Get unique values from a specific group for top N - r

I'm working with a "movie" dataset. I have a movie "title" column (col no 1) and a "overall_score" column (col no 13).
Apparently multiple movies has scored 10, so when I make the top 10, it only shows me all movie with score 10.
But I only want the score 10, 9, 8 and so on until 1 to appear only 3 times. I tired using the slice function but wasn't successful in that, what do you think I'm doing wrong?
Here's my code:
movie2 <- movie_reviews %>%
arrange(desc(Overall)) %>%
group_by(uid, title) %>%
head(10) %>% slice(13:3)

If you don't care about which movies are within the score subgroups, then you could just use row_number to assign a unique number per Overall group.
library(dplyr)
set.seed(1)
movie_reviews <- data.frame(
uid = 1:100,
title = paste("title", 1:100),
Overall = sample(1:10, 100, replace=T)
)
movie2 <- movie_reviews %>%
group_by(Overall) %>%
mutate(rn = row_number()) %>%
ungroup() %>%
filter(rn < 4)%>%
select(-rn) %>%
arrange(Overall)
> movie2
# A tibble: 30 × 4
uid title Overall rn
<int> <chr> <int> <int>
1 4 title 4 1 3
2 9 title 9 1 2
3 64 title 64 1 1
4 23 title 23 2 1
5 82 title 82 2 2
6 87 title 87 2 3
7 8 title 8 3 3
8 57 title 57 3 2
9 80 title 80 3 1
10 27 title 27 4 1
# … with 20 more rows

Related

Replace 'middle' frequencies with averaged frequency values

I have this type of data, with frequency data and position data grouped by rowid:
df
rowid word f position
1 2 i 700 1
2 2 'm 600 2
3 2 fine 1 3
4 3 how 400 1
5 3 's 500 2
6 3 the 700 3
7 3 weather 20 4
8 4 it 390 1
9 4 's 500 2
10 4 really 177 3
11 4 very 200 4
12 4 cold 35 5
13 5 i 700 1
14 5 love 199 2
15 5 you 400 3
The task I'm facing seems simple: in those rowids where there are more than 3 positions, I need to replace the frequencies of all middle positions with their average. The following approach works but seems over-convoluted, so I'm almost certain there will be a more straightforward dplyrway to get the desired output:
df %>%
group_by(rowid) %>%
# filter for 'middle' positions:
filter(position != first(position) & position != last(position)) %>%
# summarise:
summarize(across(position),
# create average frequency:
f_middle_position = mean(f, na.rm = TRUE),
# concatenate words:
word = str_c(word, collapse = " ")
) %>%
filter(!duplicated(f_middle_position)) %>%
# join with df:
left_join(df, ., by = c("rowid", "position")) %>%
# remove rows other than #1,#2, and last:
group_by(rowid) %>%
# create row count:
mutate(rn = row_number()) %>%
# filter first, second, and last row per group:
filter(rn %in% c(1, 2, last(rn))) %>%
# transfer frequencies for middle positions:
mutate(f = ifelse(is.na(f_middle_position), f, f_middle_position)) %>%
# make more changes:
mutate(
# change position labels:
position = ifelse(position == first(position), 1,
ifelse(position == last(position), 2, 1.5)),
# update word:
word = ifelse(is.na(word.y), word.x, word.y)
) %>%
# remove obsolete variables:
select(-c(f_middle_position, word.y, word.x,rn))
A tibble: 12 × 4
# Groups: rowid [4]
rowid f position word
<dbl> <dbl> <dbl> <chr>
1 2 700 1 i
2 2 600 1.5 'm
3 2 1 2 fine
4 3 400 1 how
5 3 600 1.5 's the
6 3 20 2 weather
7 4 390 1 it
8 4 292. 1.5 's really very
9 4 35 2 cold
10 5 700 1 i
11 5 199 1.5 love
12 5 400 2 you
How can this result be obtained in a more concise way in dplyr and, preferably without the left_join, which causes problems with my actual data?
Data:
df <- data.frame(
rowid = c(2,2,2,3,3,3,3,4,4,4,4,4,5,5,5),
word = c("i","'m","fine",
"how","'s","the","weather",
"it","'s","really", "very","cold",
"i","love","you"),
f = c(700,600,1,
400,500,700,20,
390,500,177,200,35,
700,199,400),
position = c(1,2,3,
1,2,3,4,
1,2,3,4,5,
1,2,3)
)
You can create a group variable pos that marks the first row with 1, the middle with 1.5, and the last with 2. Then group the data by rowid and pos and apply mean() and paste() on f and word respectively.
library(dplyr)
df %>%
group_by(rowid) %>%
mutate(pos = case_when(position == 1 ~ 1, position == n() ~ 2, TRUE ~ 1.5)) %>%
group_by(rowid, pos) %>%
summarise(f = mean(f), word = paste(word, collapse = ' '), .groups = 'drop')
# # A tibble: 12 × 4
# rowid pos f word
# <dbl> <dbl> <dbl> <chr>
# 1 2 1 700 i
# 2 2 1.5 600 'm
# 3 2 2 1 fine
# 4 3 1 400 how
# 5 3 1.5 600 's the
# 6 3 2 20 weather
# 7 4 1 390 it
# 8 4 1.5 292. 's really very
# 9 4 2 35 cold
# 10 5 1 700 i
# 11 5 1.5 199 love
# 12 5 2 400 you

apply function or loop within mutate

Let's say I have a data frame. I would like to mutate new columns by subtracting each pair of the existing columns. There are rules in the matching columns. For example, in the below codes, the prefix is all same for the first component (base_g00) of the subtraction and the same for the second component (allow_m00). Also, the first component has numbers from 27 to 43 for the id and the second component's id is from 20 to 36 also can be interpreted as (1st_id-7). I am wondering for the following code, can I write in a apply function or loops within mutate format to make the codes simpler. Thanks so much for any suggestions in advance!
pred_error<-y07_13%>%mutate(annual_util_1=base_g0027-allow_m0020,
annual_util_2=base_g0028-allow_m0021,
annual_util_3=base_g0029-allow_m0022,
annual_util_4=base_g0030-allow_m0023,
annual_util_5=base_g0031-allow_m0024,
annual_util_6=base_g0032-allow_m0025,
annual_util_7=base_g0033-allow_m0026,
annual_util_8=base_g0034-allow_m0027,
annual_util_9=base_g0035-allow_m0028,
annual_util_10=base_g0036-allow_m0029,
annual_util_11=base_g0037-allow_m0030,
annual_util_12=base_g0038-allow_m0031,
annual_util_13=base_g0039-allow_m0032,
annual_util_14=base_g0040-allow_m0033,
annual_util_15=base_g0041-allow_m0034,
annual_util_16=base_g0042-allow_m0035,
annual_util_17=base_g0043-allow_m0036)
I think a more idiomatic tidyverse approach would be to reshape your data so those column groups are encoded as a variable instead of as separate columns which have the same semantic meaning.
For instance,
library(dplyr); library(tidyr); library(stringr)
y07_13 <- tibble(allow_m0021 = 1:5,
allow_m0022 = 2:6,
allow_m0023 = 11:15,
base_g0028 = 5,
base_g0029 = 3:7,
base_g0030 = 100)
y07_13 %>%
mutate(row = row_number()) %>%
pivot_longer(-row) %>%
mutate(type = str_extract(name, "allow_m|base_g"),
num = str_remove(name, type) %>% as.numeric(),
group = num - if_else(type == "allow_m", 20, 27)) %>%
select(row, type, group, value) %>%
pivot_wider(names_from = type, values_from = value) %>%
mutate(annual_util = base_g - allow_m)
Result
# A tibble: 15 x 5
row group allow_m base_g annual_util
<int> <dbl> <dbl> <dbl> <dbl>
1 1 1 1 5 4
2 1 2 2 3 1
3 1 3 11 100 89
4 2 1 2 5 3
5 2 2 3 4 1
6 2 3 12 100 88
7 3 1 3 5 2
8 3 2 4 5 1
9 3 3 13 100 87
10 4 1 4 5 1
11 4 2 5 6 1
12 4 3 14 100 86
13 5 1 5 5 0
14 5 2 6 7 1
15 5 3 15 100 85
Here is vectorised base R approach -
base_cols <- paste0("base_g00", 27:43)
allow_cols <- paste0("allow_m00", 20:36)
new_cols <- paste0("annual_util", 1:17)
y07_13[new_cols] <- y07_13[base_cols] - y07_13[allow_cols]
y07_13

paste together first and last value by group

I have a df that looks like this:
group sequence link
90 1 11|S1
90 2 10|S1
90 3 12|10
91 1 9|10
91 2 13|9
93 1 15|20
...
How can I store the first and last value of the linkvariable in each group as a new variable?
Desired output is:
group sequence link Key
90 1 11|S1 11|S1, 12|10
90 2 10|S1 11|S1, 12|10
90 3 12|10 11|S1, 12|10
91 1 9|10 9|10, 13|9
91 2 13|9 9|10,13|9
93 1 15|20
....
You could do:
library(dplyr)
df %>%
group_by(group) %>%
mutate(
Key = paste(link[1], link[n()], sep = ", ")
)
Though that wouldn't match your desired output. In your example data frame, you have e.g. the group 91 where there's only 1 value. The above code would give you 9|10 repeatedly both as beginning and end.
If you'd like to only display one value in such cases, you can do:
df %>%
group_by(group) %>%
mutate(
Key = case_when(
n() > 1 ~ paste(link[1], link[n()], sep = ", "),
TRUE ~ as.character(link)
)
)
I think you could use arrange() and slice() to find the first/last links in your data. My solution is a lengthier than #arg0naut91's, but is perhaps more intuitive.
Create toy data frame...
df <- data.frame(group=rep(letters,3), # create toy data frame
sequence=rep(1:3,26),
link=sample(9:13,78,T)) %>%
arrange(group,sequence) %>% # arrange data
group_by(group,link) %>% sample_n(1) %>% # remove any duplicate link values (to create uneven sequence var)
ungroup() %>% arrange(group,sequence) # arrange again to view
glimpse(df)
Find first and last links. Add them as new columns to the data frame.
df <- df %>% arrange(group,link) %>% group_by(group) %>%
slice(1) %>% mutate(link.first=link) %>% # find first link for each group
select(group,link.first) %>% left_join(df,.) # add to original data frame
df <- df %>% arrange(group,link) %>% group_by(group) %>%
slice(n()) %>% mutate(link.last=link) %>% # find last link for each group
select(group,link.last) %>% left_join(df,.) # add to original data frame
df %>% mutate(key=paste(link.first,link.last,sep=', ')) # paste links to form key
# A tibble: 62 x 6
group sequence link link.first link.last key
<fct> <int> <int> <int> <int> <chr>
1 a 1 10 10 12 10, 12
2 a 2 12 10 12 10, 12
3 b 2 9 9 11 9, 11
4 b 3 11 9 11 9, 11
5 c 1 13 9 13 9, 13
6 c 2 12 9 13 9, 13
7 c 3 9 9 13 9, 13
8 d 1 9 9 13 9, 13
9 d 3 13 9 13 9, 13
10 e 1 11 9 11 9, 11
Since I used sample() with replacement to generate the data, there may be some group's with only one row (i.e., the same first and last link values), which can be filtered out.
df %>% filter(link.first==link.last)
# A tibble: 2 x 5
group sequence link link.first link.last
<fct> <int> <int> <int> <int>
1 k 2 9 9 9
2 z 1 9 9 9
df %>% count(group) %>% filter(n==1)

dplyr collapse 'tail' rows into larger groups

library(tidyverse)
df <- tibble(a = as.factor(1:20), b = c(50, 20, 13, rep(2, 10), rep(1, 7)))
How do I make dplyr look at this data frame df and collapse all these occurences of 2 into a single summed group, and collapse all the occurrences of 1 into a single summed group? And also keep the rest of the data frame.
Turn this:
# A tibble: 20 x 2
a b
<fct> <dbl>
1 1 50
2 2 20
3 3 13
4 4 2
5 5 2
6 6 2
7 7 2
8 8 2
9 9 2
10 10 2
11 11 2
12 12 2
13 13 2
14 14 1
15 15 1
16 16 1
17 17 1
18 18 1
19 19 1
20 20 1
into this:
# A tibble: 5 x 2
a b
<fct> <dbl>
1 1 50
2 2 20
3 3 13
4 grp2 20
5 grp1 7
[Edit] - I fixed the example data. Sorry about that.
We group by a manufactured sortkey to maintain sort order. We used the fact that b is in descending order in the input but if that is not the case in your actual data then replace sortkey = -b with the more general sortkey = data.table::rleid(b) or the longer sortkey = cumsum(coalesce(b != lag(b), FALSE)) .
We also convert b to the group names giving a new a. It wasn't clear which groups are to be converted to grp... form. Hard-coded 1 and 2? Any group with more than one row? Groups at the end with more than one row? At any rate it would be easy enough to change the condition in the if_else once that were clarified.
Finally perform the summation and then remove the sortkey.
df %>%
group_by(sortkey = -b, a = paste0(if_else(b %in% 1:2, "grp", ""), b)) %>%
summarize(b = sum(b)) %>%
ungroup %>%
select(-sortkey)
giving:
# A tibble: 5 x 2
a b
<chr> <int>
1 50 50
2 20 20
3 13 13
4 grp2 20
5 grp1 7
Here's a way. I have converted a from factor to character to make things easier. You can convert it back to factor if you want. Also your test data was a bit wrong.
df <- tibble(a = as.character(1:20), b = c(50, 20, 13, rep(2, 10), rep(1, 7)))
df %>%
mutate(
a = case_when(
b == 1 ~ "grp1",
b == 2 ~ "grp2",
TRUE ~ a
)
) %>%
group_by(a) %>%
summarise(b = sum(b))
# A tibble: 5 x 2
a b
<chr> <dbl>
1 1 50
2 2 20
3 3 13
4 grp1 7
5 grp2 20
This is an approach which gives you the desired names for groups & where you don't need to think in advance how many cases like that you would need (e.g. it would create grp3, grp4, ... depending on the number in b).
library(dplyr)
df %>%
mutate(
grp = as.numeric(lag(df$b) != df$b),
grp = cumsum(ifelse(is.na(grp), 0, grp))
) %>% group_by(grp) %>%
mutate(
a = ifelse(n() > 1, paste0("grp", b), a),
b = sum(b)
) %>% ungroup() %>% distinct(a, b)
Output:
a b
<chr> <dbl>
1 1 50
2 2 20
3 3 13
4 grp2 20
5 grp1 7
Note that the code could be also condensed but that leads to a certain lack of readability in my opinion:
df %>%
group_by(grp = cumsum(ifelse(is.na(as.numeric(lag(df$b) != df$b)), 0, as.numeric(lag(df$b) != df$b)))) %>%
mutate(
a = ifelse(n() > 1, paste0("grp", b), a),
b = sum(b)
) %>% ungroup() %>% distinct(a, b)

R dplyr - select values from one column based on position of a specific value in another column

I am working with gait-cycle data. I have 8 events marked for each id and gait trial. The values "LFCH" and "RFCH" occurs twice in each trial, as these represent the beginning and the end of the gait cycles from left and right leg.
Sample Data Frame:
df <- data.frame(ID = rep(1:5, each = 16),
Gait_nr = rep(1:2, each = 8, times=5),
Frame = rep(c(1,5,7,9,10,15,22,25), times = 10),
Marks = rep(c("LFCH", "LHL", "RFCH", "LTO", "RHL", "LFCH", "RTO", "RFCH"), times =10)
head(df,8)
ID Gait_nr Frame Marks
1 1 1 1 LFCH
2 1 1 5 LHL
3 1 1 7 RFCH
4 1 1 9 LTO
5 1 1 10 RHL
6 1 1 15 LFCH
7 1 1 22 RTO
8 1 1 25 RFCH
I wold like to create something like
Total_gait_left = Frame[The last time Marks == "LFCH"] - Frame[The first time Marks == "LFCH"]
My current code solves the problem, but depends on the position of the Frame values rather than actual values in Marks. Any individual not following the normal gait pattern will have wrong values produced by the code.
library(tidyverse)
l <- df %>% group_by(ID, Gait_nr) %>% filter(grepl("L.+", Marks)) %>%
summarize(Total_gait = Frame[4] - Frame[1],
Side = "left")
r <- df %>% group_by(ID, Gait_nr) %>% filter(grepl("R.+", Marks)) %>%
summarize(Total_gait = Frame[4] - Frame[1],
Side = "right")
val <- union(l,r, by=c("ID", "Gait_nr", "Side")) %>% arrange(ID, Gait_nr, Side)
Can you help me make my code more stable by helping me change e.g. Frame[4] to something like Frame[Marks=="LFCH" the last time ]?
If both LFCH and RFCH happen exactly twice, you can filter and then use diff in summarize:
df %>%
group_by(ID, Gait_nr) %>%
summarise(
left = diff(Frame[Marks == 'LFCH']),
right = diff(Frame[Marks == 'RFCH'])
)
# A tibble: 10 x 4
# Groups: ID [?]
# ID Gait_nr left right
# <int> <int> <dbl> <dbl>
# 1 1 1 14 18
# 2 1 2 14 18
# 3 2 1 14 18
# 4 2 2 14 18
# 5 3 1 14 18
# 6 3 2 14 18
# 7 4 1 14 18
# 8 4 2 14 18
# 9 5 1 14 18
#10 5 2 14 18
We can use first and last from the dplyr package.
library(dplyr)
df2 <- df %>%
filter(Marks %in% "LFCH") %>%
group_by(ID, Gait_nr) %>%
summarise(Total_gait = last(Frame) - first(Frame)) %>%
ungroup()
df2
# # A tibble: 10 x 3
# ID Gait_nr Total_gait
# <int> <int> <dbl>
# 1 1 1 14
# 2 1 2 14
# 3 2 1 14
# 4 2 2 14
# 5 3 1 14
# 6 3 2 14
# 7 4 1 14
# 8 4 2 14
# 9 5 1 14
# 10 5 2 14

Resources