How To Split Data Into Test And Train In R

Splitting Data into Grooming & Testing Sets in R (Example Code)

In this article yous'll acquire how to split a data frame into training and testing data sets in the R programming language.

Creating Example Information

set.seed(                  1043857                  )                  # Creating example data                  df_all                  <-                  information.frame(x                  =                  round                  (runif(                  100                  )                  ,                  two                  )                  ,                  y                  =                  round                  (runif(                  100                  )                  ,                  two                  )                  ,                  z                  =                  sample(letters,                  100                  ,                  supervene upon                  =                  TRUE                  )                  )                  head(df_all)                  # Head of instance data                                    #      x    y z                                    # one 0.83 0.67 grand                                    # 2 0.17 0.88 x                                    # 3 0.88 0.21 grand                                    # 4 0.15 0.43 h                                    # v 0.74 0.29 r                                    # six 0.80 0.11 thou

set.seed(1043857) # Creating example information df_all <- data.frame(ten = round(runif(100), 2), y = round(runif(100), ii), z = sample(letters, 100, supervene upon = TRUE)) head(df_all) # Head of case data # 10 y z # ane 0.83 0.67 one thousand # two 0.17 0.88 x # 3 0.88 0.21 grand # 4 0.15 0.43 h # 5 0.74 0.29 r # 6 0.80 0.11 k

Example: Creating Railroad train & Test Data with sample Part

ind_split                  <-                  sample(c(rep(                  "railroad train"                  ,                  0.75                  *                  nrow(df_all)                  )                  ,                  # Create dummy for splitting                                    rep(                  "examination"                  ,                  0.25                  *                  nrow(df_all)                  )                  )                  )                  # 75% train; 25% exam data

ind_split <- sample(c(rep("railroad train", 0.75 * nrow(df_all)), # Create dummy for splitting rep("exam", 0.25 * nrow(df_all)))) # 75% train; 25% test data

df_train                  <-                  df_all[ind_split                  ==                  "railroad train"                  ,                  ]                  # Creating train data set                  head(df_train)                  # Head of railroad train information                                    #      x    y z                                    # one 0.83 0.67 g                                    # two 0.17 0.88 x                                    # 3 0.88 0.21 yard                                    # 4 0.15 0.43 h                                    # five 0.74 0.29 r                                    # 6 0.fourscore 0.eleven grand

df_train <- df_all[ind_split == "train", ] # Creating train information set head(df_train) # Head of train information # 10 y z # 1 0.83 0.67 g # 2 0.17 0.88 x # 3 0.88 0.21 yard # 4 0.15 0.43 h # 5 0.74 0.29 r # vi 0.eighty 0.11 k

df_test                  <-                  df_all[ind_split                  ==                  "examination"                  ,                  ]                  # Creating test data set                  caput(df_test)                  # Head of test data                                    #       10    y z                                    # 7  0.86 0.58 q                                    # 12 0.31 0.24 f                                    # 17 0.43 0.89 i                                    # 20 0.64 0.03 b                                    # 26 0.30 0.48 o                                    # 27 0.83 0.04 c

df_test <- df_all[ind_split == "test", ] # Creating test data set head(df_test) # Head of test data # x y z # vii 0.86 0.58 q # 12 0.31 0.24 f # 17 0.43 0.89 i # twenty 0.64 0.03 b # 26 0.30 0.48 o # 27 0.83 0.04 c