Baseline Classifier on a Messy Tabular Dataset

Asked at Coinbase · ML Fundamentals & Algorithms · Difficulty: Medium · Last asked Dec 19, 2025

You are given a tabular dataset for a binary classification task, loaded into a pandas DataFrame `df` with a label column `target`. The data is deliberately messy: some numeric columns have missing values, some categorical columns spell the same category in several ways (for example `"NY"`, `"ny "`, `"New York"`), one column mixes numbers and strings, and an identifier column is present. Build a clean, working baseline classifier end to end and explain your choices.

Loading the problem…