pkg/R/plot.R

   1 #' @title plot curves
   2 #'
   3 #' @description Plot a range of curves in data
   4 #'
   5 #' @param data Object of class Data
   6 #' @param indices Range of indices (integers or dates)
   7 #'
   8 #' @export
   9 plotCurves <- function(data, indices=seq_len(data$getSize()))
  10 {
  11     yrange = quantile( range( sapply( indices, function(i) {
  12         serie = c(data$getCenteredSerie(i))
  13         if (!all(is.na(serie)))
  14             range(serie, na.rm=TRUE)
  15         c()
  16     }) ), probs=c(0.05,0.95) )
  17     par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
  18     for (i in seq_along(indices))
  19     {
  20         plot(data$getSerie(indices[i]), type="l", ylim=yrange,
  21             xlab=ifelse(i==1,"Temps (en heures)",""), ylab=ifelse(i==1,"PM10",""))
  22         if (i < length(indices))
  23             par(new=TRUE)
  24     }
  25 }
  26
  27 #' @title plot measured / predicted
  28 #'
  29 #' @description Plot measured curve (in black) and predicted curve (in red)
  30 #'
  31 #' @param data Object return by \code{getData}
  32 #' @param pred Object as returned by \code{getForecast}
  33 #' @param index Index in forecasts
  34 #'
  35 #' @export
  36 plotPredReal <- function(data, pred, index)
  37 {
  38     horizon = length(pred$getSerie(1))
  39     measure = data$getSerie(pred$getIndexInData(index)+1)[1:horizon]
  40     yrange = range( pred$getSerie(index), measure )
  41     par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lwd=3)
  42     plot(measure, type="l", ylim=yrange, xlab="Temps (en heures)", ylab="PM10")
  43     par(new=TRUE)
  44     plot(pred$getSerie(index), type="l", col="#0000FF", ylim=yrange, xlab="", ylab="")
  45 }
  46
  47 #' @title Compute filaments
  48 #'
  49 #' @description Get similar days in the past + "past tomorrow", as black as distances are small
  50 #'
  51 #' @param data Object as returned by \code{getData}
  52 #' @param index Index in data
  53 #' @param limit Number of neighbors to consider
  54 #' @param plot Should the result be plotted?
  55 #'
  56 #' @export
  57 computeFilaments <- function(data, index, limit=60, plot=TRUE)
  58 {
  59     index = dateIndexToInteger(index, data)
  60     ref_serie = data$getCenteredSerie(index)
  61     if (any(is.na(ref_serie)))
  62         stop("computeFilaments requires a serie without NAs")
  63     L = length(ref_serie)
  64
  65     # Determine indices of no-NAs days followed by no-NAs tomorrows
  66     first_day = ifelse(length(data$getCenteredSerie(1))<L, 2, 1)
  67     fdays_indices = c()
  68     for (i in first_day:(index-1))
  69     {
  70         if ( !any(is.na(data$getSerie(i)) | is.na(data$getSerie(i+1))) )
  71             fdays_indices = c(fdays_indices, i)
  72     }
  73
  74     distances = sapply(fdays_indices, function(i) {
  75         sqrt( sum( (ref_serie - data$getCenteredSerie(i))^2 ) / L )
  76     })
  77     indices = sort(distances, index.return=TRUE)$ix[1:min(limit,length(distances))]
  78     yrange = quantile( range( ref_serie, sapply( indices, function(i) {
  79         ii = fdays_indices[i]
  80         serie = c(data$getCenteredSerie(ii), data$getCenteredSerie(ii+1))
  81         if (!all(is.na(serie)))
  82             return (range(serie, na.rm=TRUE))
  83         c()
  84     }) ), probs=c(0.05,0.95) )
  85     grays = gray.colors(20, 0.1, 0.9) #TODO: 20 == magic number
  86     min_dist = min(distances[indices])
  87     max_dist = max(distances[indices])
  88     color_values = floor( 19.5 * (distances[indices]-min_dist) / (max_dist-min_dist) ) + 1
  89     plot_order = sort(color_values, index.return=TRUE, decreasing=TRUE)$ix
  90     colors = c(grays[ color_values[plot_order] ], "#FF0000")
  91     if (plot)
  92     {
  93         par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lwd=2)
  94         for ( i in seq_len(length(indices)+1) )
  95         {
  96             ii = ifelse(i<=length(indices), fdays_indices[ indices[plot_order[i]] ], index)
  97             plot(c(data$getCenteredSerie(ii),data$getCenteredSerie(ii+1)),
  98                 ylim=yrange, type="l", col=colors[i],
  99                 xlab=ifelse(i==1,"Temps (en heures)",""), ylab=ifelse(i==1,"PM10 centré",""))
 100             if (i <= length(indices))
 101                 par(new=TRUE)
 102         }
 103         abline(v=24, lty=2, col=colors()[56])
 104     }
 105     list("indices"=c(fdays_indices[ indices[plot_order] ],index), "colors"=colors)
 106 }
 107
 108 #' @title Plot similarities
 109 #'
 110 #' @description Plot histogram of similarities (weights)
 111 #'
 112 #' @param pred Object as returned by \code{getForecast}
 113 #' @param index Index in forecasts (not in data)
 114 #'
 115 #' @export
 116 plotSimils <- function(pred, index)
 117 {
 118     weights = pred$getParams(index)$weights
 119     if (is.null(weights))
 120         stop("plotSimils only works on 'Neighbors' forecasts")
 121     par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
 122     hist(pred$getParams(index)$weights, nclass=20, xlab="Poids", ylab="Effectif")
 123 }
 124
 125 #' @title Plot error
 126 #'
 127 #' @description Draw error graphs, potentially from several runs of \code{getForecast}
 128 #'
 129 #' @param err Error as returned by \code{getError}
 130 #' @param cols Colors for each error (default: 1,2,3,...)
 131 #'
 132 #' @seealso \code{\link{plotPredReal}}, \code{\link{plotFilaments}}, \code{\link{plotSimils}}
 133 #'   \code{\link{plotFbox}}
 134 #'
 135 #' @export
 136 plotError <- function(err, cols=seq_along(err))
 137 {
 138     if (!is.null(err$abs))
 139         err = list(err)
 140     par(mfrow=c(2,2), mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lwd=2)
 141     L = length(err)
 142     yrange = range( sapply(1:L, function(index) ( err[[index]]$abs$day ) ), na.rm=TRUE )
 143     for (i in seq_len(L))
 144     {
 145         plot(err[[i]]$abs$day, type="l", xlab=ifelse(i==1,"Temps (heures)",""),
 146             ylab=ifelse(i==1,"Moyenne |y - y_hat|",""), ylim=yrange, col=cols[i])
 147         if (i < L)
 148             par(new=TRUE)
 149     }
 150     yrange = range( sapply(1:L, function(index) ( err[[index]]$abs$indices ) ), na.rm=TRUE )
 151     for (i in seq_len(L))
 152     {
 153         plot(err[[i]]$abs$indices, type="l", xlab=ifelse(i==1,"Temps (jours)",""),
 154             ylab=ifelse(i==1,"Moyenne |y - y_hat|",""), ylim=yrange, col=cols[i])
 155         if (i < L)
 156             par(new=TRUE)
 157     }
 158     yrange = range( sapply(1:L, function(index) ( err[[index]]$MAPE$day ) ), na.rm=TRUE )
 159     for (i in seq_len(L))
 160     {
 161         plot(err[[i]]$MAPE$day, type="l", xlab=ifelse(i==1,"Temps (heures)",""),
 162             ylab=ifelse(i==1,"MAPE moyen",""), ylim=yrange, col=cols[i])
 163         if (i < L)
 164             par(new=TRUE)
 165     }
 166     yrange = range( sapply(1:L, function(index) ( err[[index]]$MAPE$indices ) ), na.rm=TRUE )
 167     for (i in seq_len(L))
 168     {
 169         plot(err[[i]]$MAPE$indices, type="l", xlab=ifelse(i==1,"Temps (jours)",""),
 170             ylab=ifelse(i==1,"MAPE moyen",""), ylim=yrange, col=cols[i])
 171         if (i < L)
 172             par(new=TRUE)
 173     }
 174 }
 175
 176 #' @title Functional boxplot
 177 #'
 178 #' @description Draw the functional boxplot on the left, and bivariate plot on the right
 179 #'
 180 #' @param data Object return by \code{getData}
 181 #' @param fiter Optional filter: return TRUE on indices to process
 182 #' @param plot_bivariate Should the bivariate plot appear?
 183 #'
 184 #' @export
 185 plotFbox <- function(data, filter=function(index) TRUE, plot_bivariate=TRUE)
 186 {
 187     if (!requireNamespace("rainbow", quietly=TRUE))
 188         stop("Functional boxplot requires the rainbow package")
 189
 190     start_index = 1
 191     end_index = data$getSize()
 192     if (length(data$getCenteredSerie(1)) < length(data$getCenteredSerie(2)))
 193     {
 194         # Shifted start (7am, or 1pm, or...)
 195         start_index = 2
 196         end_index = data$getSize() - 1
 197     }
 198
 199     series_matrix = sapply(start_index:end_index, function(index) {
 200         as.matrix(data$getSerie(index))
 201     })
 202     # Remove NAs. + filter TODO: merge with previous step: only one pass required...
 203     nas_indices = seq_len(ncol(series_matrix))[ sapply( 1:ncol(series_matrix),
 204         function(index) ( !filter(index) || any(is.na(series_matrix[,index])) ) ) ]
 205     series_matrix = series_matrix[,-nas_indices]
 206
 207     series_fds = rainbow::fds(seq_len(nrow(series_matrix)), series_matrix)
 208     if (plot_bivariate)
 209         par(mfrow=c(1,2))
 210     par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
 211     rainbow::fboxplot(series_fds, "functional", "hdr", xlab="Temps (heures)", ylab="PM10",
 212         plotlegend=FALSE, lwd=2)
 213     if (plot_bivariate)
 214         rainbow::fboxplot(series_fds, "bivariate", "hdr", plotlegend=FALSE)
 215 }
 216
 217 #' @title Functional boxplot on filaments
 218 #'
 219 #' @description Draw the functional boxplot on filaments obtained by \code{computeFilaments}
 220 #'
 221 #' @param data Object return by \code{getData}
 222 #' @param indices Indices as output by \code{computeFilaments}
 223 #'
 224 #' @export
 225 plotFilamentsBox = function(data, indices, ...)
 226 {
 227     past_neighbs_indices = head(indices,-1)
 228     plotFbox(data, function(i) i %in% past_neighbs_indices, plot_bivariate=FALSE)
 229     par(new=TRUE)
 230     # "Magic" found at http://stackoverflow.com/questions/13842560/get-xlim-from-a-plot-in-r
 231     usr <- par("usr")
 232     yr <- (usr[4] - usr[3]) / 27
 233     plot(data$getSerie(tail(indices,1)), type="l", lwd=2, lty=2,
 234         ylim=c(usr[3] + yr, usr[4] - yr), xlab="", ylab="")
 235 }
 236
 237 #' @title Plot relative conditional variability / absolute variability
 238 #'
 239 #' @description Draw the relative conditional variability / absolute variability based on on
 240 #'   filaments obtained by \code{computeFilaments}
 241 #'
 242 #' @param data Object return by \code{getData}
 243 #' @param indices Indices as output by \code{computeFilaments}
 244 #'
 245 #' @export
 246 plotRelativeVariability = function(data, indices, ...)
 247 {
 248     #plot left / right separated by vertical line brown dotted
 249     #median of 3 runs for random length(indices) series
 250     ref_series = t( sapply(indices, function(i) {
 251         c( data$getSerie(i), data$getSerie(i+1) )
 252     }) )
 253     ref_var = apply(ref_series, 2, sd)
 254
 255     # Determine indices of no-NAs days followed by no-NAs tomorrows
 256     first_day = ifelse(length(data$getCenteredSerie(1))<length(ref_series[1,]), 2, 1)
 257     fdays_indices = c()
 258     for (i in first_day:(tail(indices,1)-1))
 259     {
 260         if ( !any(is.na(data$getSerie(i)) | is.na(data$getSerie(i+1))) )
 261             fdays_indices = c(fdays_indices, i)
 262     }
 263
 264     # TODO: 3 == magic number
 265     random_var = matrix(nrow=3, ncol=48)
 266     for (mc in seq_len(nrow(random_var)))
 267     {
 268         random_indices = sample(fdays_indices, length(indices))
 269         random_series = t( sapply(random_indices, function(i) {
 270             c( data$getSerie(i), data$getSerie(i+1) )
 271         }) )
 272         random_var[mc,] = apply(random_series, 2, sd)
 273     }
 274     random_var = apply(random_var, 2, median)
 275
 276     yrange = range(ref_var, random_var)
 277     par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
 278     plot(ref_var, type="l", col=1, lwd=3, ylim=yrange, xlab="Temps (heures)", ylab="Écart-type")
 279     par(new=TRUE)
 280     plot(random_var, type="l", col=2, lwd=3, ylim=yrange, xlab="", ylab="")
 281     abline(v=24, lty=2, col=colors()[56])
 282 }