0 of 3 done
Most of an ML engineer’s time goes here, not on the model. “Garbage in, garbage out” is the iron law: no algorithm rescues bad data. Part of the ML Engineer track.
- Missing Data & Cleaning — [Intermediate] — handling missing values, outliers, and dirty data
- Feature Engineering & Scaling — [Intermediate] — turning raw data into useful features, normalization/standardization
- Train/Val/Test Splits — [Intermediate] — holding out data honestly so evaluation measures generalization
Related
- Classical ML — the algorithms this data feeds
- Model Evaluation — why the test split must stay untouched
Jump to a subject
Software engineering
Infrastructure & security
Computer science
Hardware & engineering