Stratified Sampling 편집하기 (부분)

==How Stratified Sampling Works==
Stratified sampling divides the dataset into mutually exclusive groups (strata) based on one or more specific features, and samples are then drawn from each group proportionally.

# '''Define Strata''': Identify one or more features for stratification (e.g., class labels in classification tasks). 
# '''Partition Data into Strata''': Divide the data based on the selected feature(s) so that each stratum contains data points with the same characteristic. 
# '''Sample Proportionally''': For each stratum, randomly sample a proportional subset that reflects the original distribution within the dataset.