我在R中运行k-means聚类，并希望使用NbClust来帮助确定最佳聚类数。我的数据集df有636,688行和7列。
当我运行NbClust(df, min.nc = 2, max.nc = 3, method = "kmeans")时，我得到：
Error: cannot allocate vector of size 1510.1 Gb
In addition: Warning messages:
1: In dist(jeu, method = "euclidean") :
Reached total allocation of 32767Mb: see help(memory.size)
2: In dist(jeu, method = "euclidean") :
Reached total allocation of 32767Mb: see help(memory.size)
3: In dist(jeu, method = "euclidean") :
Reached total allocation of 32767Mb: see help(memory.size)
4: In dist(jeu, method = "euclidean") :
Reached total allocation of 32767Mb: see help(memory.size)
这是我的sessionInfo：
R version 3.0.2 (2013-09-25)
Platform: x86_64-w64-mingw32/x64 (64-bit)
locale:
[1] LC_COLLATE=English_United Kingdom.1252 LC_CTYPE=English_United Kingdom.1252 LC_MONETARY=English_United Kingdom.1252
[4] LC_NUMERIC=C LC_TIME=English_United Kingdom.1252
attached base packages:
[1] stats4 grid stats graphics grDevices utils datasets methods base
other attached packages:
[1] clusterSim_0.43-3 fpc_2.1-6 flexmix_2.3-11 mclust_4.2 cluster_1.14.4 MASS_7.3-29
[7] flexclust_1.3-4 modeltools_0.2-21 lattice_0.20-23 NbClust_1.4 rattle_2.6.26
loaded via a namespace (and not attached):
[1] ade4_1.6-2 class_7.3-9 e1071_1.6-2 nnet_7.3-7 parallel_3.0.2 R2HTML_2.2.1 rgl_0.93.996 tools_3.0.2
有没有办法绕过尺寸限制？我知道我可以对样本进行采样并运行NbClust，但我发现我的样本必须是完整数据集的1％才能使NbClust能够运行 - 样本I相信它太小而无法提供帮助，即使我经常进行多次迭代。即使是小样本 - NbClust也很慢。我查看了cluster.stats包中的fpc，但这需要一个距离矩阵 - 这对我和我的大数据集都是不可能的。