我正在使用Python的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>
用彩条创建数据的轮廓图。我已经成功完成了无数次,即使使用相同变量的其他层也是如此。但是,当值变小时(约为1E-12),轮廓的某些部分将显示白色。颜色栏也不显示白色。有谁知道这是什么原因以及如何解决?故障轮廓附在下面。
matplotlib.pyplot.contourf
答案 0 :(得分:4)
鉴于新信息,matplotlib无法为您的数据设置正确的级别数(在文档中为see parameters),从而使数据不被绘制。要解决此问题,您需要告诉matplotlib使用plt.contourf(..., extend="max")
或plt.contourf(..., extend="both")
contourf()
显示带有colormap that doesn't include white的白色区域有几个原因。
从不绘制NaN值。
如果在绘图之前屏蔽数据,则数据不会出现在绘图中。但是您应该知道是否屏蔽了数据。
尽管如此,如果您使用Tick locator = LogLocator()之类的东西,则可能会掩盖您的数据。
有时matplotlib设置的级别不正确,从而使您的某些数据无法绘制。
要解决此问题,您可以使用plt.contourf(..., extend=EXTENDS)
,其中EXTENDS可以是"neither", "both", "min", "max"
contourf plots whitespace over finite data. Past answers do not correct
答案 1 :(得分:0)
如果X和Y向量数据点的间距不相等,则图中的白色部分也会出现。在那种情况下,最好使用函数tricontourf()。