[talweg.git] / pkg / R / plot.R

#' Plot curves
#'
#' Plot a range of curves in data.
#'
#' @inheritParams computeError
#' @param indices Range of indices (integers or dates)
#'
#' @export
plotCurves <- function(data, indices=seq_len(data$getSize()))
{
	series = data$getSeries(indices)
	yrange = quantile(series, probs=c(0.025,0.975), na.rm=TRUE)
	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	matplot(series, type="l", ylim=yrange, xlab="Time (hours)", ylab="PM10")
}

#' Plot error
#'
#' Draw error graphs, potentially from several runs of \code{computeForecast()}.
#'
#' @param err Error as returned by \code{computeError()}
#' @param cols Colors for each error (default: 1,2,3,...)
#'
#' @seealso \code{\link{plotCurves}}, \code{\link{plotPredReal}},
#'   \code{\link{plotSimils}}, \code{\link{plotFbox}}, \code{\link{computeFilaments}},
#'   \code{\link{plotFilamentsBox}}, \code{\link{plotRelVar}}
#'
#' @export
plotError <- function(err, cols=seq_along(err))
{
	if (!is.null(err$abs))
		err = list(err)
	par(mfrow=c(2,2), mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lty=1)
	L = length(err)

	yrange = range( sapply(1:L, function(i) ( err[[i]]$abs$day ) ), na.rm=TRUE )
	matplot( sapply( seq_len(L), function(i) err[[i]]$abs$day ), type="l",
		xlab="Time (hours)", ylab="Mean |y - y_hat|", ylim=yrange, col=cols, lwd=2 )

	yrange = range( sapply(1:L, function(i) ( err[[i]]$abs$indices ) ), na.rm=TRUE )
	matplot( sapply( seq_len(L), function(i) err[[i]]$abs$indices ), type="l",
		xlab="Time (days)", ylab="Mean |y - y_hat|", ylim=yrange, col=cols, lwd=2 )

	yrange = range( sapply(1:L, function(i) ( err[[i]]$MAPE$day ) ), na.rm=TRUE )
	matplot( sapply( seq_len(L), function(i) err[[i]]$MAPE$day ), type="l",
		xlab="Time (hours)", ylab="Mean MAPE", ylim=yrange, col=cols, lwd=2 )

	yrange = range( sapply(1:L, function(i) ( err[[i]]$MAPE$indices ) ), na.rm=TRUE )
	matplot( sapply( seq_len(L), function(i) err[[i]]$MAPE$indices ), type="l",
		xlab="Time (days)", ylab="Mean MAPE", ylim=yrange, col=cols, lwd=2 )
}

#' Plot measured / predicted
#'
#' Plot measured curve (in black) and predicted curve (in blue).
#'
#' @inheritParams computeError
#' @param index Index in forecasts (integer or date)
#'
#' @export
plotPredReal <- function(data, pred, index)
{
	prediction = pred$getForecast(index)
	measure = data$getSerie( pred$getIndexInData(index) )[1:length(pred$getForecast(1))]

	# Remove the common part, where prediction == measure
	dot_mark <- ifelse(prediction[1]==measure[1], which.max(prediction==measure), 0)
	prediction = prediction[(dot_mark+1):length(prediction)]
	measure = measure[(dot_mark+1):length(measure)]

	yrange = range(measure, prediction)
	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lwd=3)
	plot(measure, type="l", ylim=yrange, xlab="Time (hours)", ylab="PM10")
	par(new=TRUE)
	plot(prediction, type="l", col="#0000FF", ylim=yrange, xlab="", ylab="")
}

#' Plot similarities
#'
#' Plot histogram of similarities (weights), for 'Neighbors' method.
#'
#' @inheritParams computeError
#' @param index Index in forecasts (integer or date)
#'
#' @export
plotSimils <- function(pred, index)
{
	weights = pred$getParams(index)$weights
	if (is.null(weights))
		stop("plotSimils only works on 'Neighbors' forecasts")
	par(mfrow=c(1,2), mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	small_weights = weights[ weights < 1/length(weights) ]
	large_weights = weights[ weights >= 1/length(weights) ]
	hist(small_weights, nclass=25, main="", xlab="Weight < 1/N", ylab="Count")
	hist(large_weights, nclass=25, main="", xlab="Weight >= 1/N", ylab="Count")
}

#' Functional boxplot
#'
#' Draw the functional boxplot on the left, and bivariate plot on the right.
#'
#' @inheritParams computeError
#' @inheritParams plotCurves
#'
#' @export
plotFbox <- function(data, indices=seq_len(data$getSize()))
{
	if (!requireNamespace("rainbow", quietly=TRUE))
		stop("Functional boxplot requires the rainbow package")

	series_matrix = data$getSeries(indices)
	# Remove series with NAs
	no_NAs_indices = sapply( 1:ncol(series_matrix),
		function(i) all(!is.na(series_matrix[,i])) )
	series_matrix = series_matrix[,no_NAs_indices]

	series_fds = rainbow::fds(seq_len(nrow(series_matrix)), series_matrix)
	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	rainbow::fboxplot(series_fds, "functional", "hdr", xlab="Time (hours)", ylab="PM10",
		plotlegend=FALSE, lwd=2)
	rainbow::fboxplot(series_fds, "bivariate", "hdr", plotlegend=FALSE)
}

#' Compute filaments
#'
#' Obtain similar days in the past, and (optionally) plot them -- as black as distances
#' are small.
#'
#' @inheritParams computeError
#' @param index Index in forecast (integer or date)
#' @param limit Number of neighbors to consider
#' @param plot Should the result be plotted?
#'
#' @return A list with
#' \itemize{
#'   \item index : index of the current serie ('today')
#'   \item neighb_indices : indices of its neighbors
#'   \item colors : colors of neighbors curves (shades of gray)
#' }
#'
#' @export
computeFilaments <- function(data, pred, index, limit=60, plot=TRUE)
{
	weights <- pred$getParams(index)$weights
	if (is.null(weights) || is.na(pred$getParams(index)$weights[1]))
		stop("computeFilaments requires a serie without NAs")

	nn <- min(limit, length(weights))
	sorted_dists = sort(-log(weights), index.return=TRUE)
	# Compute colors for each neighbor (from darkest to lightest), if weights differ
	if ( any( weights != weights[1] ) )
	{
		min_dist = min(sorted_dists$x[1:nn])
		max_dist = max(sorted_dists$x[1:nn])
		color_values = floor(19.5*(sorted_dists$x[1:nn]-min_dist)/(max_dist-min_dist)) + 1
		colors = gray.colors(20,0.1,0.9)[color_values] #TODO: 20 == magic number
	}
	else
		colors <- rep(colors()[17], length(weights))

	if (plot)
	{
		# Complete series with (past and present) tomorrows
		ref_serie = c( data$getCenteredSerie( pred$getIndexInData(index)-1 ),
			data$getCenteredSerie( pred$getIndexInData(index) ) )
		centered_series = rbind(
			data$getCenteredSeries( pred$getParams(index)$indices-1 ),
			data$getCenteredSeries( pred$getParams(index)$indices ) )
		yrange = range( ref_serie,
			quantile(centered_series, probs=c(0.025,0.975), na.rm=TRUE) )
		par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lwd=2)
		for (i in nn:1)
		{
			plot(centered_series[,sorted_dists$ix[i]], ylim=yrange, type="l", col=colors[i],
				xlab=ifelse(i==1,"Time (hours)",""), ylab=ifelse(i==1,"Centered PM10",""))
			par(new=TRUE)
		}
		# Also plot ref curve, in red
		plot(ref_serie, ylim=yrange, type="l", col="#FF0000", xlab="", ylab="")
		dot_mark <- 0.5 + which.max( pred$getForecast(1) ==
			data$getSerie( pred$getIndexInData(1) )[1:length(pred$getForecast(1))] )
		abline(v=24+dot_mark, lty=2, col=colors()[56], lwd=1)
	}

	list(
		"index"=pred$getIndexInData(index),
		"neighb_indices"=pred$getParams(index)$indices[sorted_dists$ix[1:nn]],
		"colors"=colors)
}

#' Functional boxplot on filaments
#'
#' Draw the functional boxplot on filaments obtained by \code{computeFilaments()}.
#'
#' @inheritParams computeError
#' @param fil Output of \code{computeFilaments}
#' @param predict_from First predicted time step
#'
#' @export
plotFilamentsBox = function(data, fil, predict_from)
{
	if (!requireNamespace("rainbow", quietly=TRUE))
		stop("Functional boxplot requires the rainbow package")

	series_matrix = rbind(
		data$getSeries(fil$neighb_indices-1), data$getSeries(fil$neighb_indices) )
	series_fds = rainbow::fds(seq_len(nrow(series_matrix)), series_matrix)

	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	rainbow::fboxplot(series_fds, "functional", "hdr", xlab="Time (hours)", ylab="PM10",
		plotlegend=FALSE, lwd=2)

	# "Magic": http://stackoverflow.com/questions/13842560/get-xlim-from-a-plot-in-r
	usr <- par("usr")
	yr <- (usr[4] - usr[3]) / 27
	par(new=TRUE)
	plot(c(data$getSerie(fil$index-1),data$getSerie(fil$index)), type="l", lwd=2, lty=2,
		ylim=c(usr[3] + yr, usr[4] - yr), xlab="", ylab="")
	abline(v=24+predict_from-0.5, lty=2, col=colors()[56])
}

#' Plot relative conditional variability / absolute variability
#'
#' Draw the relative conditional variability / absolute variability based on filaments
#' obtained by \code{computeFilaments()}.
#'
#' @inheritParams computeError
#' @inheritParams plotFilamentsBox
#'
#' @export
plotRelVar = function(data, fil, predict_from)
{
	ref_var = c( apply(data$getSeries(fil$neighb_indices-1),1,sd),
		apply(data$getSeries(fil$neighb_indices),1,sd) )
	tdays = .getNoNA2(data, 2, fil$index)
	global_var = c(
		apply(data$getSeries(tdays-1),1,sd),
		apply(data$getSeries(tdays),1,sd) )

	yrange = range(ref_var, global_var)
	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	plot(ref_var, type="l", col=1, lwd=3, ylim=yrange,
		xlab="Time (hours)", ylab="Standard deviation")
	par(new=TRUE)
	plot(global_var, type="l", col=2, lwd=3, ylim=yrange, xlab="", ylab="")
	abline(v=24+predict_from-0.5, lty=2, col=colors()[56])
}
Commit	Line	Data
	1	#' Plot curves
	2	#'
	3	#' Plot a range of curves in data.
	4	#'
	5	#' @inheritParams computeError
	6	#' @param indices Range of indices (integers or dates)
	7	#'
	8	#' @export
	9	plotCurves <- function(data, indices=seq_len(data$getSize()))
	10	{
	11	series = data$getSeries(indices)
	12	yrange = quantile(series, probs=c(0.025,0.975), na.rm=TRUE)
	13	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	14	matplot(series, type="l", ylim=yrange, xlab="Time (hours)", ylab="PM10")
	15	}
	16
	17	#' Plot error
	18	#'
	19	#' Draw error graphs, potentially from several runs of \code{computeForecast()}.
	20	#'
	21	#' @param err Error as returned by \code{computeError()}
	22	#' @param cols Colors for each error (default: 1,2,3,...)
	23	#'
	24	#' @seealso \code{\link{plotCurves}}, \code{\link{plotPredReal}},
	25	#' \code{\link{plotSimils}}, \code{\link{plotFbox}}, \code{\link{computeFilaments}},
	26	#' \code{\link{plotFilamentsBox}}, \code{\link{plotRelVar}}
	27	#'
	28	#' @export
	29	plotError <- function(err, cols=seq_along(err))
	30	{
	31	if (!is.null(err$abs))
	32	err = list(err)
	33	par(mfrow=c(2,2), mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lty=1)
	34	L = length(err)
	35
	36	yrange = range( sapply(1:L, function(i) ( err[[i]]$abs$day ) ), na.rm=TRUE )
	37	matplot( sapply( seq_len(L), function(i) err[[i]]$abs$day ), type="l",
	38	xlab="Time (hours)", ylab="Mean \|y - y_hat\|", ylim=yrange, col=cols, lwd=2 )
	39
	40	yrange = range( sapply(1:L, function(i) ( err[[i]]$abs$indices ) ), na.rm=TRUE )
	41	matplot( sapply( seq_len(L), function(i) err[[i]]$abs$indices ), type="l",
	42	xlab="Time (days)", ylab="Mean \|y - y_hat\|", ylim=yrange, col=cols, lwd=2 )
	43
	44	yrange = range( sapply(1:L, function(i) ( err[[i]]$MAPE$day ) ), na.rm=TRUE )
	45	matplot( sapply( seq_len(L), function(i) err[[i]]$MAPE$day ), type="l",
	46	xlab="Time (hours)", ylab="Mean MAPE", ylim=yrange, col=cols, lwd=2 )
	47
	48	yrange = range( sapply(1:L, function(i) ( err[[i]]$MAPE$indices ) ), na.rm=TRUE )
	49	matplot( sapply( seq_len(L), function(i) err[[i]]$MAPE$indices ), type="l",
	50	xlab="Time (days)", ylab="Mean MAPE", ylim=yrange, col=cols, lwd=2 )
	51	}
	52
	53	#' Plot measured / predicted
	54	#'
	55	#' Plot measured curve (in black) and predicted curve (in blue).
	56	#'
	57	#' @inheritParams computeError
	58	#' @param index Index in forecasts (integer or date)
	59	#'
	60	#' @export
	61	plotPredReal <- function(data, pred, index)
	62	{
	63	prediction = pred$getForecast(index)
	64	measure = data$getSerie( pred$getIndexInData(index) )[1:length(pred$getForecast(1))]
	65
	66	# Remove the common part, where prediction == measure
	67	dot_mark <- ifelse(prediction[1]==measure[1], which.max(prediction==measure), 0)
	68	prediction = prediction[(dot_mark+1):length(prediction)]
	69	measure = measure[(dot_mark+1):length(measure)]
	70
	71	yrange = range(measure, prediction)
	72	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lwd=3)
	73	plot(measure, type="l", ylim=yrange, xlab="Time (hours)", ylab="PM10")
	74	par(new=TRUE)
	75	plot(prediction, type="l", col="#0000FF", ylim=yrange, xlab="", ylab="")
	76	}
	77
	78	#' Plot similarities
	79	#'
	80	#' Plot histogram of similarities (weights), for 'Neighbors' method.
	81	#'
	82	#' @inheritParams computeError
	83	#' @param index Index in forecasts (integer or date)
	84	#'
	85	#' @export
	86	plotSimils <- function(pred, index)
	87	{
	88	weights = pred$getParams(index)$weights
	89	if (is.null(weights))
	90	stop("plotSimils only works on 'Neighbors' forecasts")
	91	par(mfrow=c(1,2), mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	92	small_weights = weights[ weights < 1/length(weights) ]
	93	large_weights = weights[ weights >= 1/length(weights) ]
	94	hist(small_weights, nclass=25, main="", xlab="Weight < 1/N", ylab="Count")
	95	hist(large_weights, nclass=25, main="", xlab="Weight >= 1/N", ylab="Count")
	96	}
	97
	98	#' Functional boxplot
	99	#'
	100	#' Draw the functional boxplot on the left, and bivariate plot on the right.
	101	#'
	102	#' @inheritParams computeError
	103	#' @inheritParams plotCurves
	104	#'
	105	#' @export
	106	plotFbox <- function(data, indices=seq_len(data$getSize()))
	107	{
	108	if (!requireNamespace("rainbow", quietly=TRUE))
	109	stop("Functional boxplot requires the rainbow package")
	110
	111	series_matrix = data$getSeries(indices)
	112	# Remove series with NAs
	113	no_NAs_indices = sapply( 1:ncol(series_matrix),
	114	function(i) all(!is.na(series_matrix[,i])) )
	115	series_matrix = series_matrix[,no_NAs_indices]
	116
	117	series_fds = rainbow::fds(seq_len(nrow(series_matrix)), series_matrix)
	118	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	119	rainbow::fboxplot(series_fds, "functional", "hdr", xlab="Time (hours)", ylab="PM10",
	120	plotlegend=FALSE, lwd=2)
	121	rainbow::fboxplot(series_fds, "bivariate", "hdr", plotlegend=FALSE)
	122	}
	123
	124	#' Compute filaments
	125	#'
	126	#' Obtain similar days in the past, and (optionally) plot them -- as black as distances
	127	#' are small.
	128	#'
	129	#' @inheritParams computeError
	130	#' @param index Index in forecast (integer or date)
	131	#' @param limit Number of neighbors to consider
	132	#' @param plot Should the result be plotted?
	133	#'
	134	#' @return A list with
	135	#' \itemize{
	136	#' \item index : index of the current serie ('today')
	137	#' \item neighb_indices : indices of its neighbors
	138	#' \item colors : colors of neighbors curves (shades of gray)
	139	#' }
	140	#'
	141	#' @export
	142	computeFilaments <- function(data, pred, index, limit=60, plot=TRUE)
	143	{
	144	weights <- pred$getParams(index)$weights
	145	if (is.null(weights) \|\| is.na(pred$getParams(index)$weights[1]))
	146	stop("computeFilaments requires a serie without NAs")
	147
	148	nn <- min(limit, length(weights))
	149	sorted_dists = sort(-log(weights), index.return=TRUE)
	150	# Compute colors for each neighbor (from darkest to lightest), if weights differ
	151	if ( any( weights != weights[1] ) )
	152	{
	153	min_dist = min(sorted_dists$x[1:nn])
	154	max_dist = max(sorted_dists$x[1:nn])
	155	color_values = floor(19.5*(sorted_dists$x[1:nn]-min_dist)/(max_dist-min_dist)) + 1
	156	colors = gray.colors(20,0.1,0.9)[color_values] #TODO: 20 == magic number
	157	}
	158	else
	159	colors <- rep(colors()[17], length(weights))
	160
	161	if (plot)
	162	{
	163	# Complete series with (past and present) tomorrows
	164	ref_serie = c( data$getCenteredSerie( pred$getIndexInData(index)-1 ),
	165	data$getCenteredSerie( pred$getIndexInData(index) ) )
	166	centered_series = rbind(
	167	data$getCenteredSeries( pred$getParams(index)$indices-1 ),
	168	data$getCenteredSeries( pred$getParams(index)$indices ) )
	169	yrange = range( ref_serie,
	170	quantile(centered_series, probs=c(0.025,0.975), na.rm=TRUE) )
	171	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5, lwd=2)
	172	for (i in nn:1)
	173	{
	174	plot(centered_series[,sorted_dists$ix[i]], ylim=yrange, type="l", col=colors[i],
	175	xlab=ifelse(i==1,"Time (hours)",""), ylab=ifelse(i==1,"Centered PM10",""))
	176	par(new=TRUE)
	177	}
	178	# Also plot ref curve, in red
	179	plot(ref_serie, ylim=yrange, type="l", col="#FF0000", xlab="", ylab="")
	180	dot_mark <- 0.5 + which.max( pred$getForecast(1) ==
	181	data$getSerie( pred$getIndexInData(1) )[1:length(pred$getForecast(1))] )
	182	abline(v=24+dot_mark, lty=2, col=colors()[56], lwd=1)
	183	}
	184
	185	list(
	186	"index"=pred$getIndexInData(index),
	187	"neighb_indices"=pred$getParams(index)$indices[sorted_dists$ix[1:nn]],
	188	"colors"=colors)
	189	}
	190
	191	#' Functional boxplot on filaments
	192	#'
	193	#' Draw the functional boxplot on filaments obtained by \code{computeFilaments()}.
	194	#'
	195	#' @inheritParams computeError
	196	#' @param fil Output of \code{computeFilaments}
	197	#' @param predict_from First predicted time step
	198	#'
	199	#' @export
	200	plotFilamentsBox = function(data, fil, predict_from)
	201	{
	202	if (!requireNamespace("rainbow", quietly=TRUE))
	203	stop("Functional boxplot requires the rainbow package")
	204
	205	series_matrix = rbind(
	206	data$getSeries(fil$neighb_indices-1), data$getSeries(fil$neighb_indices) )
	207	series_fds = rainbow::fds(seq_len(nrow(series_matrix)), series_matrix)
	208
	209	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	210	rainbow::fboxplot(series_fds, "functional", "hdr", xlab="Time (hours)", ylab="PM10",
	211	plotlegend=FALSE, lwd=2)
	212
	213	# "Magic": http://stackoverflow.com/questions/13842560/get-xlim-from-a-plot-in-r
	214	usr <- par("usr")
	215	yr <- (usr[4] - usr[3]) / 27
	216	par(new=TRUE)
	217	plot(c(data$getSerie(fil$index-1),data$getSerie(fil$index)), type="l", lwd=2, lty=2,
	218	ylim=c(usr[3] + yr, usr[4] - yr), xlab="", ylab="")
	219	abline(v=24+predict_from-0.5, lty=2, col=colors()[56])
	220	}
	221
	222	#' Plot relative conditional variability / absolute variability
	223	#'
	224	#' Draw the relative conditional variability / absolute variability based on filaments
	225	#' obtained by \code{computeFilaments()}.
	226	#'
	227	#' @inheritParams computeError
	228	#' @inheritParams plotFilamentsBox
	229	#'
	230	#' @export
	231	plotRelVar = function(data, fil, predict_from)
	232	{
	233	ref_var = c( apply(data$getSeries(fil$neighb_indices-1),1,sd),
	234	apply(data$getSeries(fil$neighb_indices),1,sd) )
	235	tdays = .getNoNA2(data, 2, fil$index)
	236	global_var = c(
	237	apply(data$getSeries(tdays-1),1,sd),
	238	apply(data$getSeries(tdays),1,sd) )
	239
	240	yrange = range(ref_var, global_var)
	241	par(mar=c(4.7,5,1,1), cex.axis=1.5, cex.lab=1.5)
	242	plot(ref_var, type="l", col=1, lwd=3, ylim=yrange,
	243	xlab="Time (hours)", ylab="Standard deviation")
	244	par(new=TRUE)
	245	plot(global_var, type="l", col=2, lwd=3, ylim=yrange, xlab="", ylab="")
	246	abline(v=24+predict_from-0.5, lty=2, col=colors()[56])
	247	}