用 ggplot2 - R 绘制数据框

Question

我正在尝试使用以下数据框创建图表。它有 10 个观测值和 25 个变量（其中一个 - ID - 只是不同观测值的 ID 列

'data.frame':    10 obs. of  25 variables:
 $ NDVI_mean   : num  0.0607 0.0552 0.5811 0.7676 0.0328 ...
 $ NDVI_sd     : num  0.0881 0.0298 0.1644 0.0937 0.0292 ...
 $ NDVI_mean.1 : num  0.0211 0.0549 0.1375 0.1207 0.024 ...
 $ NDVI_sd.1   : num  0.0111 0.0195 0.0227 0.0701 0.0197 ...
 $ NDVI_mean.2 : num  0.0703 0.0715 0.6832 0.769 0.0418 ...
 $ NDVI_sd.2   : num  0.0938 0.0298 0.1601 0.0674 0.0402 ...
 $ NDVI_mean.3 : num  0.0636 0.0552 0.6829 0.732 0.0292 ...
 $ NDVI_sd.3   : num  0.0912 0.0222 0.1613 0.1102 0.0355 ...
 $ NDVI_mean.4 : num  0.092 0.0781 0.6947 0.5256 0.056 ...
 $ NDVI_sd.4   : num  0.0879 0.0211 0.158 0.0686 0.0328 ...
 $ NDVI_mean.5 : num  0.1047 0.091 0.4251 0.3573 0.0722 ...
 $ NDVI_sd.5   : num  0.0441 0.013 0.0585 0.0368 0.0156 ...
 $ NDVI_mean.6 : num  0.0547 0.0654 0.5912 0.6098 0.0404 ...
 $ NDVI_sd.6   : num  0.0874 0.0195 0.2143 0.0975 0.0287 ...
 $ NDVI_mean.7 : num  0.1047 0.0882 0.6914 0.6532 0.0689 ...
 $ NDVI_sd.7   : num  0.0843 0.0177 0.1553 0.0653 0.0299 ...
 $ NDVI_mean.8 : num  0.0859 0.071 0.6905 0.6866 0.0556 ...
 $ NDVI_sd.8   : num  0.0809 0.018 0.1624 0.0866 0.0311 ...
 $ NDVI_mean.9 : num  0.0949 0.1204 0.1434 0.2849 0.1231 ...
 $ NDVI_sd.9   : num  0.00951 0.00719 0.01228 0.03483 0.01023 ...
 $ NDVI_mean.10: num  0.0854 0.0752 0.6712 0.7326 0.0628 ...
 $ NDVI_sd.10  : num  0.0789 0.0212 0.1471 0.0951 0.0326 ...
 $ NDVI_mean.11: num  0.0942 0.0986 0.6434 0.7741 0.0899 ...
 $ NDVI_sd.11  : num  0.0735 0.0188 0.1299 0.0765 0.0277 ...
 $ ID          : int  1 2 3 4 5 6 7 8 9 10

Image of the dataframe

我想创建一个具有以下特征的图表：

X 轴应该是不同的 NDVI_mean 变量（NDVI_mean - NDVI_mean.1 - NDVI_mean.2 - 等等）

Y 轴应该是这些变量的值

+

我希望图表包含对应于 10 个观察值的 10 条线

我对 ggplot2 完全陌生。我已经设法使用此代码创建了一些图表，但这不是我想要的

ggplot2(NDVIdf)+geom_lines(aes(x=NDVI_mean, y=ID))

编辑

dput 的输出(NDVIdf)

> dput(NDVIdf)
structure(list(NDVI_mean = c(0.0607135413903215, 0.0551773354119158, 
0.58106114381679, 0.767559372904067, 0.0327779986531678, 0.0178615320775541, 
0.242088217197272, 0.20999285277774, 0.0393074640533382, 0.362654323805063
), NDVI_sd = c(0.0881409040764672, 0.0297587817960566, 0.164350402694459, 
0.0937447350009958, 0.0291673504162979, 0.0328954778667684, 0.154674728930805, 
0.143054126101431, 0.0394783704067313, 0.0311605700206721), NDVI_mean.1 = c(0.0210982854397687, 
0.0549092171312182, 0.137518549485032, 0.120670383289592, 0.0240424367577284, 
0.0159096452554129, 0.0672761385275565, 0.0341803495552938, -0.00134448575083377, 
0.016580205828644), NDVI_sd.1 = c(0.0110723260269658, 0.01951851232517, 
0.0227065359549684, 0.0700670943356275, 0.0196608837546022, 0.0121199724795787, 
0.0585473350749026, 0.0227914450038557, 0.0135058392129466, 0.0150912377421709
), NDVI_mean.2 = c(0.0703180934388137, 0.0714783174472453, 0.683213190781725, 
0.769036956136717, 0.0418112830812162, 0.0284743048998433, 0.23348292946483, 
0.235929665998861, 0.0450296798850473, 0.193100322654342), NDVI_sd.2 = c(0.0937618859311873, 
0.0298498793436821, 0.160085159223464, 0.0673810570033997, 0.0402149180186587, 
0.0397066821267195, 0.150683537602667, 0.145471088294412, 0.0437922991992655, 
0.0141486994532874), NDVI_mean.3 = c(0.063601350404069, 0.0551904437586304, 
0.682930851671194, 0.731967082842268, 0.0291583347580934, 0.0193111448443503, 
0.319631300233593, 0.166050320085929, 0.0366014763086276, 0.221872499210234
), NDVI_sd.3 = c(0.0912293427166813, 0.0222453701937956, 0.161322107465979, 
0.110207844254988, 0.0355049011856384, 0.0349930810428516, 0.226276210965238, 
0.140438890801978, 0.0376830428032925, 0.0220584182188743), NDVI_mean.4 = c(0.0919804842383724, 
0.0781265501499422, 0.694671427954745, 0.525632176936541, 0.055980386796576, 
0.0444207693277835, 0.426953378337129, 0.160783372015251, 0.0609390942283722, 
0.280378773041507), NDVI_sd.4 = c(0.0879414801936151, 0.0210552190044792, 
0.158033594194682, 0.0685669288657517, 0.0327713833848639, 0.0354769286383367, 
0.252469606866754, 0.12982572565032, 0.036836867665617, 0.0411465416161875
), NDVI_mean.5 = c(0.104738543138272, 0.0909713368375085, 0.42508525657118, 
0.357320549164012, 0.0721572385527876, 0.0663794698314188, 0.23911562990616, 
0.142111328436142, 0.0838823297267412, 0.251654432686439), NDVI_sd.5 = c(0.0441048632295888, 
0.0130326877444498, 0.0585279766430101, 0.0368348303398042, 0.0155510862094617, 
0.0192137216652464, 0.0585475549304422, 0.0736886597614494, 0.0212806524351524, 
0.0259938407933158), NDVI_mean.6 = c(0.054670992223019, 0.065381296775636, 
0.591168574495215, 0.609806323819807, 0.0403625648315437, 0.00811946347995523, 
0.310916693477462, 0.158498269033413, 0.0443372830506701, 0.371097291211943
), NDVI_sd.6 = c(0.0874297350407085, 0.0195485598323798, 0.214314782421285, 
0.0974716364190341, 0.0286726835375469, 0.0464844141552075, 0.124111018323546, 
0.128024962302557, 0.0389891785245309, 0.0776972260415738), NDVI_mean.7 = c(0.104681150177595, 
0.0882044567680745, 0.691354972003269, 0.65319672247453, 0.0689026139683328, 
0.0564115948034715, 0.345466555679804, 0.189094867024672, 0.0757218905916805, 
0.473698360613464), NDVI_sd.7 = c(0.0842570243914433, 0.0176701260206802, 
0.15529028462675, 0.0653161775993753, 0.0298643217871498, 0.0350802264835342, 
0.179839784256719, 0.123083052319927, 0.0381348337459403, 0.0504351117371588
), NDVI_mean.8 = c(0.08591308296691, 0.0710169977816541, 0.69050244219096, 
0.686550053201553, 0.05559868259279, 0.0386856009179078, 0.385427380396624, 
0.191494000375466, 0.0582605982908748, 0.634092671211804), NDVI_sd.8 = c(0.0809198942299164, 
0.0179761950231585, 0.162375086927031, 0.0865933396475938, 0.0311069109690036, 
0.0341123644056808, 0.221678551331174, 0.123510636808576, 0.0352156193862569, 
0.037441682380018), NDVI_mean.9 = c(0.0948832729278301, 0.120400127877444, 
0.143375746425582, 0.284877572639076, 0.123096886134381, 0.111171634746743, 
0.223262233590715, 0.120538120679937, 0.0971369124338333, 0.233815280325772
), NDVI_sd.9 = c(0.00951402820331221, 0.00718755312976778, 0.0122787887859583, 
0.0348298462083254, 0.0102326571562652, 0.00707392292527096, 
0.047660749828529, 0.0127667426992843, 0.00615732059786784, 0.0341929453840942
), NDVI_mean.10 = c(0.0853653180560701, 0.0751638478379656, 0.671240397847597, 
0.732629951796317, 0.0628200256114987, 0.0389376153602489, 0.261580310922298, 
0.237551383820751, 0.0543488352606212, 0.729810364384283), NDVI_sd.10 = c(0.0788860954632659, 
0.0212295674726634, 0.147125862744127, 0.0951195938807548, 0.0325819971840338, 
0.0313179762667036, 0.149062631594425, 0.123975319460501, 0.0295479331978356, 
0.0347570926406439), NDVI_mean.11 = c(0.0941656718689444, 0.0985743153705462, 
0.643407964040386, 0.774084527469533, 0.0899420980061257, 0.0535166413991826, 
0.303595683796766, 0.245633779631581, 0.0643377575135575, 0.697236179516483
), NDVI_sd.11 = c(0.0735030069394199, 0.0187835191570716, 0.129850840148202, 
0.0764938134743573, 0.0276954256603995, 0.0260888900038652, 0.122217568202193, 
0.0934074608484564, 0.0272831553843282, 0.026010072370012), ID = 1:10), .Names = c("NDVI_mean", 
"NDVI_sd", "NDVI_mean.1", "NDVI_sd.1", "NDVI_mean.2", "NDVI_sd.2", 
"NDVI_mean.3", "NDVI_sd.3", "NDVI_mean.4", "NDVI_sd.4", "NDVI_mean.5", 
"NDVI_sd.5", "NDVI_mean.6", "NDVI_sd.6", "NDVI_mean.7", "NDVI_sd.7", 
"NDVI_mean.8", "NDVI_sd.8", "NDVI_mean.9", "NDVI_sd.9", "NDVI_mean.10", 
"NDVI_sd.10", "NDVI_mean.11", "NDVI_sd.11", "ID"), row.names = c("1", 
"2", "3", "4", "5", "6", "7", "8", "9", "10"), class = "data.frame")

Answer 1

您的数据是宽格式，但ggplot 要求您先将其转换为长格式。也就是说，每次测量的每个观察值都应该有一行。在你的情况下，你应该有一个 12*2*10 行的数据框，三列是：观察 (1-10)、统计 (mean.1, sd.2, ...) 和值。

您可以使用tidyr 的gather 函数轻松地将数据重新格式化为长格式：

library(tidyr)

NDVIdf_forplot <- gather(NDVIdf, key = statistic, value = value, -ID)
ggplot(NDVIdf_forplot, aes(x = statistic, y = value) + geom_line()

gather 函数采用两个参数：键和值，它们为新的长格式数据帧中的相关字段设置列名。在这种情况下，key 是赋予新列的名称，该名称将之前的列标题（mean、sd、...）和 value 编码为之前的值（0.0607、...）。

表达式-id 告诉gather 不要删除ID 列，因此新数据框中的每条记录仍与正确的ID 相关联。这样您就可以使用它来分隔 ggplot 调用中的不同行，如下所示：

ggplot(NDVIdf_forplot, aes(x = statistic, y= value, group = ID)) + geom_line()

而且，如果您想对不同的观察结果进行不同的着色并包含图例，则可以使用颜色美学：

ggplot(NDVIdf_forplot, aes(x = statistic, y = value, group = ID, colour = ID)) + 
    geom_line()