Full API

Splits data, builds a preprocessing pipeline, and fits the model.

Signature

model.train(algorithm=None)

What it does

• Splits rows into train / test sets (before any preprocessing)
• Builds a ColumnTransformer: median imputation + standard scaling for numerics; mode imputation + OneHotEncoding for categoricals
• Wraps it in a Pipeline with the chosen estimator
• Fits on training rows only
• Prints a plain-English summary of every decision

Example

model.train()                           # auto-select algorithm
model.train(algorithm="random_forest")  # explicit