我有一个df，它有3个分层分组级别和4种类型的成本。对于项目，组是一对多，并且使用版本对项目进行一对多项目。 costs2.1a看起来像这样（~500krows）：
+-------+---------+----------+----------+---------+---------+---------+
| group | project | versions | amount1 | amount2 | amount3 | amount4 |
+-------+---------+----------+----------+---------+---------+---------+
| 1316 | 142 | 0 | 370662 | 370662 | 288264 | 4688 |
| 1316 | 142 | 1 | 174242 | 174242 | 134837 | 0 |
| 1316 | 142 | 1 | 174242 | 174242 | 134837 | 159242 |
| 1316 | 142 | 2 | -246912 | -246912 | -191073 | 0 |
| 1316 | 142 | 2 | -246912 | -246912 | -191073 | 0 |
| 1316 | 142 | 2 | -246912 | -246912 | -191073 | 60591 |
| 1316 | 142 | 2 | -246912 | -246912 | -191073 | 5000 |
| 1316 | 142 | 3 | -32437 | -32437 | -25101 | 0 |
| 1316 | 142 | 3 | -32437 | -32437 | -25101 | 0 |
| 1316 | 142 | 3 | -32437 | -32437 | -25101 | 0 |
| 1316 | 142 | 3 | -32437 | -32437 | -25101 | 0 |
| 1316 | 143 | 0 | 620515 | 620515 | 480186 | 411400 |
| 1316 | 143 | 1 | -31113 | -31113 | -24077 | 0 |
| 1316 | 151 | 0 | 515269 | 515269 | 398982 | 510 |
| 1316 | 151 | 1 | 85380 | 85380 | 65213 | 85380 |
+-------+---------+----------+----------+---------+---------+---------+
我想先group by，'group'和'project'变量'versions'。然后我想删除'amount 1' : 'amount 4'上的重复项。然后，我想返回并删除'amount 1' : 'amount 3'（仍然持有分组）if 'amount 4' = 0的重复项。
我试图通过几种方式找出答案，包括使用if循环，但最接近答案的是这段代码：
library(dplyr)
costs2.1a <- cbind(mit2.1filter[ , 2:4], costsSub) %>%
group_by(DISASTER_NUMBER, PW_NUMBER, VERSION_NUMBER) %>%
distinct(PROJECT_AMOUNT, TOTAL_ELIGIBLE, TOTAL_OBLIGATED, MITIGATION_COST, .keep_all = TRUE)
costs2.1b <- costs2.1a[which(distinct(costs2.1a, PROJECT_AMOUNT, TOTAL_ELIGIBLE, TOTAL_OBLIGATED, .keep_all = TRUE) & costs2.1a$MITIGATION_COST != 0), ]
我知道它不是我想要的，因为costs2.1a有173,871行，costs2.1b有366,968行;这没有意义。
如果有人知道我在这段代码中做错了什么，或者以其他方式做我想做的事，那将非常感谢帮助！
修改
Heres一些输入数据，谢谢Jazzaro。
> dput(head(cbind(mit2.1filter[ , 2:4], costsSub), 30))
structure(list(DISASTER_NUMBER = c(1301L, 1301L, 1301L, 1301L,
1301L, 1301L, 1301L, 1301L, 1302L, 1302L, 1302L, 1302L, 1302L,
1302L, 1302L, 1302L, 1302L, 1302L, 1303L, 1303L, 1303L, 1305L,
1305L, 1305L, 1306L, 1306L, 1306L, 1306L, 1306L, 1306L), PW_NUMBER = c(6L,
35L, 70L, 71L, 83L, 121L, 121L, 125L, 9L, 37L, 37L, 58L, 60L,
62L, 65L, 124L, 124L, 124L, 184L, 184L, 184L, 10L, 28L, 29L,
1235L, 1235L, 1349L, 1349L, 1349L, 1349L), VERSION_NUMBER = c(0L,
0L, 0L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 0L, 0L, 1L,
2L, 0L, 1L, 2L, 0L, 0L, 0L, 0L, 1L, 0L, 1L, 2L, 2L), PROJECT_AMOUNT = c(6787,
16402, 33843, 28169, 12148, 217010, -151101, 13786, 1625, 60260,
79755, 39894, 11352, 7521, 2950, 82010, 20316, 2646, 243190,
-111638, -14021, 3364, 11421, 7534, 184012, -184012, 229357,
0, -65000, -65000), TOTAL_ELIGIBLE = c(6787, 16402, 33843, 28169,
12148, 217010, -151101, 13786, 1625, 60260, 79755, 39894, 11352,
7521, 2950, 82010, 20316, 2646, 243190, -111638, -14021, 3364,
11421, 7534, 184012, -184012, 229357, 0, -65000, -65000), TOTAL_OBLIGATED = c(5347,
12921, 26320, 21907, 9745, 168768, -117512, 10721, 1306, 46878,
62026, 31740, 9032, 5984, 2347, 64607, 15855, 2058, 189129, -86391,
-10850, 2703, 9087, 5994, 139624, -139624, 174030, 0, -49320,
-49320), MITIGATION_COST = c(1780, 1575, 1500, 6600, 0, 3885,
0, 6230, 0, 0, 0, 10222, 1410, 1821, 528, 0, 0, 0, 0, 27187,
0, 1050, 3280, 204, 87984, 0, 65000, 0, -65000, 0)), .Names = c("DISASTER_NUMBER",
"PW_NUMBER", "VERSION_NUMBER", "PROJECT_AMOUNT", "TOTAL_ELIGIBLE",
"TOTAL_OBLIGATED", "MITIGATION_COST"), row.names = c(NA, 30L), class = "data.frame")
> dput(head(costs2.1a, 30)) ##this is output data!!
structure(list(DISASTER_NUMBER = c(1301L, 1301L, 1301L, 1301L,
1301L, 1301L, 1301L, 1301L, 1302L, 1302L, 1302L, 1302L, 1302L,
1302L, 1302L, 1302L, 1302L, 1302L, 1303L, 1303L, 1303L, 1305L,
1305L, 1305L, 1306L, 1306L, 1306L, 1306L, 1306L, 1306L), PW_NUMBER = c(6L,
35L, 70L, 71L, 83L, 121L, 121L, 125L, 9L, 37L, 37L, 58L, 60L,
62L, 65L, 124L, 124L, 124L, 184L, 184L, 184L, 10L, 28L, 29L,
1235L, 1235L, 1349L, 1349L, 1349L, 1349L), VERSION_NUMBER = c(0L,
0L, 0L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 0L, 0L, 1L,
2L, 0L, 1L, 2L, 0L, 0L, 0L, 0L, 1L, 0L, 1L, 2L, 2L), PROJECT_AMOUNT = c(6787,
16402, 33843, 28169, 12148, 217010, -151101, 13786, 1625, 60260,
79755, 39894, 11352, 7521, 2950, 82010, 20316, 2646, 243190,
-111638, -14021, 3364, 11421, 7534, 184012, -184012, 229357,
0, -65000, -65000), TOTAL_ELIGIBLE = c(6787, 16402, 33843, 28169,
12148, 217010, -151101, 13786, 1625, 60260, 79755, 39894, 11352,
7521, 2950, 82010, 20316, 2646, 243190, -111638, -14021, 3364,
11421, 7534, 184012, -184012, 229357, 0, -65000, -65000), TOTAL_OBLIGATED = c(5347,
12921, 26320, 21907, 9745, 168768, -117512, 10721, 1306, 46878,
62026, 31740, 9032, 5984, 2347, 64607, 15855, 2058, 189129, -86391,
-10850, 2703, 9087, 5994, 139624, -139624, 174030, 0, -49320,
-49320), MITIGATION_COST = c(1780, 1575, 1500, 6600, 0, 3885,
0, 6230, 0, 0, 0, 10222, 1410, 1821, 528, 0, 0, 0, 0, 27187,
0, 1050, 3280, 204, 87984, 0, 65000, 0, -65000, 0)), .Names = c("DISASTER_NUMBER",
"PW_NUMBER", "VERSION_NUMBER", "PROJECT_AMOUNT", "TOTAL_ELIGIBLE",
"TOTAL_OBLIGATED", "MITIGATION_COST"), row.names = c(NA, -30L
), class = c("grouped_df", "tbl_df", "tbl", "data.frame"), vars = list(
DISASTER_NUMBER, PW_NUMBER, VERSION_NUMBER), drop = TRUE, indices = list(
0L, 1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 12L, 13L,
14L, 15L, 16L, 17L, 18L, 19L, 20L, 21L, 22L, 23L, 24L, 25L,
26L, 27L, 28:29), group_sizes = c(1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 2L), biggest_group_size = 2L, labels = structure(list(
DISASTER_NUMBER = c(1301L, 1301L, 1301L, 1301L, 1301L, 1301L,
1301L, 1301L, 1302L, 1302L, 1302L, 1302L, 1302L, 1302L, 1302L,
1302L, 1302L, 1302L, 1303L, 1303L, 1303L, 1305L, 1305L, 1305L,
1306L, 1306L, 1306L, 1306L, 1306L), PW_NUMBER = c(6L, 35L,
70L, 71L, 83L, 121L, 121L, 125L, 9L, 37L, 37L, 58L, 60L,
62L, 65L, 124L, 124L, 124L, 184L, 184L, 184L, 10L, 28L, 29L,
1235L, 1235L, 1349L, 1349L, 1349L), VERSION_NUMBER = c(0L,
0L, 0L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 0L, 0L,
1L, 2L, 0L, 1L, 2L, 0L, 0L, 0L, 0L, 1L, 0L, 1L, 2L)), row.names = c(NA,
-29L), class = "data.frame", vars = list(DISASTER_NUMBER, PW_NUMBER,
VERSION_NUMBER), drop = TRUE, .Names = c("DISASTER_NUMBER",
"PW_NUMBER", "VERSION_NUMBER")))
> dput(droplevels(head(costs2.1a, 30))) ##this is output data!!
structure(list(DISASTER_NUMBER = c(1301L, 1301L, 1301L, 1301L,
1301L, 1301L, 1301L, 1301L, 1302L, 1302L, 1302L, 1302L, 1302L,
1302L, 1302L, 1302L, 1302L, 1302L, 1303L, 1303L, 1303L, 1305L,
1305L, 1305L, 1306L, 1306L, 1306L, 1306L, 1306L, 1306L), PW_NUMBER = c(6L,
35L, 70L, 71L, 83L, 121L, 121L, 125L, 9L, 37L, 37L, 58L, 60L,
62L, 65L, 124L, 124L, 124L, 184L, 184L, 184L, 10L, 28L, 29L,
1235L, 1235L, 1349L, 1349L, 1349L, 1349L), VERSION_NUMBER = c(0L,
0L, 0L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 0L, 0L, 1L,
2L, 0L, 1L, 2L, 0L, 0L, 0L, 0L, 1L, 0L, 1L, 2L, 2L), PROJECT_AMOUNT = c(6787,
16402, 33843, 28169, 12148, 217010, -151101, 13786, 1625, 60260,
79755, 39894, 11352, 7521, 2950, 82010, 20316, 2646, 243190,
-111638, -14021, 3364, 11421, 7534, 184012, -184012, 229357,
0, -65000, -65000), TOTAL_ELIGIBLE = c(6787, 16402, 33843, 28169,
12148, 217010, -151101, 13786, 1625, 60260, 79755, 39894, 11352,
7521, 2950, 82010, 20316, 2646, 243190, -111638, -14021, 3364,
11421, 7534, 184012, -184012, 229357, 0, -65000, -65000), TOTAL_OBLIGATED = c(5347,
12921, 26320, 21907, 9745, 168768, -117512, 10721, 1306, 46878,
62026, 31740, 9032, 5984, 2347, 64607, 15855, 2058, 189129, -86391,
-10850, 2703, 9087, 5994, 139624, -139624, 174030, 0, -49320,
-49320), MITIGATION_COST = c(1780, 1575, 1500, 6600, 0, 3885,
0, 6230, 0, 0, 0, 10222, 1410, 1821, 528, 0, 0, 0, 0, 27187,
0, 1050, 3280, 204, 87984, 0, 65000, 0, -65000, 0)), .Names = c("DISASTER_NUMBER",
"PW_NUMBER", "VERSION_NUMBER", "PROJECT_AMOUNT", "TOTAL_ELIGIBLE",
"TOTAL_OBLIGATED", "MITIGATION_COST"), row.names = c(NA, -30L
), vars = list(DISASTER_NUMBER, PW_NUMBER, VERSION_NUMBER), drop = TRUE, indices = list(
0L, 1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 12L, 13L,
14L, 15L, 16L, 17L, 18L, 19L, 20L, 21L, 22L, 23L, 24L, 25L,
26L, 27L, 28:29), group_sizes = c(1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 2L), biggest_group_size = 2L, labels = structure(list(
DISASTER_NUMBER = c(1301L, 1301L, 1301L, 1301L, 1301L, 1301L,
1301L, 1301L, 1302L, 1302L, 1302L, 1302L, 1302L, 1302L, 1302L,
1302L, 1302L, 1302L, 1303L, 1303L, 1303L, 1305L, 1305L, 1305L,
1306L, 1306L, 1306L, 1306L, 1306L), PW_NUMBER = c(6L, 35L,
70L, 71L, 83L, 121L, 121L, 125L, 9L, 37L, 37L, 58L, 60L,
62L, 65L, 124L, 124L, 124L, 184L, 184L, 184L, 10L, 28L, 29L,
1235L, 1235L, 1349L, 1349L, 1349L), VERSION_NUMBER = c(0L,
0L, 0L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 0L, 0L,
1L, 2L, 0L, 1L, 2L, 0L, 0L, 0L, 0L, 1L, 0L, 1L, 2L)), row.names = c(NA,
-29L), class = "data.frame", vars = list(DISASTER_NUMBER, PW_NUMBER,
VERSION_NUMBER), drop = TRUE, .Names = c("DISASTER_NUMBER",
"PW_NUMBER", "VERSION_NUMBER")), class = c("grouped_df", "tbl_df",
"tbl", "data.frame"))
答案 0 :(得分：0)
第一步相对简单，我们使用gather重新塑造，使用distinct仅保留唯一值，然后重新塑造。请注意，我们完全丢失了amount2，因为示例数据中没有值是唯一的。
d <- costs2.1a %>%
gather('key', 'value', amount1:amount4) %>%
distinct(group, project, versions, value, .keep_all = TRUE) %>%
spread(key, value)
第二部分更难，因为我们无法有条件地使用distinct。所以我们再次重新整形（d2），然后仅对子集（distinct）使用d3，对其余数据（d4）不执行任何操作，然后绑定然后一起重新塑造。
d2 <- d %>%
gather('key', 'value', amount1:amount3)
d3 <- d2 %>%
filter(amount4 == 0) %>%
distinct(group, project, versions, value, .keep_all = TRUE)
d4 <- d2 %>% filter(amount4 != 0)
bind_rows(d3, d4) %>%
spread(key, value)
costs2.1a <- as.data.frame(structure(list(DISASTER_NUMBER = c(1301L, 1301L, 1301L, 1301L,
1301L, 1301L, 1301L, 1301L, 1302L, 1302L, 1302L, 1302L, 1302L,
1302L, 1302L, 1302L, 1302L, 1302L, 1303L, 1303L, 1303L, 1305L,
1305L, 1305L, 1306L, 1306L, 1306L, 1306L, 1306L, 1306L),
PW_NUMBER = c(6L,
35L, 70L, 71L, 83L, 121L, 121L, 125L, 9L, 37L, 37L, 58L, 60L,
62L, 65L, 124L, 124L, 124L, 184L, 184L, 184L, 10L, 28L, 29L,
1235L, 1235L, 1349L, 1349L, 1349L, 1349L),
VERSION_NUMBER = c(0L,
0L, 0L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 1L, 0L, 0L, 0L, 0L, 0L, 1L,
2L, 0L, 1L, 2L, 0L, 0L, 0L, 0L, 1L, 0L, 1L, 2L, 2L),
PROJECT_AMOUNT = c(6787,
16402, 33843, 28169, 12148, 217010, -151101, 13786, 1625, 60260,
79755, 39894, 11352, 7521, 2950, 82010, 20316, 2646, 243190,
-111638, -14021, 3364, 11421, 7534, 184012, -184012, 229357,
0, -65000, -65000),
TOTAL_ELIGIBLE = c(6787, 16402, 33843, 28169,
12148, 217010, -151101, 13786, 1625, 60260, 79755, 39894, 11352,
7521, 2950, 82010, 20316, 2646, 243190, -111638, -14021, 3364,
11421, 7534, 184012, -184012, 229357, 0, -65000, -65000),
TOTAL_OBLIGATED = c(5347,
12921, 26320, 21907, 9745, 168768, -117512, 10721, 1306, 46878,
62026, 31740, 9032, 5984, 2347, 64607, 15855, 2058, 189129, -86391,
-10850, 2703, 9087, 5994, 139624, -139624, 174030, 0, -49320,
-49320),
MITIGATION_COST = c(1780, 1575, 1500, 6600, 0, 3885,
0, 6230, 0, 0, 0, 10222, 1410, 1821, 528, 0, 0, 0, 0, 27187,
0, 1050, 3280, 204, 87984, 0, 65000, 0, -65000, 0)),
.Names = c("DISASTER_NUMBER",
"PW_NUMBER", "VERSION_NUMBER", "PROJECT_AMOUNT", "TOTAL_ELIGIBLE",
"TOTAL_OBLIGATED", "MITIGATION_COST"), row.names = c(NA, -30L
)))
names(costs2.1a) <- c('group', 'project', 'versions', 'amount1', 'amount2', 'amount3', 'amount4')