Calculate euclidean distance with R - r

I have data where rows are points and columns are coordinates x,y,z.
I'd like to calculate euclidean distance between points in couple, as 3-4, 11-12, 18-19 and so on... for example, I dont' need distance between 3 and 11, 12, 18
The problem is that I have to analize 1074 tables with 1000 rows or more, so I'm searching a way to do it automatically, maybe considering tha fact that I want to calculate distance between an odd number and the even following one. I don't care too much about the output format, but pls consider that after I have to select only distances <3.2, so a dataframe format will be great.
THANK YOU! :*

How about something like this:
First, I'll make some fake data
set.seed(4304)
df <- data.frame(
x = runif(1000, -1, 1),
y = runif(1000, -1, 1),
z = runif(1000, -1,1)
)
Make a sequence of values from 1 to the number of rows of your dataset by 2s.
s <- seq(1, nrow(df), by=2)
Use sapply() to make the distance between each pair of points.
out <- sapply(s, function(i){
sqrt(sum((df[i,] - df[(i+1), ])^2))
})
Organize the distances into a data frame
res <- data.frame(
pair = paste(rownames(df)[s], rownames(df)[(s+1)], sep="-"),
dist=out)
head(res)
# pair dist
# 1 1-2 1.379992
# 2 3-4 1.303511
# 3 5-6 1.242302
# 4 7-8 1.257228
# 5 9-10 1.107484
# 6 11-12 1.392247

Here is a function that can be applied to a data.frame or matrix holding the data.
DistEucl <- function(X){
i <- cumsum(seq_len(nrow(X)) %% 2 == 1)
sapply(split(X, i), function(Y){
sqrt(sum((Y[1, ] - Y[2, ])^2))
})
}
DistEucl(df1)
# 1 2 3 4
#1.229293 1.234273 1.245567 1.195319
With the data in DaveArmstrong's answer, the results are the same except for a names attribute in the above function's return value.
out2 <- DistEucl(df)
all.equal(out, out2)
#[1] "names for current but not for target"
identical(out, unname(out2))
#[1] TRUE
Data in the question
x <- c(13.457, 13.723, 15.319, 15.713, 18.446, 19.488, 19.762, 19.743)
y <- c(28.513, 29.656, 28.510, 27.342, 28.827, 28.24, 29.841, 30.942)
z <- c(40.513, 40.147, 43.281, 43.218, 43.095, 43.443, 40.094, 40.559)
df1 <- data.frame(x, y, z)

Related

Converting Lists into Data Frames

I am trying to make a data frame with 3 columns (and 100 rows) containing random numbers such that the random numbers in each row add to 72.
I am using this code to generate the random numbers:
random_numbers <- diff(c(0, sort(sample(72, 2)), 72))
Although, I can't "fit" these random numbers into a data frame because of the format. For example:
i <- 1:100
d <- data.frame(i)
d$rand <- diff(c(0, sort(sample(72, 2)), 72))
Error in `$<-.data.frame`(`*tmp*`, rand, value = c(37, 21, 14)) :
replacement has 3 rows, data has 100
I had another idea in which at least I can create all 100 random numbers:
results <- list()
for (i in 1:100) {
r_i <- diff(c(0, sort(sample(72, 2)), 72))
results[[i]] <- r_i
}
results[1]
# [[1]]
# [1] 3 19 50
results[2]
# [[1]]
# [1] 16 11 45
But I am not sure how I can I can create a data frame with 3 columns and 100 rows from this data.
I know how to do this in the "classical" sense:
i <- 1:100
r_1 <- rnorm(5, 5, 100)
r_2 <- rnorm(5, 5, 100)
r_3 <- rnorm(5, 5, 100)
d <- data.frame(i, r_1, r_2, r_3)
d = data.frame(i, r_1, r_2, r_3)
But of course, in the above data frame, these 3 numbers will most certainly not add to 72.
Is it possible to take the 100 random numbers results that I generated above and then place them into a data frame?
We may use replicate with n specified as the number of rows of 'd' and assign new columns from the matrix output
d[paste0("r_", 1:3)] <- t(replicate(nrow(d),
diff(c(0, sort(sample(72, 2)), 72))))
-testing for equality
> all(rowSums(d[-1]) == 72)
[1] TRUE

Trying to make a script calculate a value (using a function) for every 24 rows

I have not been able to find a solution to a problem similar to this on StackOverflow. I hope someone can help!
I am using the R environment.
I have data from turtle nests. There are two types of hourly data in each nest. The first is hourly Temperature, and it has an associated hourly Development (amount of "anatomical" embryonic development").
I am calculating a weighted median. In this case, the median is temperature and it is weighted by development.
I have a script here that I am using to calculated weighted median:
weighted.median <- function(x, w, probs=0.5, na.rm=TRUE) {
x <- as.numeric(as.vector(x))
w <- as.numeric(as.vector(w))
if(anyNA(x) || anyNA(w)) {
ok <- !(is.na(x) | is.na(w))
x <- x[ok]
w <- w[ok]
}
stopifnot(all(w >= 0))
if(all(w == 0)) stop("All weights are zero", call.=FALSE)
#'
oo <- order(x)
x <- x[oo]
w <- w[oo]
Fx <- cumsum(w)/sum(w)
#'
result <- numeric(length(probs))
for(i in seq_along(result)) {
p <- probs[i]
lefties <- which(Fx <= p)
if(length(lefties) == 0) {
result[i] <- x[1]
} else {
left <- max(lefties)
result[i] <- x[left]
if(Fx[left] < p && left < length(x)) {
right <- left+1
y <- x[left] + (x[right]-x[left]) * (p-Fx[left])/(Fx[right]- Fx[left])
if(is.finite(y)) result[i] <- y
}
}
}
names(result) <- paste0(format(100 * probs, trim = TRUE), "%")
return(result)
}
So from the function you can see that I need two input vectors, x and w (which will be temperature and development, respectively).
The problem I'm having is that I have hourly temperature traces that last anywhere from 5 days to 53 days (i.e., 120 hours to 1272 hours).
I would like to calculate the daily weighted median for all days within a nest (i.e., take the 24 rows of x and w, and calculate the weighted median, then move onto rows 25-48, and so forth.) The output vector would therefore be a list of daily weighted medians with length n/24 (where n is the total number of rows in x).
In other words, I would like to analyse my data automatically, in a fashion equivalent to manually doing this (nest1 is the datasheet for Nest 1 which contains two vectors, temp and devo (devo is the weight))):
`weighted.median(nest1$temp[c(1,1:24)],nest1$devo[c(1,1:24)],na.rm=TRUE)`
followed by
weighted.median(nest1$temp[c(1,25:48)],nest1$devo[c(1,25:48)],na.rm=TRUE)
followed by
weighted.median(nest1$temp[c(1,49:72)],nest1$devo[c(1,49:72)],na.rm=TRUE)
all the way to
`weighted.median(nest1$temp[c(1,n-23:n)],nest1$devo[c(1,n-23:n)],na.rm=TRUE)`
I'm afraid I don't even know where to start. Any help or clues would be very much appreciated.
The main idea is to create a new column for day 1, day 2, ..., day n/24, split the dataframe into subsets by day, and apply your function to each subset.
First I create some sample data:
set.seed(123)
n <- 120 # number of rows
nest1 <- data.frame(temp = rnorm(n), devo = rpois(n, 5))
Create the splitting variable:
nest1$day <- rep(1:(nrow(nest1)/24), each = 24)
Then, use the by() function to split nest1 by nest1$day and apply the function to each subset:
out <- by(nest1, nest1$day, function(d) {
weighted.median(d$temp, d$devo, na.rm = TRUE)
})
data.frame(day = dimnames(out)[[1]], x = as.vector(out))
# day x
# 1 1 -0.45244433
# 2 2 0.15337312
# 3 3 0.07071673
# 4 4 0.23873174
# 5 5 -0.27694709
Instead of using by, you can also use the group_by + summarise functions from the dplyr package:
library(dplyr)
nest1 %>%
group_by(day) %>%
summarise(x = weighted.median(temp, devo, na.rm = TRUE))
# # A tibble: 5 x 2
# day x
# <int> <dbl>
# 1 1 -0.452
# 2 2 0.153
# 3 3 0.0707
# 4 4 0.239
# 5 5 -0.277

R: find consecutive occurrence of a number

first define some function to bind list rowwise and column wise
# a function to append vectors row wise
rbindlist <- function(list) {
n <- length(list)
res <- NULL
for (i in seq(n)) res <- rbind(res, list[[i]])
return(res)
}
cbindlist <- function(list) {
n <- length(list)
res <- NULL
for (i in seq(n)) res <- cbind(res, list[[i]])
return(res)
}
# generate sample data
sample.dat <- list()
set.seed(123)
for(i in 1:365){
vec1 <- sample(c(0,1), replace=TRUE, size=5)
sample.dat[[i]] <- vec1
}
dat <- rbindlist(sample.dat)
dat has five columns. Each column is a location and has 365 days of the year (365 rows) with values 1 or 0.
I have another dataframe (see below) which has certain days of the year for each column (location) in dat.
# generate second sample data
set.seed(123)
sample.dat1 <- list()
for(i in 1:5){
vec1 <- sort(sample(c(258:365), replace=TRUE, size=4), decreasing = F)
sample.dat1[[i]] <- vec1
}
dat1 <- cbindlist(sample.dat1)
I need to use dat1 to subset days in dat to do a calculation. An example below:
1) For location 1 (first column in both dat1 and dat):
In column 1 of dat, select the days from 289 till 302 (using dat1), find the longest consecutive occurrence of 1.
Repeat it and this time select the days from 303 (302 + 1) till 343 from dat, find the longest consecutive occurrence of 1.
Repeat it for 343 till 353: select the days from 344 (343 + 1) till 353, find the longest consecutive occurrence of 1.
2) Do this for all the columns
If I want to do sum of 1s, I can do this:
dat <- as.tibble(dat)
dat1 <- as.tibble(dat1)
pmap(list(dat,dat1), ~ {
range1 <- ..2[1]
range2 <- ..2[2]
range3 <- ..2[3]
range4 <- ..2[4]
sum.range1 <- sum(..1[range1:range2]) # this will generate sum between range 1 and range 2
sum.range2 <- sum(..1[range2:range3]) # this will generate sum between range 2 and range 3
sum.range3 <- sum(..1[range3:range4]) # this will generate sum between range 3 and range 4
c(sum.range1=sum.range1,sum.range2=sum.range2,sum.range3=sum.range3)
})
For longest consequtive occurrence of 1 between each range, I thought of using the rle function. Example below:
pmap(list(dat,dat1), ~ {
range1 <- ..2[1]
range2 <- ..2[2]
range3 <- ..2[3]
range4 <- ..2[4]
spell.range1 <- rle(..1[range1:range2]) # sort the data, this shows the longest run of ANY type (0 OR 1)
spell.1.range1 <- tapply(spell.range1$lengths, spell.range1$values, max)[2] # this should select the maximum consequtive run of 1
spell.range2 <- rle(..1[range2:range3]) # sort the data, this shows the longest run of ANY type (0 OR 1)
spell.1.range2 <- tapply(spell.range2$lengths, spell.range2$values, max)[2] # this should select the maximum consequtive run of 1
spell.range3 <- rle(..1[range3:range4]) # sort the data, this shows the longest run of ANY type (0 OR 1)
spell.1.range3 <- tapply(spell.range3$lengths, spell.range3$values, max)[2] # this should select the maximum consequtive run of 1
c(spell.1.range1 = spell.1.range1, spell.1.range2 = spell.1.range2, spell.1.range3 = spell.1.range3)
})
I get an error which I think is because I am not using the rle function properly here. I would really like to keep the code as above since
my others code are in the same pattern and format of the outputs is suited for my need, so I would appreciate if someone can suggest how to fix it.
OP's code does work for me. So, without a specific error message it is impossible to understand why the code is not working for the OP.
However, the sample datasets created by the OP are matrices (before they were coerced to tibble) and I felt challenged to find a way to solve the task in base R without using purrr:
To find the number of consecutive occurences of a particular value val in a vector x we can use the following function:
max_rle <- function(x, val) {
y <- rle(x)
len <- y$lengths[y$value == val]
if (length(len) > 0) max(len) else NA
}
Examples:
max_rle(c(0, 1, 1, 1, 1, 0, 0, 1, 0, 1, 1), 1)
[1] 4
max_rle(c(0, 1, 1, 1, 1, 0, 0, 1, 0, 1, 1), 0)
[1] 2
# find consecutive occurrences in column batches
lapply(seq_len(ncol(dat1)), function(col_num) {
start <- head(dat1[, col_num], -1L)
end <- tail(dat1[, col_num], -1L) - 1
sapply(seq_along(start), function(range_num) {
max_rle(dat[start[range_num]:end[range_num], col_num], 1)
})
})
[[1]]
[1] 8 4 5
[[2]]
[1] 4 5 2
[[3]]
[1] NA 3 4
[[4]]
[1] 5 5 4
[[5]]
[1] 3 2 3
The first lapply() loops over the columns of dat and dat1, resp. The second sapply() loops over the row ranges stored in dat1 and subsets dat accordingly.

Computing pairwise Hamming distance between all rows of two integer matrices/data frames

I have two data frames, df1 with reference data and df2 with new data. For each row in df2, I need to find the best (and the second best) matching row to df1 in terms of hamming distance.
I used e1071 package to compute hamming distance. Hamming distance between two vectors x and y can be computed as for example:
x <- c(356739, 324074, 904133, 1025460, 433677, 110525, 576942, 526518, 299386,
92497, 977385, 27563, 429551, 307757, 267970, 181157, 3796, 679012, 711274,
24197, 610187, 402471, 157122, 866381, 582868, 878)
y <- c(356739, 324042, 904133, 959893, 433677, 110269, 576942, 2230, 267130,
92496, 960747, 28587, 429551, 438825, 267970, 181157, 36564, 677220,
711274, 24485, 610187, 404519, 157122, 866413, 718036, 876)
xm <- sapply(x, intToBits)
ym <- sapply(y, intToBits)
distance <- sum(sapply(1:ncol(xm), function(i) hamming.distance(xm[,i], ym[,i])))
and the resulting distance is 25. Yet I need to do this for all rows of df1 and df2. A trivial method takes a double loop nest and looks terribly slow.
Any ideas how to do this more efficiently? In the end I need to append to df2:
a column with the row id from df1 that gives the lowest distance;
a column with the lowest distance;
a column with the row id from df1 that gives the 2nd lowest distance;
a column with the second lowest distance.
Thanks.
Fast computation of hamming distance between two integers vectors of equal length
As I said in my comment, we can do:
hmd0 <- function(x,y) sum(as.logical(xor(intToBits(x),intToBits(y))))
to compute hamming distance between two integers vectors of equal length x and y. This only uses R base, yet is more efficient than e1071::hamming.distance, because it is vectorized!
For the example x and y in your post, this gives 25. (My other answer will show what we should do, if we want pairwise hamming distance.)
Fast hamming distance between a matrix and a vector
If we want to compute the hamming distance between a single y and multiple xs, i.e., the hamming distance between a vector and a matrix, we can use the following function.
hmd <- function(x,y) {
rawx <- intToBits(x)
rawy <- intToBits(y)
nx <- length(rawx)
ny <- length(rawy)
if (nx == ny) {
## quick return
return (sum(as.logical(xor(rawx,rawy))))
} else if (nx < ny) {
## pivoting
tmp <- rawx; rawx <- rawy; rawy <- tmp
tmp <- nx; nx <- ny; ny <- tmp
}
if (nx %% ny) stop("unconformable length!") else {
nc <- nx / ny ## number of cycles
return(unname(tapply(as.logical(xor(rawx,rawy)), rep(1:nc, each=ny), sum)))
}
}
Note that:
hmd performs computation column-wise. It is designed to be CPU cache friendly. In this way, if we want to do some row-wise computation, we should transpose the matrix first;
there is no obvious loop here; instead, we use tapply().
Fast hamming distance computation between two matrices/data frames
This is what you want. The following function foo takes two data frames or matrices df1 and df2, computing the distance between df1 and each row of df2. argument p is an integer, showing how many results you want to retain. p = 3 will keep the smallest 3 distances with their row ids in df1.
foo <- function(df1, df2, p) {
## check p
if (p > nrow(df2)) p <- nrow(df2)
## transpose for CPU cache friendly code
xt <- t(as.matrix(df1))
yt <- t(as.matrix(df2))
## after transpose, we compute hamming distance column by column
## a for loop is decent; no performance gain from apply family
n <- ncol(yt)
id <- integer(n * p)
d <- numeric(n * p)
k <- 1:p
for (i in 1:n) {
distance <- hmd(xt, yt[,i])
minp <- order(distance)[1:p]
id[k] <- minp
d[k] <- distance[minp]
k <- k + p
}
## recode "id" and "d" into data frame and return
id <- as.data.frame(matrix(id, ncol = p, byrow = TRUE))
colnames(id) <- paste0("min.", 1:p)
d <- as.data.frame(matrix(d, ncol = p, byrow = TRUE))
colnames(d) <- paste0("mindist.", 1:p)
list(id = id, d = d)
}
Note that:
transposition is done at the beginning, according to reasons before;
a for loop is used here. But this is actually efficient because there is considerable computation done in each iteration. It is also more elegant than using *apply family, since we ask for multiple output (row id id and distance d).
Experiment
This part uses small dataset to test/demonstrate our functions.
Some toy data:
set.seed(0)
df1 <- as.data.frame(matrix(sample(1:10), ncol = 2)) ## 5 rows 2 cols
df2 <- as.data.frame(matrix(sample(1:6), ncol = 2)) ## 3 rows 2 cols
Test hmd first (needs transposition):
hmd(t(as.matrix(df1)), df2[1, ]) ## df1 & first row of df2
# [1] 2 4 6 2 4
Test foo:
foo(df1, df2, p = 2)
# $id
# min1 min2
# 1 1 4
# 2 2 3
# 3 5 2
# $d
# mindist.1 mindist.2
# 1 2 2
# 2 1 3
# 3 1 3
If you want to append some columns to df2, you know what to do, right?
Please don't be surprised why I take another section. This part gives something relevant. It is not what OP asks for, but may help any readers.
General hamming distance computation
In the previous answer, I start from a function hmd0 that computes hamming distance between two integer vectors of the same length. This means if we have 2 integer vectors:
set.seed(0)
x <- sample(1:100, 6)
y <- sample(1:100, 6)
we will end up with a scalar:
hmd0(x,y)
# 13
What if we want to compute pairwise hamming distance of two vectors?
In fact, a simple modification to our function hmd will do:
hamming.distance <- function(x, y, pairwise = TRUE) {
nx <- length(x)
ny <- length(y)
rawx <- intToBits(x)
rawy <- intToBits(y)
if (nx == 1 && ny == 1) return(sum(as.logical(xor(intToBits(x),intToBits(y)))))
if (nx < ny) {
## pivoting
tmp <- rawx; rawx <- rawy; rawy <- tmp
tmp <- nx; nx <- ny; ny <- tmp
}
if (nx %% ny) stop("unconformable length!") else {
bits <- length(intToBits(0)) ## 32-bit or 64 bit?
result <- unname(tapply(as.logical(xor(rawx,rawy)), rep(1:ny, each = bits), sum))
}
if (pairwise) result else sum(result)
}
Now
hamming.distance(x, y, pairwise = TRUE)
# [1] 0 3 3 2 5 0
hamming.distance(x, y, pairwise = FALSE)
# [1] 13
Hamming distance matrix
If we want to compute the hamming distance matrix, for example,
set.seed(1)
x <- sample(1:100, 5)
y <- sample(1:100, 7)
The distance matrix between x and y is:
outer(x, y, hamming.distance) ## pairwise argument has no effect here
# [,1] [,2] [,3] [,4] [,5] [,6] [,7]
# [1,] 2 3 4 3 4 4 2
# [2,] 7 6 3 4 3 3 3
# [3,] 4 5 4 3 6 4 2
# [4,] 2 3 2 5 6 4 2
# [5,] 4 3 4 3 2 0 2
We can also do:
outer(x, x, hamming.distance)
# [,1] [,2] [,3] [,4] [,5]
# [1,] 0 5 2 2 4
# [2,] 5 0 3 5 3
# [3,] 2 3 0 2 4
# [4,] 2 5 2 0 4
# [5,] 4 3 4 4 0
In the latter situation, we end up with a symmetric matrix with 0 on the diagonal. Using outer is inefficient here, but it is still more efficient than writing R loops. Since our hamming.distance is written in R code, I would stay with using outer. In my answer to this question, I demonstrate the idea of using compiled code. This of course requires writing a C version of hamming.distance, but I will not show it here.
Here's an alternative solution that uses only base R, and should be very fast, especially when your df1 and df2 have many rows. The main reason for this is that it does not use any R-level looping for calculating the Hamming distances, such as for-loops, while-loops, or *apply functions. Instead, it uses matrix multiplication for computing the Hamming distance. In R, this is much faster than any approach using R-level looping. Also note that using an *apply function will not necessarily make your code any faster than using a for loop. Two other efficiency-related features of this approach are: (1) It uses partial sorting for finding the best two matches for each row in df2, and (2) It stores the entire bitwise representation of df1 in one matrix (same for df2), and does so in one single step, without using any R-level loops.
The function that does all the work:
# INPUT:
# X corresponds to your entire df1, but is a matrix
# Y corresponds to your entire df2, but is a matrix
# OUTPUT:
# Matrix with four columns corresponding to the values
# that you specified in your question
fun <- function(X, Y) {
# Convert integers to bits
X <- intToBits(t(X))
# Reshape into matrix
dim(X) <- c(ncols * 32, nrows)
# Convert integers to bits
Y <- intToBits(t(Y))
# Reshape into matrix
dim(Y) <- c(ncols * 32, nrows)
# Calculate pairwise hamming distances using matrix
# multiplication.
# Columns of H index into Y; rows index into X.
# The code for the hamming() function was retrieved
# from this page:
# https://johanndejong.wordpress.com/2015/10/02/faster-hamming-distance-in-r-2/
H <- hamming(X, Y)
# Now, for each row in Y, find the two best matches
# in X. In other words: for each column in H, find
# the two smallest values and their row indices.
t(apply(H, 2, function(h) {
mindists <- sort(h, partial = 1:2)
c(
ind1 = which(h == mindists[1])[1],
val1 = mindists[1],
hmd2 = which(h == mindists[2])[1],
val2 = mindists[2]
)
}))
}
To call the function on some random data:
# Generate some random test data with no. of columns
# corresponding to your data
nrows <- 1000
ncols <- 26
# X corresponds to your df1
X <- matrix(
sample(1e6, nrows * ncols, replace = TRUE),
nrow = nrows,
ncol = ncols
)
# Y corresponds to your df2
Y <- matrix(
sample(1e6, nrows * ncols, replace = TRUE),
nrow = nrows,
ncol = ncols
)
res <- fun(X, Y)
The above example with 1000 rows in both X (df1) and Y (df2) took about 1.1 - 1.2 seconds to run on my laptop.

R: Rolling window function with adjustable window and step-size for irregularly spaced observations

Say there is a 2-column data frame with a time or distance column which sequentially increases and an observation column which may have NAs here and there. How can I efficiently use a sliding window function to get some statistic, say a mean, for the observations in a window of duration X (e.g. 5 seconds), slide the window over Y seconds (e.g. 2.5 seconds), repeat... The number of observations in the window is based on the time column, thus both the number of observations per window and the number of observations to slide the window may vary The function should accept any window size up to the number of observations and a step size.
Here is sample data (see "Edit:" for a larger sample set)
set.seed(42)
dat <- data.frame(time = seq(1:20)+runif(20,0,1))
dat <- data.frame(dat, measure=c(diff(dat$time),NA_real_))
dat$measure[sample(1:19,2)] <- NA_real_
head(dat)
time measure
1 1.914806 1.0222694
2 2.937075 0.3490641
3 3.286140 NA
4 4.830448 0.8112979
5 5.641746 0.8773504
6 6.519096 1.2174924
Desired Output for the specific case of a 5 second window, 2.5 second step, first window from -2.5 to 2.5, na.rm=FALSE:
[1] 1.0222694
[2] NA
[3] NA
[4] 1.0126639
[5] 0.9965048
[6] 0.9514456
[7] 1.0518228
[8] NA
[9] NA
[10] NA
Explanation: In the desired output the very first window looks for times between -2.5 and 2.5. One observation of measure is in this window, and it is not an NA, thus we get that observation: 1.0222694. The next window is from 0 to 5, and there is an NA in the window, so we get NA. Same for the window from 2.5 to 7.5. The next window is from 5 to 10. There are 5 observations in the window, none are NA. So, we get the average of those 5 observations (i.e. mean(dat[dat$time >5 & dat$time <10,'measure']) )
What I tried: Here is what I tried for the specific case of a window where the step size is 1/2 the window duration:
windo <- 5 # duration in seconds of window
# partition into groups depending on which window(s) an observation falls in
# When step size >= window/2 and < window, need two grouping vectors
leaf1 <- round(ceiling(dat$time/(windo/2))+0.5)
leaf2 <- round(ceiling(dat$time/(windo/2))-0.5)
l1 <- tapply(dat$measure, leaf1, mean)
l2 <- tapply(dat$measure, leaf2, mean)
as.vector(rbind(l2,l1))
Not flexible, not elegant, not efficient. If step size isn't 1/2 window size, the approach will not work, as is.
Any thoughts on a general solution to this kind of problem? Any solution is acceptable. The faster the better, though I prefer solutions using base R, data.table, Rcpp, and/or parallel computation. In my real data set, there are several millions of observations contained in a list of data frames (max data frame is ~400,000 observations).
Below is a extra info: A larger sample set
Edit: As per request, here is a larger, more realistic example dataset with many more NAs and the minimum time span (~0.03). To be clear, though, the list of data frames contains small ones like the one above, as well as ones like the following and larger:
set.seed(42)
dat <- data.frame(time = seq(1:50000)+runif(50000, 0.025, 1))
dat <- data.frame(dat, measure=c(diff(dat$time),NA_real_))
dat$measure[sample(1:50000,1000)] <- NA_real_
dat$measure[c(350:450,3000:3300, 20000:28100)] <- NA_real_
dat <- dat[-c(1000:2000, 30000:35000),]
# a list with a realistic number of observations:
dat <- lapply(1:300,function(x) dat)
Here is an attempt with Rcpp. The function assumes that data is sorted according to time. More testing would be advisable and adjustments could be made.
#include <Rcpp.h>
using namespace Rcpp;
// [[Rcpp::export]]
NumericVector rollAverage(const NumericVector & times,
NumericVector & vals,
double start,
const double winlen,
const double winshift) {
int n = ceil((max(times) - start) / winshift);
NumericVector winvals;
NumericVector means(n);
int ind1(0), ind2(0);
for(int i=0; i < n; i++) {
if (times[0] < (start+winlen)) {
while((times[ind1] <= start) &
(times[ind1+1] <= (start+winlen)) &
(ind1 < (times.size() - 1))) {
ind1++;
}
while((times[ind2+1] <= (start+winlen)) & (ind2 < (times.size() - 1))) {
ind2++;
}
if (times[ind1] >= start) {
winvals = vals[seq(ind1, ind2)];
means[i] = mean(winvals);
} else {
means[i] = NA_REAL;
}
} else {
means[i] = NA_REAL;
}
start += winshift;
}
return means;
}
Testing it:
set.seed(42)
dat <- data.frame(time = seq(1:20)+runif(20,0,1))
dat <- data.frame(dat, measure=c(diff(dat$time),NA_real_))
dat$measure[sample(1:19,2)] <- NA_real_
rollAverage(dat$time, dat$measure, -2.5, 5.0, 2.5)
#[1] 1.0222694 NA NA 1.0126639 0.9965048 0.9514456 1.0518228 NA NA NA
With your list of data.frames (using data.table):
set.seed(42)
dat <- data.frame(time = seq(1:50000)+runif(50000, 0.025, 1))
dat <- data.frame(dat, measure=c(diff(dat$time),NA_real_))
dat$measure[sample(1:50000,1000)] <- NA_real_
dat$measure[c(350:450,3000:3300, 20000:28100)] <- NA_real_
dat <- dat[-c(1000:2000, 30000:35000),]
# a list with a realistic number of observations:
dat <- lapply(1:300,function(x) dat)
library(data.table)
dat <- lapply(dat, setDT)
for (ind in seq_along(dat)) dat[[ind]][, i := ind]
#possibly there is a way to avoid these copies?
dat <- rbindlist(dat)
system.time(res <- dat[, rollAverage(time, measure, -2.5, 5.0, 2.5), by=i])
#user system elapsed
#1.51 0.02 1.54
print(res)
# i V1
# 1: 1 1.0217126
# 2: 1 0.9334415
# 3: 1 0.9609050
# 4: 1 1.0123473
# 5: 1 0.9965922
# ---
#6000596: 300 1.1121296
#6000597: 300 0.9984581
#6000598: 300 1.0093060
#6000599: 300 NA
#6000600: 300 NA
Here is a function that gives the same result for your small data frame. It's not particularly quick: it takes several seconds to run on one of the larger datasets in your second dat example.
rolling_summary <- function(DF, time_col, fun, window_size, step_size, min_window=min(DF[, time_col])) {
# time_col is name of time column
# fun is function to apply to the subsetted data frames
# min_window is the start time of the earliest window
times <- DF[, time_col]
# window_starts is a vector of the windows' minimum times
window_starts <- seq(from=min_window, to=max(times), by=step_size)
# The i-th element of window_rows is a vector that tells us the row numbers of
# the data-frame rows that are present in window i
window_rows <- lapply(window_starts, function(x) { which(times>=x & times<x+window_size) })
window_summaries <- sapply(window_rows, function(w_r) fun(DF[w_r, ]))
data.frame(start_time=window_starts, end_time=window_starts+window_size, summary=window_summaries)
}
rolling_summary(DF=dat,
time_col="time",
fun=function(DF) mean(DF$measure),
window_size=5,
step_size=2.5,
min_window=-2.5)
Here are some functions that will give the same output on your first example:
partition <- function(x, window, step = 0){
a = x[x < step]
b = x[x >= step]
ia = rep(0, length(a))
ib = cut(b, seq(step, max(b) + window, by = window))
c(ia, ib)
}
roll <- function(df, window, step = 0, fun, ...){
tapply(df$measure, partition(df$time, window, step), fun, ...)
}
roll_steps <- function(df, window, steps, fun, ...){
X = lapply(steps, roll, df = df, window = window, fun = fun, ...)
names(X) = steps
X
}
Output for your first example:
> roll_steps(dat, 5, c(0, 2.5), mean)
$`0`
1 2 3 4 5
NA 1.0126639 0.9514456 NA NA
$`2.5`
0 1 2 3 4
1.0222694 NA 0.9965048 1.0518228 NA
You can also ignore missing values this way easily:
> roll_steps(dat, 5, c(0, 2.5), mean, na.rm = TRUE)
$`0`
1 2 3 4 5
0.7275438 1.0126639 0.9514456 0.9351326 NaN
$`2.5`
0 1 2 3 4
1.0222694 0.8138012 0.9965048 1.0518228 0.6122983
This can also be used for a list of data.frames:
> x = lapply(dat2, roll_steps, 5, c(0, 2.5), mean)
Ok, how about this.
library(data.table)
dat <- data.table(dat)
setkey(dat, time)
# function to compute a given stat over a time window on a given data.table
window_summary <- function(start_tm, window_len, stat_fn, my_dt) {
pos_vec <- my_dt[, which(time>=start_tm & time<=start_tm+window_len)]
return(stat_fn(my_dt$measure[pos_vec]))
}
# a vector of window start times
start_vec <- seq(from=-2.5, to=dat$time[nrow(dat)], by=2.5)
# sapply'ing the function above over vector of start times
# (in this case, getting mean over 5 second windows)
result <- sapply(start_vec, window_summary,
window_len=5, stat_fn=mean, my_dt=dat)
On my machine, it processes the first 20,000 rows of your large dataset in 13.06781 secs; all rows in 51.58614 secs
Here's another attempt to use pure data.table approach and its between function.
Have compared Rprof against the above answers (except #Rolands answer) and it seems the most optimized one.
Haven't tested for bugs though, but if you"ll like it, I'll expand the answer.
Using your dat from above
library(data.table)
Rollfunc <- function(dat, time, measure, wind = 5, slide = 2.5, FUN = mean, ...){
temp <- seq.int(-slide, max(dat$time), by = slide)
temp <- cbind(temp, temp + wind)
setDT(dat)[, apply(temp, 1, function(x) FUN(measure[between(time, x[1], x[2])], ...))]
}
Rollfunc(dat, time, measure, 5, 2.5)
## [1] 1.0222694 NA NA 1.0126639 0.9965048 0.9514456 1.0518228 NA NA
## [10] NA
You can also specify the functions and its arguments, i.e., for example:
Rollfunc(dat, time, measure, 5, 2.5, max, na.rm = TRUE)
will also work
Edit: I did some benchnarks against #Roland and his method clearly wins (by far), so I would go with the Rcpp aproach

Resources