机器学习中按Treatment与Bio分组拆分训练/测试/验证集的方法

需要将包含Control和TP两个Treatment组的数据集拆分为训练集、测试集和验证集，每组下有Bio 1-4的子分组。拆分规则为：若Control 1被分到训练集，则所有Control 1和TP 1样本都需归入训练集（实际数据中各Bio分组样本量不均）。

示例数据集：

Sample    Treatment Bio  285.945846 286.9638976 288.1004758 288.8109355
Control1_A13   Control   1 0.003535191 0.001777255 0.004729780 0.002364995
Control1_A14   Control   1 0.005063256 0.000110063 0.006249624 0.001041584
Control1_A15   Control   1 0.004262099 0.000836256 0.004277461 0.002699177
Control2_B13   Control   2 0.002411720 0.000466887 0.001129674 0.001109870
Control2_B14   Control   2 0.003085647 0.001831629 0.002482230 0.000000000
Control2_B15   Control   2 0.001996473 0.001060616 0.003995243 0.001369387
Control3_C13   Control   3 0.000299744 0.000851944 0.002808119 0.004065315
Control3_C14   Control   3 0.003187073 0.000591202 0.006833653 0.001713096
Control3_C15   Control   3 0.003692511 0.000262144 0.004673039 0.000126174
Control4_D13   Control   4 0.003369294 0.001087459 0.005171894 0.000675702
Control4_D14   Control   4 0.003818057 0.000838719 0.005513885 0.000458708
Control4_D15   Control   4 0.002572840 0.000257058 0.003537029 0.000009040
LX2+TP1_E1          TP   1 0.003347067 0.001231945 0.008181087 0.004436654
LX2+TP1_E2          TP   1 0.001552547 0.001463769 0.008864838 0.002728083
LX2+TP1_E3          TP   1 0.003224648 0.000812735 0.008518836 0.004303950
LX2+TP2_F1          TP   2 0.001705551 0.000182659 0.000911028 0.000240785
LX2+TP2_F2          TP   2 0.000760944 0.000759464 0.002486596 0.002377735
LX2+TP2_F3          TP   2 0.001034440 0.000647382 0.008146538 0.001028800
LX2+TP3_G1          TP   3 0.003660741 0.001260433 0.008046637 0.003182006
LX2+TP3_G2          TP   3 0.001802459 0.000547580 0.004882082 0.004121552
LX2+TP3_G3          TP   3 0.003590003 0.000089100 0.002801237 0.000403527
LX2+TP4_H1          TP   4 0.002831592 0.001534135 0.009151124 0.003021942
LX2+TP4_H2          TP   4 0.001863099 0.000959953 0.008284829 0.005169246
LX2+TP4_H3          TP   4 0.005649448 0.001959382 0.011814467 0.004110110

尝试过两种方法，但都存在问题：

• 方法1：仅按Treatment分组拆分，未考虑Bio分组的关联规则

set.seed(1234)
inTraining <- createDataPartition(vis_data2$Treatment, p=0.6, list=FALSE)
training.set <- vis_data2[inTraining,]
Totalvalidation.set <- vis_data2[-inTraining,]
# 拆分剩余40%为测试集和验证集各20%
inValidation <- createDataPartition(Totalvalidation.set$Treatment, p=0.5, list=FALSE)
testing.set <- Totalvalidation.set[inValidation,]
validation.set <- Totalvalidation.set[-inValidation,]

• 方法2：尝试按Bio分组拆分，但运行时常出现NA值，且分组不符合要求

set.seed(1)
# 拆分训练集和验证集
Y1 = vis_data2[,1] # 定义处理组列
g1 = vis_data2[,3] # 定义分组列
final_vis_data <- sample.split(Y1,SplitRatio = 0.5,group = g1)
table(Y1,final_vis_data) # 查看拆分比例
split(final_vis_data,g1) # 尝试保持同组样本在一起
full_train_set <- vis_data2[ final_vis_data,]
test.set <- vis_data2[!final_vis_data,]

# 拆分训练集为训练集和测试集
Y2 = full_train_set[,1] # 定义处理组列
g2 = full_train_set[,3] # 定义分组列
final_vis_data2 <- sample.split(Y2,SplitRatio = 0.5,group = g2)
table(Y2,final_vis_data2) # 查看拆分比例
split(final_vis_data2,g2) # 尝试保持同组样本在一起
test.set <- full_train_set[final_vis_data2,1:3]
validation.set <- full_train_set[!final_vis_data2,1:3]

核心思路是：以Bio编号为基本拆分单元（保证Control N和TP N绑定），先对Bio单元进行比例拆分，再批量分配对应单元的所有样本到训练/测试/验证集。

完整代码 #

set.seed(1234) # 设置随机种子保证结果可重复
# 替换为你的数据读取方式，比如read.csv或read.table
vis_data2 <- read.table("your_data_file.txt", header = TRUE)

# 提取所有唯一的Bio分组编号
bio_groups <- unique(vis_data2$Bio)

# 第一步：拆分训练集的Bio单元（示例为60%比例）
train_bio <- sample(bio_groups, size = round(length(bio_groups)*0.6))
remaining_bio <- setdiff(bio_groups, train_bio)

# 第二步：从剩余单元拆分测试集和验证集（各占剩余的50%）
test_bio <- sample(remaining_bio, size = round(length(remaining_bio)*0.5))
val_bio <- setdiff(remaining_bio, test_bio)

# 第三步：根据Bio单元分配对应所有样本
training.set <- vis_data2[vis_data2$Bio %in% train_bio, ]
testing.set <- vis_data2[vis_data2$Bio %in% test_bio, ]
validation.set <- vis_data2[vis_data2$Bio %in% val_bio, ]

# 验证拆分结果
cat("训练集包含的Bio分组：", paste(train_bio, collapse = ","), "\n")
cat("测试集包含的Bio分组：", paste(test_bio, collapse = ","), "\n")
cat("验证集包含的Bio分组：", paste(val_bio, collapse = ","), "\n")
# 查看各数据集的Treatment和Bio分布
table(training.set$Treatment, training.set$Bio)
table(testing.set$Treatment, testing.set$Bio)
table(validation.set$Treatment, validation.set$Bio)

代码说明 #

• bio_groups提取所有唯一Bio编号，确保拆分时以整个Bio组为单位，保证Control和TP的同编号样本绑定
• sample()对Bio单元随机抽样，完全符合需求中"同Bio编号样本同属一个数据集"的规则
• 可根据实际需求调整拆分比例，比如修改round(length(bio_groups)*0.6)中的数值
• 最后通过%in%筛选样本，批量分配效率高且不会出现分组错误

内容的提问来源于stack exchange，提问作者Ruth Walker