R 对15组进行迭代计算

我正在为我的一些数据做一些QA/QC工作，我有兴趣查看我的数据在每个组中被标记为可能更容易出错的百分比。我有15个小组，我想重复这一点，但我不知道如何最好地做到这一点。也许是一个循环

# pulling out group 1 data
group_1 <- filter(flow_group_df, GROUP == 1) 

#looking at number of flagged occurrences in group 1
group_1_flagged <- length(which(group_1 == "flagged")) 

#total number of checked occurrences that havent been flagged
class_1_checked <- length(which(group_1 == "checked"))

#拉出第1组数据
第1组考虑R的内置
stats
库，使用
aggregate
进行完全聚合：

agg_df <- aggregate(num_col ~ GROUP + group_1, flow_group_df, length)


要用随机数据演示：

set.seed(72318)
flow_group_df <- data.frame(GROUP = c("julia", "r", "pandas"),
                            group_1 = sample(c("flagged", "checked"), 60, replace=TRUE),
                            num_col = runif(60, 0, 100))

set.seed（72318）
flow\u group\u df只需按操作分组
flow\u group\u df%>%group\u by（group，group\u 1）%>%summary（n=n（））
如果
group\u 1
中有更多的组，请执行
过滤器
flow\u group\u df%>%filter（group\u 1%在%c中（'flagged'，'checked'））%%>%group\u by（group，group\u 1）%>%summary（n=n（））
或只需
表格（df[c（“集团”、“集团1”）]
set.seed(72318)
flow_group_df <- data.frame(GROUP = c("julia", "r", "pandas"),
                            group_1 = sample(c("flagged", "checked"), 60, replace=TRUE),
                            num_col = runif(60, 0, 100))

agg_df <- aggregate(num_col ~ GROUP + group_1, flow_group_df, length)    
agg_df <- with(agg_df, agg_df[order(GROUP, group_1),])                      # ORDER BY GROUPS
row.names(agg_df) <- NULL                                                   # RESET ROW NAMES
colnames(agg_df)[3] <- "count"                                              # RENAME KEY COL

agg_df         
#    GROUP group_1   count
# 1  julia checked      10
# 2  julia flagged      10
# 3 pandas checked       8
# 4 pandas flagged      12
# 5      r checked       7
# 6      r flagged      13

flow_group_df$group_1_count <- with(flow_group_df, ave(num_col, GROUP, group_1, FUN=length))
flow_group_df$group_1_pct <- with(flow_group_df, ave(num_col, GROUP, group_1, FUN=length)) /
                                  with(flow_group_df, ave(num_col, GROUP, FUN=length))

flow_group_df <- with(flow_group_df, flow_group_df[order(GROUP, group_1),])    # ORDER BY GROUPS
row.names(flow_group_df) <- NULL                                               # RESET ROW NAMES

tail(flow_group_df, 20)
#    GROUP group_1   num_col group_1_count group_1_pct
# 41     r checked  8.128056             7        0.35
# 42     r checked 86.439911             7        0.35
# 43     r checked 75.488474             7        0.35
# 44     r checked 88.120510             7        0.35
# 45     r checked 43.058268             7        0.35
# 46     r checked 46.662674             7        0.35
# 47     r checked 42.329505             7        0.35
# 48     r flagged 94.959380            13        0.65
# 49     r flagged 64.817015            13        0.65
# 50     r flagged 61.118952            13        0.65
# 51     r flagged 69.104977            13        0.65
# 52     r flagged 98.078729            13        0.65
# 53     r flagged 74.857959            13        0.65
# 54     r flagged 83.813440            13        0.65
# 55     r flagged 99.069011            13        0.65
# 56     r flagged 62.298414            13        0.65
# 57     r flagged 14.335920            13        0.65
# 58     r flagged 70.404048            13        0.65
# 59     r flagged 18.744892            13        0.65
# 60     r flagged 21.598072            13        0.65