with df like below
df <- data.frame(
name = rep(c("A", "B", "C"),2),
type = c("10", "10", "10","20", "20", "20"),
val = c(1,2,3,4,5,6)
)
> df
name type val
1 A 10 1
2 B 10 2
3 C 10 3
4 A 20 4
5 B 20 5
6 C 20 6
>
the expected output is
I need to add val of all records with name C to val of records with name A for the corresponding type with a new name AC. Need an output keeping name C and without it.
output1
name type val
1 A 10 1
2 B 10 2
3 C 10 3
4 AC 10 4
5 A 20 4
6 B 20 5
7 C 20 6
8 AC 20 10
output2
name type val
1 AC 10 4
2 B 10 2
4 AC 20 10
5 B 20 5
>
prefer dplyr based solution
Here is one way,
library(dplyr)
df %>%
mutate(new = as.integer(name %in% c('A', 'C'))) %>%
group_by(type, new) %>%
summarise(name = paste0(name, collapse = ''), val = sum(val)) %>%
ungroup() %>%
select(-new)
# A tibble: 4 × 3
# type name val
# <fctr> <chr> <dbl>
#1 10 B 2
#2 10 AC 4
#3 20 B 5
#4 20 AC 10
To get the other output then,
df %>%
mutate(new = as.integer(name %in% c('A', 'C'))) %>%
group_by(type, new) %>%
summarise(name = paste0(name, collapse = ''), val = sum(val)) %>%
ungroup() %>%
select(-new) %>%
filter(nchar(name) > 1) %>%
bind_rows( df) %>%
arrange(val)
# A tibble: 8 × 3
# type name val
# <fctr> <chr> <dbl>
#1 10 A 1
#2 10 B 2
#3 10 C 3
#4 10 AC 4
#5 20 A 4
#6 20 B 5
#7 20 C 6
#8 20 AC 10
Here is another (requires tidyr as well as dplyr)
df1 <- df %>% group_by(type) %>%
summarise(AC=sum(val[name %in% c("A","C")]),B=val[name=="B"]) %>%
gather(key=name,value=val,-type) %>%
arrange(type)
Here is one option using data.table
library(data.table)
rbindlist(list(df, setDT(df)[, .(name = "AC", val = sum(val[as.character(name) %chin%
c("A", "C")])) , .(type)][, names(df), with = FALSE]))[order(type, name)]
# name type val
#1: A 10 1
#2: B 10 2
#3: C 10 3
#4: AC 10 4
#5: A 20 4
#6: B 20 5
#7: C 20 6
#8: AC 20 10
Or with dplyr
library(dplyr)
df %>%
filter(name %in% c("A", "C")) %>%
group_by(type) %>%
summarise(name = 'AC', val = sum(val)) %>%
full_join(df, ., on = 'type') %>%
arrange(type, val)
# name type val
#1 A 10 1
#2 B 10 2
#3 C 10 3
#4 AC 10 4
#5 A 20 4
#6 B 20 5
#7 C 20 6
#8 AC 20 10
Related
data <- tibble(time = c(1,1,2,2), a = c(1,2,3,4), b =c(4,3,2,1), c = c(1,1,1,1))
The result will look like this
result <- tibble(
t = c(1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,2),
firm1 = c("a","a","a","b","b","b","c","c","c","a","a","a","b","b","b","c","c","c"),
firm2 = c("a","b","c","a","b","c","a","b","c","a","b","c","a","b","c","a","b","c"),
value = c(6,10,5,10,14,9,5,9,4,14,10,9,10,6,5,9,5,4))
result
The function could be
function(x, y){sum(x, y)}
Basically I am looking for a tidy solution to expand.grid data at each point of time and apply functions across columns. Can anyone help?
I tried this, but I could not have time in front of the pairs.
expected_result<-expand.grid(names(data[-1]), names(data[-1])) %>%
mutate(value = map2(Var1, Var2, ~ fun1(data[.x], data[.y])))
expected_result
Use exand.grid you get all possible combination of columns, split the data by time and apply fun for each row of tmp.
library(dplyr)
library(purrr)
tmp <- expand.grid(firm1 = names(data[-1]), firm2 = names(data[-1]))
fun <- function(x, y) sum(x, y)
result <- data %>%
group_split(time) %>%
map_df(~cbind(time = .x$time[1], tmp,
value = apply(tmp, 1, function(x) fun(.x[[x[1]]], .x[[x[2]]]))))
result
# time firm1 firm2 value
#1 1 a a 6
#2 1 b a 10
#3 1 c a 5
#4 1 a b 10
#5 1 b b 14
#6 1 c b 9
#7 1 a c 5
#8 1 b c 9
#9 1 c c 4
#10 2 a a 14
#11 2 b a 10
#12 2 c a 9
#13 2 a b 10
#14 2 b b 6
#15 2 c b 5
#16 2 a c 9
#17 2 b c 5
#18 2 c c 4
You may also do this in base R -
result <- do.call(rbind, by(data, data$time, function(x) {
cbind(time = x$time[1], tmp,
value = apply(tmp, 1, function(y) fun(x[[y[1]]], x[[y[2]]])))
}))
We may use
library(dplyr)
library(tidyr)
library(purrr)
data1 <- data %>%
group_by(time) %>%
summarise(across(everything(), sum, na.rm = TRUE), .groups = 'drop') %>%
pivot_longer(cols = -time) %>%
group_split(time)
map_dfr(data1, ~ {dat <- .x
crossing(firm1 = dat$name, firm2 = dat$name) %>%
mutate(value = c(outer(dat$value, dat$value, FUN = `+`))) %>%
mutate(time = first(dat$time), .before = 1)})
-output
# A tibble: 18 × 4
time firm1 firm2 value
<dbl> <chr> <chr> <dbl>
1 1 a a 6
2 1 a b 10
3 1 a c 5
4 1 b a 10
5 1 b b 14
6 1 b c 9
7 1 c a 5
8 1 c b 9
9 1 c c 4
10 2 a a 14
11 2 a b 10
12 2 a c 9
13 2 b a 10
14 2 b b 6
15 2 b c 5
16 2 c a 9
17 2 c b 5
18 2 c c 4
I have a list here, and I wish to mutate a new column with unique values for each list relative to the mutation. For example, I want to mutate a column named ID as n >= 1.
Naturally, on a dataframe I would do this:
dat %>% mutate(id = row_number())
For a list, I would do this:
dat%>% map(~ mutate(., ID = row_number()))
And I would get an output likeso:
dat <- list(data.frame(x=c("a", "b" ,"c", "d", "e" ,"f" ,"g") ), data.frame(y=c("p", "lk", "n", "m", "g", "f", "t")))
[[1]]
x id
1 a 1
2 b 2
3 c 3
4 d 4
5 e 5
6 f 6
7 g 7
[[2]]
y id
1 p 1
2 lk 2
3 n 3
4 m 4
5 g 5
6 f 6
7 t 7
Though, how would I mutate a new column ID such that the row number continues from the first list.
Expected output:
[[1]]
x id
1 a 1
2 b 2
3 c 3
4 d 4
5 e 5
6 f 6
7 g 7
[[2]]
y id
1 p 8
2 lk 9
3 n 10
4 m 11
5 g 12
6 f 13
7 t 14
An option is to bind them into a single dataset, create the 'id' with row_number(), split by 'grp', loop over the list and remove any columns that have all NA values
library(dplyr)
library(purrr)
dat %>%
bind_rows(.id = 'grp') %>%
mutate(id = row_number()) %>%
group_split(grp) %>%
map(~ .x %>%
select(where(~ any(!is.na(.))), -grp))
-output
#[[1]]
# A tibble: 7 x 2
# x id
# <chr> <int>
#1 a 1
#2 b 2
#3 c 3
#4 d 4
#5 e 5
#6 f 6
#7 g 7
#[[2]]
# A tibble: 7 x 2
# y id
# <chr> <int>
#1 p 8
#2 lk 9
#3 n 10
#4 m 11
#5 g 12
#6 f 13
#7 t 14
Or an easier approach is to unlist (assuming single column), get the sequence, add a new column with map2
map2(dat, relist(seq_along(unlist(dat)), skeleton = dat),
~ .x %>% mutate(id = .y))
Or using a for loop
dat[[1]]$id <- seq_len(nrow(dat[[1]]))
for(i in seq_along(dat)[-1]) dat[[i]]$id <-
seq(tail(dat[[i-1]]$id, 1) + 1, length.out = nrow(dat[[i]]), by = 1)
Take an example dataframe like so (the real dataframe has more columns):
df <- data.frame(A = seq(1, 3, 1),
B = seq(4, 6, 1))
I can use pivot_longer to collect my columns of interest (A and B) like so:
library(dplyr)
library(tidyr)
df <- df %>%
pivot_longer(cols = c("A", "B"), names_to = "Letter", values_to = "Number")
df
Letter Number
<chr> <dbl>
1 A 1
2 B 4
3 A 2
4 B 5
5 A 3
6 B 6
Now let's say I have another column C in my dataframe, making it no longer tidy
C <- seq(7, 12, 1)
df_2 <- data.frame(df, C)
df_2
Letter Number C
1 A 1 7
2 B 4 8
3 A 2 9
4 B 5 10
5 A 3 11
6 B 6 12
I want to use pivot_longer again to make df_2 tidy and get this output:
data.frame(Letter = c(rep("A", 3), rep("B", 3), rep("C", 3)),
Number = seq(1, 12, 1))
Letter Number
1 A 1
2 A 2
3 A 3
4 B 4
5 B 5
6 B 6
7 C 7
8 C 8
9 C 9
10 C 10
11 C 11
12 C 12
Using the same strategy creates an error though:
df_2 %>%
pivot_longer(cols = "C", names_to = "Letter", values_to = "Number")
Error: Failed to create output due to bad names.
* Choose another strategy with `names_repair`
Setting names_repair to minimal runs but doesn't produce the output I want.
Follow it like this
library(tidyverse)
df <- data.frame(A = seq(1, 3, 1),
B = seq(4, 6, 1))
df <- df %>%
pivot_longer(cols = c("A", "B"), names_to = "Letter", values_to = "Number")
C <- seq(7, 12, 1)
df_2 <- data.frame(C)
df_2 <- df_2 %>% pivot_longer(cols = C, names_to = "Letter", values_to = "Number")
df_result <- rbind(df, df_2)
Output
> df_result
# A tibble: 12 x 2
Letter Number
<chr> <dbl>
1 A 1
2 B 4
3 A 2
4 B 5
5 A 3
6 B 6
7 C 7
8 C 8
9 C 9
10 C 10
11 C 11
12 C 12
Maybe try this if it is helpful:
library(tidyverse)
#Code
df_2 %>% pivot_longer(everything()) %>%
arrange(name) %>% group_by(name) %>%
filter(!duplicated(value))
Output:
# A tibble: 12 x 2
# Groups: name [3]
name value
<chr> <dbl>
1 A 1
2 A 2
3 A 3
4 B 4
5 B 5
6 B 6
7 C 7
8 C 8
9 C 9
10 C 10
11 C 11
12 C 12
We could do this easily with stack
library(dplyr)
stack(df_2)[2:1] %>%
distinct %>%
set_names(c("Letter", "Number"))
-output
# Letter Number
#1 A 1
#2 A 2
#3 A 3
#4 B 4
#5 B 5
#6 B 6
#7 C 7
#8 C 8
#9 C 9
#10 C 10
#11 C 11
#12 C 12
Or an option with unnest/enframe
library(tidyr)
library(tibble)
unclass(df_2) %>%
enframe(name = "Letter", value = "Number") %>%
unnest(c(Number)) %>%
distinct
Or using melt
library(reshape2)
melt(df_2) %>%
distinct()
Or in a single line in base R
unique(stack(df_2)[2:1])
I am trying to figure out how to sum values belonging to category a and b by factor file, but also keep the original data.
library(dplyr)
df <- data.frame(ID = 1:20, values = runif(20), category = rep(letters[1:5], 4), file = as.factor(sort(rep(1:5, 4))))
ID values category file
1 1 0.65699229 a 1
2 2 0.70506478 b 1
3 3 0.45774178 c 1
4 4 0.71911225 d 1
5 5 0.93467225 e 1
6 6 0.25542882 a 2
7 7 0.46229282 b 2
8 8 0.94001452 c 2
9 9 0.97822643 d 2
10 10 0.11748736 e 2
11 11 0.47499708 a 3
12 12 0.56033275 b 3
13 13 0.90403139 c 3
14 14 0.13871017 d 3
15 15 0.98889173 e 3
16 16 0.94666823 a 4
17 17 0.08243756 b 4
18 18 0.51421178 c 4
19 19 0.39020347 d 4
20 20 0.90573813 e 4
so that
df[1,2] will be added to df[2,2] to category 'ab' for file 1
df[6,2] will be added to df[7,2] to category 'ab' for file 2
etc.
So far I have this:
df %>%
filter(category %in% c('a' , 'b')) %>%
group_by(file) %>%
summarise(values = sum(values))
Problem
I would like to change the category of the summed values to "ab" and append it to the original data frame in the same pipeline.
Desired output:
ID values category file
1 1 0.65699229 a 1
2 2 0.70506478 b 1
3 3 0.45774178 c 1
4 4 0.71911225 d 1
5 5 0.93467225 e 1
6 6 0.25542882 a 2
7 7 0.46229282 b 2
8 8 0.94001452 c 2
9 9 0.97822643 d 2
10 10 0.11748736 e 2
11 11 0.47499708 a 3
12 12 0.56033275 b 3
13 13 0.90403139 c 3
14 14 0.13871017 d 3
15 15 0.98889173 e 3
16 16 0.94666823 a 4
17 17 0.08243756 b 4
18 18 0.51421178 c 4
19 19 0.39020347 d 4
20 20 0.90573813 e 4
21 21 1.25486225 ab 1
22 22 1.87216325 ab 2
23 23 1.36548126 ab 3
This will get you the result
df %>% bind_rows(
df %>%
filter(category %in% c('a' , 'b')) %>%
group_by(file) %>%
mutate(values = sum(values), category = paste0(category,collapse='')) %>%
filter(row_number() == 1 & n() > 1)
) %>% mutate(ID = row_number())
BTW the code pro produce the dataframe in the example is this one:
df <- data.frame(ID = 1:20, values = runif(20), category = rep(letters[1:5], 4), file = as.factor(sort(rep(1:4, 5))))
now lets say you want to sum multiple columns, you need to provide the list in a vector:
cols = c("values") # columns to be sum
df %>% bind_rows(
df %>%
filter(category %in% c('a' , 'b')) %>%
group_by(file) %>%
mutate_at(vars(cols), sum) %>%
mutate(category = paste0(category,collapse='')) %>%
filter(row_number() == 1 & n() > 1)
) %>% mutate(ID = row_number())
library(dplyr)
df1 %>%
filter(category %in% c('a' , 'b')) %>%
group_by(file) %>%
filter(n_distinct(category) > 1) %>%
summarise(values = sum(values)) %>%
mutate(category="ab",
ID=max(df1$ID)+1:n()) %>%
bind_rows(df1, .)
#> Warning in bind_rows_(x, .id): binding factor and character vector,
#> coercing into character vector
#> Warning in bind_rows_(x, .id): binding character and factor vector,
#> coercing into character vector
#> ID values category file
#> 1 1 0.62585921 a 1
#> 2 2 0.61865851 b 1
#> 3 3 0.05274456 c 1
#> 4 4 0.68156961 d 1
.
.
.
#> 19 19 0.43239411 d 5
#> 20 20 0.85886314 e 5
#> 21 21 1.24451773 ab 1
#> 22 22 0.99001810 ab 2
#> 23 23 1.25331943 ab 3
This data.table approach uses a self-join to get all of the possible two-character combinations.
library(data.table)
setDT(df)
df_self_join <- df[df, on = .(file), allow.cartesian = T
][category != i.category,
.(category = paste0(i.category, category), values = values + i.values, file)
][order(category), .(ID = .I + nrow(df), values, category, file)]
rbindlist(list(df, df_self_join))
ID values category file
1: 1 0.76984382 a 1
2: 2 0.54311583 b 1
3: 3 0.23462016 c 1
4: 4 0.60179043 d 1
...
20: 20 0.03534223 e 5
21: 21 1.31295965 ab 1
22: 22 0.51666175 ab 2
23: 23 1.02305754 ab 3
24: 24 1.00446399 ac 1
25: 25 0.96910373 ac 2
26: 26 0.87795389 ac 4
#total of 80 rows
Here is pretty close dplyr translation:
library(dplyr)
tib <- as_tibble(df)
inner_join(tib, tib, by = 'file')%>%
filter(ID.x != ID.y)%>%
transmute(category = paste0(category.x, category.y)
, values = values.x + values.y
, file)%>%
arrange(category)%>%
bind_rows(tib, .)%>%
mutate(ID = row_number())%>%
filter(category == 'ab') #filter added to show the "ab" files
# A tibble: 3 x 4
ID values category file
<int> <dbl> <chr> <fct>
1 21 1.31 ab 1
2 22 0.517 ab 2
3 23 1.02 ab 3
I have data:
rowID incidentID participant.type
1 1 A
2 1 B
3 2 A
4 3 A
5 3 B
6 3 C
7 4 B
8 4 C
And I would like to end up with:
rowID incident participant.type participant.type.1 participant.type.2
1 1 A B
2 2 A
3 3 A B C
4 4 B C
I tried the spread but can't achieve one line per incident; I don't think I have a way of creating a key-value pair so I wonder if there is some other method for doing this.
Before using spread(), you need to create a proper key argument.
df %>% select(-rowID) %>%
group_by(incidentID) %>%
mutate(id = 1:n()) %>%
spread(id, participant.type)
# incidentID `1` `2` `3`
# <int> <fct> <fct> <fct>
# 1 1 A B NA
# 2 2 A NA NA
# 3 3 A B C
# 4 4 B C NA
Since your grouping is based on the row order within the icidentID column. The following simple solution will also work.
It is just filtering the dataframe and then merging in the end.
It is probably not the best solution in terms of effective use of computing power, but it is easy to understand.
library(tidyverse)
df <-
tribble(
~rowID, ~incidentID, ~participant.type,
1, 1, "A",
2, 1, "B",
3, 2, "A",
4, 3, "A",
5, 3, "B",
6, 3, "C",
7, 4, "B",
8, 4, "C")
df_1 <- df %>%
select(-rowID) %>%
group_by(incidentID) %>%
filter(row_number()==1)
df_2 <- df %>%
select(-rowID) %>%
group_by(incidentID) %>%
filter(row_number()==2) %>%
rename(participant.type.1 = participant.type)
df_3 <- df %>%
select(-rowID) %>%
group_by(incidentID) %>%
filter(row_number()==3) %>%
rename(participant.type.2 = participant.type)
full_join(df_1, full_join(df_2, df_3))
Result:
Joining, by = "incidentID"
Joining, by = "incidentID"
# A tibble: 4 x 4
# Groups: incidentID [?]
incidentID participant.type participant.type.1 participant.type.2
<dbl> <chr> <chr> <chr>
1 1 A B NA
2 2 A NA NA
3 3 A B C
4 4 B C NA
Here's my solution:
df %>%
select(-rowID) %>%
group_by(incidentID) %>%
nest() %>%
mutate(data = map_chr(data, ~str_c(.x$participant.type, collapse = '_'))) %>%
separate(data, paste0('participant.type.', 0:2)) %>%
mutate_at(2:4, ~replace_na(.x, ''))
We can use reshape2::dcast for this
reshape2::dcast(df, insidentID ~ participant.type)
# insidentID A B C
# 1 1 <NA> B <NA>
# 2 8 <NA> B <NA>
# 3 12 <NA> <NA> C
# 4 16 A <NA> <NA>
# 5 24 <NA> B <NA>
# 6 27 <NA> B C
# 7 29 <NA> <NA> C
with the data
set.seed(123)
df <- data.frame(insidentID = sample(0:30, 8L, replace = TRUE),
participant.type = sample(LETTERS[1:3], 8L, replace = TRUE),
stringsAsFactors = FALSE)
df
# insidentID participant.type
# 1 8 B
# 2 24 B
# 3 12 C
# 4 27 B
# 5 29 C
# 6 1 B
# 7 16 A
# 8 27 C
The 'related question' link provided by #markus shows a variety of other solutions, including what appears to be the most concise in a tidyverse format:
df %>%
group_by(incidentID) %>%
mutate(rn = paste0("newcolumn",row_number())) %>%
spread(rn, participant.type)
gives:
incidentID newcolumn1 newcolumn2 newcolumn3
<int> <fct> <fct> <fct>
1 1 A B NA
2 2 A NA NA
3 3 A B C
4 4 B C NA
A