Function to get data manipulated
seasonweekdata<- function(data1,coordinates=coor){
data1<-data1 %>% dplyr::select("StartStation.Id","EndStation.Id","Start.Date","StartStation.Name",
"EndStation.Name")
###merge coordinate
#########################################################################
a<- coor
colnames(a)<- c("StartStation.Id","s.lat","s.lon")
x<-join(data1,a)
aa<- coor
colnames(aa)<- c("EndStation.Id","e.lat","e.lon")
#a<-setnames(coor,old=c("id","lat","long"),new=c("EndStation.Id","e.lat","e.lon"))
#aa<-setnames(coor,old=c("id","lat","long"),new=c("StartStation.Id","s.lat","s.lon"))
z<-join(x,aa)
z<- na.omit(z)
x<- count(z,vars = c("StartStation.Id","EndStation.Id"))
data1<- join(z,x)
#################weekdays############################################################
data1$weekdays<-weekdays(
strptime(data1$Start.Date,format="%d/%m/%Y %H:%M",tz="UTC")
)
########################################seasons
return(data1)
##########same start and end stations #########################################
#attach(bikedata)
# bikedata$samestation<- rep(0,nrow(bikedata))
#head(bikedata)
#bikedata$samestation[which((StartStation.Id==EndStation.Id))]<- bikedata$freq[which((StartStation.Id==EndStation.Id))]
str(data1)
}
Function for plots
q<- sample_n(seasonweekdata(data1 = bikedata),6233518)
## Joining by: StartStation.Id
## Joining by: EndStation.Id
## Joining by: StartStation.Id, EndStation.Id
######################starting station
s.top<- dplyr::select(q,"s.lat","s.lon","StartStation.Name")
sc<- count(s.top$StartStation.Name)
colnames(sc)<- c("StartStation.Name","Freq")
scj<- join(sc,s.top)
## Joining by: StartStation.Name
scj<- scj[!duplicated(scj),]
scj<-arrange(scj,desc(Freq))[1:5,]
#########################ending station
e.top<- dplyr::select(q,"e.lat","e.lon","EndStation.Name")
ec<- count(e.top$EndStation.Name)
colnames(ec)<- c("EndStation.Name","Freq1")
ecj<- join(ec,e.top)
## Joining by: EndStation.Name
ecj<- ecj[!duplicated(ecj),]
ecj<-arrange(ecj,desc(Freq1))[1:5,]
Creating a frequency data
q1<- q[1:(nrow(q)/2),]
q2<- q[3166760:6233518,]
map<- get_map(location = c(lon =mean(q$s.lon),lat = mean(q$s.lat)),
source="google",maptype="roadmap",zoom=11)
## Map from URL : http://maps.googleapis.com/maps/api/staticmap?center=51.508906,-0.125838&zoom=11&size=640x640&scale=2&maptype=roadmap&language=en-EN&sensor=false
plot<- ggmap(map)+
geom_segment(data=q1,aes(x=s.lon,y=s.lat,xend=e.lon, yend=e.lat,alpha=freq),
colour= "black",size=0.00001)+
scale_alpha_continuous(range = c(0, 0.05))+
ggtitle("2017 bicycle trips")+
theme(plot.title = element_text(hjust = 0.5,size = 13))+
xlim(range(q$s.lon))+
ylim(range(q$s.lat))+
geom_point(data=q1,aes(x=s.lon,y=s.lat),size=0.001,alpha=0.3)+
geom_label_repel(data=scj, aes(x=s.lon,y=s.lat,label=StartStation.Name),
color="blue",nudge_y = 0.03,nudge_x = -0.1,label.size = 0.05,
size=5,alpha=0.6
)+
# arrow =arrow(length =unit(0.5,"cm")))+
geom_label_repel(data=ecj, aes(x=e.lon,y=e.lat,label=EndStation.Name),
color="red",segment.size = 0.5,segment.colour = "red",
#arrow = arrow(length = unit(0.5,"cm")),
nudge_y = -0.03,nudge_x = 0.15,size=5,alpha=0.6)
## Scale for 'x' is already present. Adding another scale for 'x', which
## will replace the existing scale.
## Scale for 'y' is already present. Adding another scale for 'y', which
## will replace the existing scale.
plot1<- plot + geom_segment(data=q2,aes(x=s.lon,y=s.lat,xend=e.lon, yend=e.lat,alpha=freq),
colour= "black",size=0.00001)+
scale_alpha_continuous(range = c(0, 0.05))
## Scale for 'alpha' is already present. Adding another scale for 'alpha',
## which will replace the existing scale.
plot1

#sfullplot(zoom = 11,
# data5= topplace(data4 = Summer))