load("data/kmeans_result.Rdata")
读出变量result,result是聚类结果,四分类。
读出来的result的大概长这样:
Id 1 2 3 4
1 0 0 0 1
2 1 0 0 0
下面的代码把他改成了这样:
Id 类别
1 4
2 1
代码在此:
load("data/kmeans_result.Rdata")
head(result)
# one-hot数据处理为单个的
source_data = result
new_result <- matrix( nrow = nrow(source_data), ncol = 1)
rownames(new_result) <- rownames(source_data)
for(i in c(1:nrow(source_data))){
temp = source_data[i,]
print(sprintf("%dth row: %d",i,(which(temp>0))))
new_result[i,1] = which(temp>0)
}
print(new_result)
colnames(new_result) = c("myCluster")
save(new_result,file="kmeans_result_single.csv.Rdata")
head(new_result)