Commit | Line | Data |
---|---|---|
3d69ff21 BA |
1 | #' @title Acquire data in a clean format |
2 | #' | |
3 | #' @description Take in input data frames and/or files containing raw data, and timezones, and | |
4 | #' output a Data object, roughly corresponding to a list where each cell contains all value | |
25b75559 | 5 | #' for one day (see \code{?Data}). |
3d69ff21 BA |
6 | #' |
7 | #' @param ts_data Time-series, as a data frame (DB style: 2 columns, first is date/time, | |
8 | #' second is value) or a CSV file | |
9 | #' @param exo_data Exogenous variables, as a data frame or a CSV file; first comlumn is dates, | |
10 | #' next block are measurements for the day, and final block are exogenous forecasts | |
11 | #' @param input_tz Timezone in the input files ("GMT" or e.g. "Europe/Paris") | |
12 | #' @param date_format How date/time are stored (e.g. year/month/day hour:minutes; | |
13 | #' see \code{strptime}) | |
14 | #' @param working_tz Timezone to work with ("GMT" or e.g. "Europe/Paris") | |
09cf9c19 | 15 | #' @param predict_at When does the prediction take place ? Integer, in hours. Default: 0 |
f17665c7 | 16 | #' @param limit Number of days to extract (default: Inf, for "all") |
3d69ff21 | 17 | #' |
25b75559 BA |
18 | #' @return A list where data[[i]] contains |
19 | #' \itemize{ | |
20 | #' \item time: vector of times | |
21 | #' \item centered_serie: centered serie | |
22 | #' \item level: corresponding level | |
23 | #' \item exo: exogenous variables | |
24 | #' \item exo_hat: predicted exogenous variables | |
25 | #' } | |
3d69ff21 | 26 | #' |
44a9990b BA |
27 | #' @examples |
28 | #' ts_data = read.csv(system.file("extdata","pm10_mesures_H_loc.csv",package="talweg")) | |
29 | #' exo_data = read.csv(system.file("extdata","meteo_extra_noNAs.csv",package="talweg")) | |
25b75559 | 30 | #' data = getData(ts_data, exo_data, limit=120) |
3d69ff21 | 31 | #' @export |
1e20780e BA |
32 | getData = function(ts_data, exo_data, input_tz="GMT", date_format="%d/%m/%Y %H:%M", |
33 | working_tz="GMT", predict_at=0, limit=Inf) | |
3d69ff21 BA |
34 | { |
35 | # Sanity checks (not full, but sufficient at this stage) | |
36 | if (!is.character(input_tz) || !is.character(working_tz)) | |
37 | stop("Bad timezone (see ?timezone)") | |
38 | input_tz = input_tz[1] | |
39 | working_tz = working_tz[1] | |
6d97bfec BA |
40 | if ( (!is.data.frame(ts_data) && !is.character(ts_data)) || |
41 | (!is.data.frame(exo_data) && !is.character(exo_data)) ) | |
613a986f | 42 | stop("Bad time-series / exogenous input (data frame or CSV file)") |
3d69ff21 BA |
43 | if (is.character(ts_data)) |
44 | ts_data = ts_data[1] | |
6d97bfec BA |
45 | if (is.character(exo_data)) |
46 | exo_data = exo_data[1] | |
09cf9c19 BA |
47 | predict_at = as.integer(predict_at)[1] |
48 | if (predict_at<0 || predict_at>23) | |
49 | stop("Bad predict_at (0-23)") | |
3d69ff21 BA |
50 | if (!is.character(date_format)) |
51 | stop("Bad date_format (character)") | |
52 | date_format = date_format[1] | |
53 | ||
54 | ts_df = | |
613a986f BA |
55 | if (is.character(ts_data)) |
56 | read.csv(ts_data) | |
57 | else | |
3d69ff21 | 58 | ts_data |
3d69ff21 | 59 | exo_df = |
613a986f | 60 | if (is.character(exo_data)) |
3d69ff21 | 61 | read.csv(exo_data) |
613a986f | 62 | else |
3d69ff21 | 63 | exo_data |
3d69ff21 BA |
64 | # Convert to the desired timezone (usually "GMT" or "Europe/Paris") |
65 | formatted_dates_POSIXlt = strptime(as.character(ts_df[,1]), date_format, tz=input_tz) | |
66 | ts_df[,1] = format(as.POSIXct(formatted_dates_POSIXlt), tz=working_tz, usetz=TRUE) | |
67 | ||
3d69ff21 BA |
68 | line = 1 #index in PM10 file (24 lines for 1 cell) |
69 | nb_lines = nrow(ts_df) | |
70 | nb_exos = ( ncol(exo_df) - 1 ) / 2 | |
25b75559 | 71 | data = list() |
3d69ff21 BA |
72 | i = 1 #index of a cell in data |
73 | while (line <= nb_lines) | |
74 | { | |
75 | time = c() | |
76 | serie = c() | |
09cf9c19 | 77 | repeat |
3d69ff21 | 78 | { |
09cf9c19 BA |
79 | { |
80 | time = c(time, ts_df[line,1]) | |
81 | serie = c(serie, ts_df[line,2]) | |
82 | line = line + 1 | |
83 | }; | |
84 | if (line >= nb_lines + 1 || as.POSIXlt(ts_df[line-1,1])$hour == predict_at) | |
85 | break | |
86 | } | |
3d69ff21 | 87 | |
f17665c7 BA |
88 | exo = as.data.frame( exo_df[i,2:(1+nb_exos)] ) |
89 | exo_hat = as.data.frame( exo_df[i,(1+nb_exos+1):(1+2*nb_exos)] ) | |
3d69ff21 BA |
90 | level = mean(serie, na.rm=TRUE) |
91 | centered_serie = serie - level | |
25b75559 BA |
92 | data[[i]] = list("time"=time, "centered_serie"=centered_serie, "level"=level, |
93 | "exo"=exo, "exo_hat"=exo_hat) | |
1e20780e BA |
94 | if (i >= limit) |
95 | break | |
96 | i = i + 1 | |
3d69ff21 | 97 | } |
25b75559 BA |
98 | start = 1 |
99 | end = length(data) | |
100 | if (length(data[[1]]$centered_serie) < length(data[[2]]$centered_serie)) | |
101 | start = 2 | |
102 | if (length(data[[length(data)]]$centered_serie) < | |
103 | length(data[[length(data)-1]]$centered_serie)) | |
f17665c7 | 104 | { |
25b75559 | 105 | end = end-1 |
f17665c7 | 106 | } |
25b75559 BA |
107 | if (start>1 || end<length(data)) |
108 | data = data[start:end] | |
f17665c7 | 109 | data |
3d69ff21 | 110 | } |