问题
I have a below-mentioned dataframe:
structure(
list(ID = c("P-1", " P-1", "P-1", "P-2", "P-3", "P-4", "P-5", "P-6", "P-7",
"P-8"),
Date = c("2020-03-16 12:11:33", "2020-03-16 13:16:04",
"2020-03-16 06:13:55", "2020-03-16 10:03:43",
"2020-03-16 12:37:09", "2020-03-16 06:40:24",
"2020-03-16 09:46:45", "2020-03-16 12:07:44",
"2020-03-16 14:09:51", "2020-03-16 09:19:23"),
Status = c("SA", "SA", "SA", "RE", "RE", "RE", "RE", "XA", "XA", "XA"),
Flag = c("L", "L", "L", NA, "K", "J", NA, NA, "H", "G"),
Value = c(5929.81, 5929.81, 5929.81, NA, 6969.33, 740.08, NA, NA, 1524.8,
NA),
Flag2 = c("CL", "CL", "CL", NA, "RY", "", NA, NA, "", NA),
Flag3 = c(NA, NA, NA, NA, "RI", "PO", NA, "SS", "DDP", NA)),
.Names=c("ID", "Date", "Status", "Flag", "Value", "Flag2", "Flag3"),
row.names=c(NA, 10L), class="data.frame")
I am using below-mentioned code:
df %>% mutate(L = ifelse(Flag == "L",1,0),
K = ifelse(Flag == "K",1,0),
# etc for Flag) %>%
mutate(sub_status = NA) %>%
mutate(sub_status = ifelse(!is.na(Flag2) & Flag3 == 0, "a", sub_status),
sub_status = ifelse(is.na(Flag2) & Flag3 != 0, "b", sub_status),
# etc for sub-status) %>%
mutate(value_class = ifelse(0 <= Value & Value <= 15000, "0-15000",
"15000-50000")) %>%
group_by(Date, status, sub_status, value_class) %>%
summarise(L = sum(L),
K = sum(K),
# etc
count = n())
Which provides me the following output:
Date Status sub_status value_class G H I J K L NA Count
2020-03-20 SA a 0-15000 0 0 0 0 1 1 0 2
2020-03-20 SA b 0-15000 0 0 0 0 1 0 0 1
................
................
I want to get the following output using the DF, where the Status column has distinct 3 values and Flag2 has either values or [null] or NA and finally Flag3 column has distinct 7 values with [null] or NA. For one distinct ID we have multiple entry of Flag3 column.
I Need to create the following dataframe, by creating a 3 group based on Value like 0-15000, 15000-50000.
- If for a distinct ID
Flag2has some value other than 0 or [null]/NA butFlag3has value 0 or [null]/NA then it would bea. - If for a distinct ID
Flag3has some value other than 0 or [null]/NA butFlag2has value 0 or [null]/NA then it would beb - If for a distinct ID both
Flag2&Flag3has some value other than 0 or [Null]/NA then it would bec - If for a distinct ID both
Flag2&Flag3has value 0 or [Null]/NA the it would bed
I want to arrange the above mentioned datafrmae in the following structure with percent and Total column.
I have mentioned the percentage like 2/5 to show that status would be divided by the Total whereas sub_status would be divided by their respective Status.
16/03/2020 0 - 15000 15000 - 50000
Status count percent L K J H G [Null] count percent L K J H G [Null] Total
SA 1 1/8 (12.50%) 1 0 0 0 0 0 0 - 0 0 0 0 0 0 1
a 1 1/1(100.00%) 1 0 0 0 0 0 0 - 0 0 0 0 0 0 1
b 0 - 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0
c 0 - 1 0 0 0 0 0 0 - 0 0 0 0 0 0 0
d 0 - 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0
RE 4 50.00% 0 1 1 0 0 2 0 - 0 0 0 0 0 0 4
a 0 - 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0
b 1 25.00% 0 0 1 0 0 1 0 - 0 0 0 0 0 0 1
c 1 25.00% 0 1 0 0 0 1 0 - 0 0 0 0 0 0 1
d 2 50.00% 0 0 0 0 0 2 0 - 0 0 0 0 0 0 2
XA 3 37.50% 0 0 0 1 1 1 0 - 0 0 0 0 0 0 3
a 0 - 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0
b 2 66.67% 0 0 0 1 0 1 0 - 0 0 0 0 0 0 2
c 0 - 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0
d 1 33.33% 0 0 0 0 1 0 0 - 0 0 0 0 0 0 1
Total 8 100.00% 1 1 0 0 1 3 0 - 0 0 0 0 0 0 8
I have mentioned the required output based on the latest date which is 16/03/2020, if the dataframe doesn't have the latest date as per startdate keep all the value 0 in the output dataframe. The percentage column is just for the reference there will be calculated percentage values,.
Also, I want to keep the structure static. For Example, if for any of the parameter are not present for a day the output structure would be same with 0 value.
For Example, Suppose date 17/03/2020 don't have any row with status SA or sub_status c the place holder for that will be there in the output with value as 0.
回答1:
Hopefully that'll be enough to get you started, to go further, I'll need an expected output that looks like it comes from R, and further explanations as to how variables are computed.
library(tidyverse)
df <- structure(
list(ID = c("P-1", " P-1", "P-1", "P-2", "P-3", "P-4", "P-5", "P-6", "P-7",
"P-8"),
Date = c("2020-03-16 12:11:33", "2020-03-16 13:16:04",
"2020-03-16 06:13:55", "2020-03-16 10:03:43",
"2020-03-16 12:37:09", "2020-03-16 06:40:24",
"2020-03-16 09:46:45", "2020-03-16 12:07:44",
"2020-03-16 14:09:51", "2020-03-16 09:19:23"),
Status = c("SA", "SA", "SA", "RE", "RE", "RE", "RE", "XA", "XA", "XA"),
Flag = c("L", "L", "L", NA, "K", "J", NA, NA, "H", "G"),
Value = c(5929.81, 5929.81, 5929.81, NA, 6969.33, 740.08, NA, NA, 1524.8,
NA),
Flag2 = c("CL", "CL", "CL", NA, "RY", "", NA, NA, "", NA),
Flag3 = c(NA, NA, NA, NA, "RI", "PO", NA, "SS", "DDP", NA)),
.Names=c("ID", "Date", "Status", "Flag", "Value", "Flag2", "Flag3"),
row.names=c(NA, 10L), class="data.frame")
df2 <- df %>%
mutate(
# add variables
Value = ifelse(0 <= Value & Value <= 15000, "0-15000", "15000-50000"),
substatus = case_when(
!is.na(Flag2) & is.na(Flag3) ~ "a",
!is.na(Flag3) & is.na(Flag2) ~ "b",
!is.na(Flag3) & !is.na(Flag2) ~ "c",
TRUE ~ "d"),
# make Date an actual date rather than a timestamp
Date = as.Date(Date),
# remove obsolete columns
Flag2 = NULL,
Flag3 = NULL,
ID = NULL,
# renames NAs into the name of the desired column
Flag = ifelse(is.na(Flag), "[Null]", Flag),
# create column of 1 for pivot
temp = 1,
# and row id
id = row_number()
) %>%
# create new columns L K etc, this also drops the Flag col
pivot_wider(names_from = "Flag", values_from = "temp", values_fill = list(temp=0)) %>%
# move `[Null]` column to the end
select(everything(), -`[Null]`, `[Null]`) %>%
mutate(
id = NULL,
count = 1,
Total = rowSums(select(., L:`[Null]`)))
df2
#> # A tibble: 10 x 12
#> Date Status Value substatus L K J H G `[Null]`
#> <date> <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 2020-03-16 SA 0-15~ a 1 0 0 0 0 0
#> 2 2020-03-16 SA 0-15~ a 1 0 0 0 0 0
#> 3 2020-03-16 SA 0-15~ a 1 0 0 0 0 0
#> 4 2020-03-16 RE <NA> d 0 0 0 0 0 1
#> 5 2020-03-16 RE 0-15~ c 0 1 0 0 0 0
#> 6 2020-03-16 RE 0-15~ c 0 0 1 0 0 0
#> 7 2020-03-16 RE <NA> d 0 0 0 0 0 1
#> 8 2020-03-16 XA <NA> b 0 0 0 0 0 1
#> 9 2020-03-16 XA 0-15~ c 0 0 0 1 0 0
#> 10 2020-03-16 XA <NA> d 0 0 0 0 1 0
#> # ... with 2 more variables: count <dbl>, Total <dbl>
# As you didn't tell what to do with NA values so I left them as NA
bind_rows(
df2 %>%
# add missing combinations of abcd
complete(nesting(Date, Status, Value), substatus) %>%
group_by(Date, Value, Status, substatus) %>%
summarize_all(~sum(., na.rm=TRUE)) %>%
group_by(Status, Value) %>%
mutate(percent = paste(round(100 * Total / sum(Total), 2), "%")) %>%
ungroup(),
df2 %>%
mutate(substatus = Status, Status = paste0(Status, "_")) %>%
group_by(Date, Value, Status, substatus) %>%
mutate(count = n()) %>%
group_by(count, add = TRUE) %>%
summarize_all(~sum(., na.rm=TRUE)) %>%
group_by(Value) %>%
mutate(percent = paste(round(100 * Total / sum(Total), 2), "%"))
) %>%
arrange(Date, Value, desc(Status)) %>%
mutate(Status = NULL) %>%
rename(Status = substatus) %>%
print(n=Inf)
#> # A tibble: 25 x 12
#> Date Value Status L K J H G `[Null]` count Total
#> <date> <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 2020-03-16 0-15~ XA 0 0 0 1 0 0 1 1
#> 2 2020-03-16 0-15~ a 0 0 0 0 0 0 0 0
#> 3 2020-03-16 0-15~ b 0 0 0 0 0 0 0 0
#> 4 2020-03-16 0-15~ c 0 0 0 1 0 0 1 1
#> 5 2020-03-16 0-15~ d 0 0 0 0 0 0 0 0
#> 6 2020-03-16 0-15~ SA 3 0 0 0 0 0 3 3
#> 7 2020-03-16 0-15~ a 3 0 0 0 0 0 3 3
#> 8 2020-03-16 0-15~ b 0 0 0 0 0 0 0 0
#> 9 2020-03-16 0-15~ c 0 0 0 0 0 0 0 0
#> 10 2020-03-16 0-15~ d 0 0 0 0 0 0 0 0
#> 11 2020-03-16 0-15~ RE 0 1 1 0 0 0 2 2
#> 12 2020-03-16 0-15~ a 0 0 0 0 0 0 0 0
#> 13 2020-03-16 0-15~ b 0 0 0 0 0 0 0 0
#> 14 2020-03-16 0-15~ c 0 1 1 0 0 0 2 2
#> 15 2020-03-16 0-15~ d 0 0 0 0 0 0 0 0
#> 16 2020-03-16 <NA> XA 0 0 0 0 1 1 2 2
#> 17 2020-03-16 <NA> a 0 0 0 0 0 0 0 0
#> 18 2020-03-16 <NA> b 0 0 0 0 0 1 1 1
#> 19 2020-03-16 <NA> c 0 0 0 0 0 0 0 0
#> 20 2020-03-16 <NA> d 0 0 0 0 1 0 1 1
#> 21 2020-03-16 <NA> RE 0 0 0 0 0 2 2 2
#> 22 2020-03-16 <NA> a 0 0 0 0 0 0 0 0
#> 23 2020-03-16 <NA> b 0 0 0 0 0 0 0 0
#> 24 2020-03-16 <NA> c 0 0 0 0 0 0 0 0
#> 25 2020-03-16 <NA> d 0 0 0 0 0 2 2 2
#> # ... with 1 more variable: percent <chr>
来源：https://stackoverflow.com/questions/61145404/how-to-pivoting-dataframe-consisting-column-with-section-and-sub-section-in-r