfp$Freq != 0)
# https://stackoverflow.com/questions/47752037/pie-chart-with-ggplot2-with-specific-order-and-percentage-annotations
# changement ordre
dfp$Var1 <- factor(dfp$Var1, levels = rev(as.character(dfp$Var1)))
# https://stackoverflow.com/questions/18209061/how-to-calculate-percentage-from-a-vector-of-counts
dfp$percent = round((dfp$Freq / sum(dfp$Freq)) * 100 , 0)
p0 <- ggplot(data=dfp, aes(x="", y = Freq, fill= factor(Var1))) +geom_bar(width=1, stat="identity")
# https://stackoverflow.com/questions/28912059/labels-on-the-pie-chart-for-small-pieces-ggplot
p0 +labs(fill="") + coord_polar("y", start=0) + labs(fill= "Nature", x="", y = "") + geom_text(color = "white", aes(label=paste(Freq, paste(percent, '%', sep=' '), sep='\n') , y=Freq / 2 + c(0, cumsum(Freq)[-length(Freq)]))) + theme(legend.position = "right",   ### Solution to part 1, no legend
        axis.ticks = element_blank(), 
        panel.grid = element_blank(), 
        axis.text  = element_blank())

```

## Graphique circulaire college
```{r pie_college, echo=FALSE, warning=FALSE}
dfp <- as.data.frame(table(df0$college))
dfp <- subset(dfp, dfp$Freq != 0)
# https://stackoverflow.com/questions/47752037/pie-chart-with-ggplot2-with-specific-order-and-percentage-annotations
# changement ordre
dfp$Var1 <- factor(dfp$Var1, levels = rev(as.character(dfp$Var1)))
# https://stackoverflow.com/questions/18209061/how-to-calculate-percentage-from-a-vector-of-counts
dfp$percent = round((dfp$Freq / sum(dfp$Freq)) * 100 , 0)
p0 <- ggplot(data=dfp, aes(x="", y = Freq, fill= factor(Var1))) +geom_bar(width=1, stat="identity")
# https://stackoverflow.com/questions/28912059/labels-on-the-pie-chart-for-small-pieces-ggplot
p0 +labs(fill="") + coord_polar("y", start=0, direction=1) + labs(fill= "college", x="", y = "") + geom_text(color = "white", aes(label=paste(Freq, paste(percent, '%', sep=' '), sep='\n') , y=Freq / 2 + c(0, cumsum(Freq)[-length(Freq)]))) + theme(legend.position = "right",   ### Solution to part 1, no legend
        axis.ticks = element_blank(), 
        panel.grid = element_blank(), 
        axis.text  = element_blank())
```

# graphes Cotisation

## Distribution du montant de Cotisation

```{r partshown, echo=FALSE}
g1 <-ggplot(df0, aes(x=df0$'cotisation_annuelle')) + geom_density() + labs(x="Costisation", y = "Part")
g1 + scale_x_continuous(breaks = c(0, 20000 , 100000 , 200000))

```
Ouch, on ne voit rien car les membres fondateurs sont également inclus dans le jeu de données

## Distribution Cotisations Adhérents

```{r densite1, echo=FALSE}
h2c <- ggplot(data=df_ad) +geom_density(data=df_ad, aes(df_ad$'cotisation_annuelle'), adjust = 1.5, size=1) + labs(x="Montant de Cotisation", y = "densité")
h2c
```


## Distribution Cotisations Adhérents

```{r densite21, echo=FALSE}
p<-ggplot(df_ad, aes(x=df_ad$cotisation_annuelle)) +
  geom_density(adjust = 1.5, size=1)+
  geom_vline(data=df_ad, aes(xintercept=mean(df_ad$cotisation_annuelle), colour="moyenne 1367 euros"),
             linetype="dashed", size=1.25)+
  geom_vline(data=df_ad, aes(xintercept=median(df_ad$cotisation_annuelle), colour="medianne 325 euros"),
             linetype="dashed", size=1.25)
p1 <- p + theme(legend.position = c(0.8, 0.8))  + labs(fill= "Département", x="Montant", y = "nombre")
p1 + scale_color_manual(values=c("red", "darkgreen", "gray"), name="légende") + scale_x_continuous(breaks = c(0, 1500, 2500, 5000 , 7500, 10000, 12500, 15000, 17500, 20000 ))

```

## Cela veut dire :
>- **médiane** : La **moitié (50%)** des adhérentes payent une cotisation inférieure à **325 euros**  
 l'autre moitié payent **plus de 325 euros** de cotisation  
\newline  
>- **moyenne** : En moyenne, un adhérent paye **1367 euros** en cotisation annuelle  
 Etant donne la distribution, une moyenne ne donne pas trop de **sens**.  
>- Autres remarques : **155 Syndicats** et assimilé ne payent pas de cotisation.

## Quelques chiffres cle




## En affinant

```{r densite41, echo=FALSE}
h2c <- ggplot(data=df_ad) +geom_density(data=df_ad, aes(df_ad$'cotisation_annuelle'), adjust = 1.5, size=1) + labs(x="Montant de Cotisation", y = "densité")
h2c + facet_grid(.~df_ad$nature_ad) + coord_cartesian(xlim = c(0, 5000))
```

## En affinant stacked density

```{r densite42, echo=FALSE}
h2c <- ggplot(data=df_ad) +geom_density(data=df_ad, aes(df_ad$'cotisation_annuelle',fill=df_ad$nature_ad), position="stack", adjust = 1.5, size=1) + labs(x="Montant de Cotisation", y = "densité")
h2c +coord_cartesian(xlim=c(0,2500)) + labs(fill= "Nature", x="Cotisation Montant", y = "Densité")
```


## Pour aller un peu plus loin
```{r densite3, echo=FALSE}
g1 <- ggplot(df_ad, aes(x=df_ad$cotisation_annuelle, ..count.., fill=factor(df_ad$nature_ad))) +
  geom_density(position = "fill") + scale_fill_brewer(palette = "Set1")
g1 + theme(legend.position = "bottom", legend.text=element_text(size=6))  + labs(fill= "Nature", x="Montant", y = "pourcentage") + scale_x_continuous(breaks = c(0, 1500, 2500, 5000 , 7500, 10000, 12500, 15000, 17500, 20000 , 22500 ))

```

## Ce que montre ce graphe
>- En parcours l'axe des absisses (Pour un montant de cotisation donnée), on a la compistion.
>- Il y a :
>- Jusqu'à **5000 euros**, les adhérents sont en grande majorité des **communes de 3500 hab**.
>- Au dela de , ce sont les **communes de 3500 à 20000 hab**
>- au dela, ce sont les **groupement de communes**

# Contributions

## On regarde la somme des cotisations par territoire  
Membres fondateurs et Adhérents  

## Contributions financières
```{r nb_cotis_col, echo=FALSE, warning=FALSE}
# tableau recap et GRAPHE
bil_nat <- summaryBy(cotisation_annuelle  ~ nature_ad , data = df0, FUN=c(sum,mean,length), fun.names=c('cotis_som','cotis_mean', 'cotis_nb'))
bil_col <- summaryBy(cotisation_annuelle  ~ df0$college , data = df0, FUN=c(sum,mean,length), fun.names=c('cotis_som','cotis_mean', 'cotis_nb'))
bil_col$percent = round((bil_col$cotisation_annuelle.cotis_som / sum(bil_col$cotisation_annuelle.cotis_som)) * 100 , 1)

h3 <- ggplot(data=bil_col, aes(x = factor(bil_col$'college'), y=bil_col$cotisation_annuelle.cotis_som )) +geom_bar(stat="identity") 
h4 <- h3 + labs(fill= "Département", x="Collège", y = "montant") + scale_y_continuous(breaks = c(10000, 250000, 500000 , 1000000, 1250000 , 1500000 , 1800000))
h4 + geom_text(size = 3, color = "white", stat='identity', aes(label=(paste(as.character(percent),"%",sep=' '))), position = position_stack(vjust = 0.5))

```

## Contributions financières dép
```{r nb_cotis_col_dep, echo=FALSE, warning=FALSE}
# tableau recap et GRAPHE
bil_dep <- summaryBy(cotisation_annuelle  ~ CodeDepartement , data = df0, FUN=c(sum,mean,length), fun.names=c('cotis_som','cotis_mean', 'cotis_nb'))
bil_dep$percent = round((bil_dep$cotisation_annuelle.cotis_som / sum(bil_dep$cotisation_annuelle.cotis_som)) * 100 , 1)

h3 <- ggplot(data=bil_dep, aes(x = factor(bil_dep$CodeDepartement), y=bil_dep$cotisation_annuelle.cotis_som )) +geom_bar(stat="identity") 
h4 <- h3 + labs(fill= "Département", x="Département", y = "montant") + scale_y_continuous(breaks = c(10000, 250000, 500000 , 1000000, 1250000 , 1500000 , 1800000)) +  scale_x_discrete(labels = abbreviate)
h4 + geom_text(size = 3, color = "white", stat='identity', aes(label=(paste(as.character(percent),"%",sep=' '))), position = position_stack(vjust = 0.5))

```
  
## Explications  
Le Doubs cotiserait à hauteur de **19,6** % à Ternum !?  
N'importe quoi me direz vous !  
Il faut regarder la donnée et constater que la Région  
a pour codedépartement : 25

## Contributions financières nature
```{r nb_cotis_col_nat, echo=FALSE, warning=FALSE}
# tableau recap et GRAPHE
bil_nat <- summaryBy(cotisation_annuelle  ~ nature_ad , data = df0, FUN=c(sum,mean,length), fun.names=c('cotis_som','cotis_mean', 'cotis_nb'))
bil_nat$percent = round((bil_nat$cotisation_annuelle.cotis_som / sum(bil_nat$cotisation_annuelle.cotis_som)) * 100 , 1)

h3 <- ggplot(data=bil_nat, aes(x = factor(bil_nat$'nature_ad'), y=bil_nat$cotisation_annuelle.cotis_som )) +geom_bar(stat="identity") 
h4 <- h3 + labs(fill= "Département", x="Nature", y = "montant") + scale_y_continuous(breaks = c(10000, 250000, 500000 , 1000000, 1250000 , 1500000 , 1800000)) +  scale_x_discrete(labels = abbreviate)
h4 + geom_text(size = 3, color = "white", stat='identity', aes(label=(paste(as.character(percent),"%",sep=' '))), position = position_stack(vjust = 0.5))

```


# Voir Cotisation et les territoires
## cotisations par departement illisible
```{r point00, echo=FALSE, message=FALSE,}
b2 <- ggplot(data=df_ad, aes(y=df_ad$'cotisation_annuelle', x=factor(df_ad$'CodeDepartement'))) + geom_point()
b2 + labs(x="département", y = "montant de cotisation")
```

## cotisations par departement illisible
```{r point01, echo=FALSE, message=FALSE,}
b2 <- ggplot(data=df_ad, aes(y=df_ad$'cotisation_annuelle', x=factor(df_ad$'CodeDepartement'))) + geom_point(alpha=0.1)
b2 + labs(x="département", y = "montant de cotisation") +geom_jitter(alpha=0.1)
```

## cotisations dispersion
```{r boxplot2, echo=FALSE, message=FALSE,}
b2 <- ggplot(data=df_ad) +geom_boxplot(data=df_ad, aes(y=df_ad$'cotisation_annuelle', x=factor(df_ad$'CodeDepartement')))
b2 + labs(x="département", y = "montant de cotisation")
```

## Explications :
Boites englobent **95 % des adhérents** pour chaque **territoire** 
  les **barres noires** représentent les **médianes** pour chaque **territoire**  
  

## cotisations dispersion nature
```{r boxplot_nature, echo=FALSE, message=FALSE, warning=FALSE}
b2 <- ggplot(data=df_ad) +geom_boxplot(data=df_ad, aes(y=df_ad$'cotisation_annuelle', x=factor(df_ad$nature_ad)))
b2 + labs(x="Nature", y = "montant de cotisation")  + scale_x_discrete(labels = abbreviate)
```

# Relation Cotisation / Nombre d'habitants 
## Modele éco' en graphes
```{r coti_hab, echo=FALSE}
p1 <- ggplot(data=df_ad) + geom_point(aes(x=df_ad$cotisation_annuelle , y=df_ad$NbreHabitants)) + theme(legend.position = "bottom")
p1 + theme(legend.position = "right") + labs(color= "college", x="Montant de Cotisation", y = "nombre d'habitants")
```

## On differencie avec les colleges
```{r coti_hab_col, echo=FALSE}
p2 <- ggplot(data=df_ad) + geom_point(aes(x=df_ad$cotisation_annuelle , y=df_ad$NbreHabitants , color=factor(df_ad$'college'))) + theme(legend.position = "right") 
p2 + theme(legend.position = "right") + labs(color= "college", x="Montant de Cotisation", y = "nombre d'habitants")
```

## On differencie avec la nature
```{r coti_hab_nat, echo=FALSE}
p3 <- ggplot(data=df_ad) + geom_point(aes(x=df_ad$cotisation_annuelle , y=df_ad$NbreHabitants , color=factor(df_ad$'nature_ad'))) + theme(legend.position = "right")
p3 + theme(legend.position = "right") + labs(color= "nature", x="Montant de Cotisation", y = "nombre d'habitants")
```

## en y ajoutant une courbe
```{r coti_hab_col_courbe, echo=FALSE, message=FALSE,}
p3 <- p2 + geom_smooth(aes(x=df_ad$cotisation_annuelle , y=df_ad$NbreHabitants), span = 0.05)
p3 + theme(legend.position = "right") + labs(color= "college", x="Montant de Cotisation", y = "nombre d'habitants")
```


## Treemap Surface Habitants
```{r figdim2, echo=FALSE, message=FALSE,}
df_ad <- subset(df0, df0$'college' != 1)
cat <- sapply(df_ad, is.factor)
recap <- summaryBy(cotisation_annuelle + NbreHabitants ~ nature_ad , data = df_ad, FUN=c(sum,mean,length))

g1 <- ggplot(recap, aes(fill=recap$cotisation_annuelle.sum, area=recap$NbreHabitants.sum, label= recap$nature_ad)) +geom_treemap() + 
  geom_treemap_text(fontface = "italic", colour = "black", place = "centre", grow = TRUE) + theme(legend.direction = 'horizontal' , legend.position = "bottom") +
  scale_fill_gradient(low="white",high="red") + labs(fill= "Montant Cotisation")
g1
# + scale_fill_conti(palette = "Spectral")
# A ameliorer

```


## Treemap Surface Cotisation
```{r figdim3, echo=FALSE, message=FALSE,}
df_ad <- subset(df0, df0$'college' != 1)
cat <- sapply(df_ad, is.factor)
recap <- summaryBy(cotisation_annuelle + NbreHabitants ~ nature_ad , data = df_ad, FUN=c(sum,mean,length))

g1 <- ggplot(recap, aes(area=recap$cotisation_annuelle.sum, fill=recap$NbreHabitants.sum, label= recap$nature_ad)) +geom_treemap() + 
  geom_treemap_text(fontface = "italic", colour = "black", place = "centre", grow = TRUE) + theme(legend.direction = 'horizontal' , legend.position = "bottom") +
  scale_fill_gradient(low="white",high="red") + labs(fill= "Montant Cotisation")
g1 +  theme(axis.text.y = element_text(angle = 90, hjust = 1))
# + scale_fill_conti(palette = "Spectral")
# A ameliorer

```


## Au final
>- Des indicateurs parlants à tous à trouver
>- Des graphes à garder, jeter ou améliorer
>- Présentation et Mise en Page à améliorer
>- Prendre garde au contexte de la donnée source
>- A échanger, à discuter !  

## TO COMPLETE :

>- Cumul des montant d'adhésion selon la date
>- *** Corrections PR ***
>- Communiquer sur V2
```{r anim_note, echo=FALSE, message=FALSE,}
# https://grunwaldlab.github.io/Reproducible-science-in-R/Extra_content---Advanced_RMarkdown.html
# https://www.r-bloggers.com/animated-plots-with-r/
```