machinelearning_R/awsPrepareData.r at master · net-map/machinelearning_R · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124

setwd("../machinelearning_R")
path<-paste(getwd(),"serverFunctions.r",sep="/")

invisible(source(path))


#recieve input facility name from command line
args = commandArgs(trailingOnly=TRUE)
#Rscript awsPrepareData.r facilityName
#default values
if (length(args)==0){
  args <- "Brocolândia"
}


invisible(source("serverFunctions.r"))

mongo <- mongo.create(host="localhost:27017",username="netmap",password = "p4gic0tb9f2m2yj37iav")

if(mongo.is.connected(mongo) == TRUE) {
  db <- "server_api_production"


  facilities <- paste(db,"facilities",sep = ".")

  #retrive list of facilities that match query and get 1st element which is the ID
  facilityID<-mongo.find.all(mongo,facilities,list(name=args))[[1]][[1]]


  zones <- paste(db,"zones",sep = ".")


  #retrive all zones from that facility
  listZones <- mongo.find.all(mongo,zones,list(facility_id=facilityID))

  rawData <- NULL

  #for each zone found
  for (z in 1:length(listZones)){

    #get ID of that zone from list of lists
    zoneID <- listZones[[z]][[1]]

    acquisition <- paste(db,"acquisitions",sep = ".")

    listAcquisitions <- mongo.find.all(mongo,acquisition,list(zone_id=zoneID))

    lista <- NULL

    for (a in 1:length(listAcquisitions)){
      listAP <- listAcquisitions[[a]]$access_points
      acquiID <- listAcquisitions[[a]]$`_id`
      #get relevant informantion from list
      temp <- lapply(listAP,function(x) return (c(x$BSSID,x$RSSI)))

      #reshape list and add relevant IDs
      temp2<- lapply(temp,montaLista,zoneID=zoneID,acquiID=acquiID)

      #create list in desirable format

      #unlist lists
      for (l in 1:length(temp2)){
         lista <- rbind(lista,unlist(temp2[[l]]))
      }


    }

    rawData <- rbind(rawData,lista)

  }


  #we DONT want strings to be turned to factors!
  rawDataDF<-data.frame(rawData,stringsAsFactors = FALSE)

  names(rawDataDF)<- c("BSSID","RSSI","idZ","acquiID")

  rawDataDF$RSSI <- as.numeric(rawDataDF$RSSI)


  molten <- melt(rawDataDF,id.vars=c("idZ","acquiID","BSSID"), value.name = "RSSI",measure.vars = c("RSSI"))

  molten$variable <- NULL

  #gambiarra
  #molten <- molten[-14,]

  attach(rawDataDF)
  tidyData <- reshape2::dcast(molten,   acquiID+ idZ  ~ BSSID, value.var = 'RSSI')
  detach(rawDataDF)


}else{
  print("Could not connect to Mongo! DAMN IT LIRA")
}


#convert to numbers
#zones <- as.numeric(args)

#source("serverFunctions.r")


dataPath <- "prepared-data"
#remove Aquisition ID, as we don't really need it from now on
tidyData<- tidyData[,-1]


#transform idZ into factor!
tidyData[,1] <- as.factor(tidyData[,1])


saveRDS(tidyData,paste(dataPath,"/",args,".rds",sep=""))