ML Workflows – ML Associate Study Guide
This domain accounts for 19% of the exam and covers the core data-preparation steps of an ML workflow on Databricks: exploratory data analysis, cleaning, feature engineering, splitting, and handling imbalance.
Exploratory Data Analysis (EDA)
Compute summary statistics (mean, median, quartiles) and visualize distributions (histograms, box plots) to understand each feature and detect outliers before modeling.
Cleaning: Missing Values and Outliers
Handle missing values by imputation (mean/median or model-based) or by dropping rows/columns, judged by how much signal is affected. Don't leave nulls that silently break training.
Unlock with a paid plan
4 more sections (about 175 words) continue below. Every paid plan unlocks all Machine Learning Associate study guides, the full question bank and mock exams, from $4.97/month.