Supervised Workflow
Prepare, validate, debug and demonstrate readiness on unfamiliar data.
01 / Prepare
Explore training data
Keep input exploration inside the training population.
12–18 min · 3 practicesA useful reference
Compare with a simple predictor that ignores X.
12–18 min · 3 practicesLearn a scale from training rows
Distinguish fitting a scale from transforming with it.
12–18 min · 3 practicesEncode categories without inventing order
Preserve a reusable categorical schema.
18–25 min · 4 practicesPrepare different feature types together
Apply each preparation to the intended columns.
12–18 min · 3 practicesKeep preparation with the estimator
Fit and predict through one pipeline.
12–18 min · 3 practicesLearn missing-value replacements safely
Fit replacement statistics inside the training workflow.
12–18 min · 3 practicesPreparation retrieval
Retrieve earlier concepts before combining them.
15–20 min · 3 practices02 / Validate and compare
What a fold does
Separate fitting, validation and final-test row roles.
12–18 min · 3 practicesRead cross-validation evidence
Interpret scores and compare on matching folds.
18–25 min · 4 practicesSettings, learned values and fit quality
Separate hyperparameters, learned parameters and generalisation evidence.
12–18 min · 3 practices03 / Select, diagnose and finish
How a search makes a choice
Understand search boundaries before model-specific tuning.
12–18 min · 3 practicesDiagnose without opening the final test
Use predictions from models that did not fit each diagnostic row.
12–18 min · 3 practicesFinish once, then report
Separate selection from final evidence.
12–18 min · 3 practicesRespect time
Validate in the direction the model will be used.
18–25 min · 4 practicesWorkflow evidence retrieval
Retrieve earlier concepts before combining them.
15–20 min · 3 practicesSupervised Workflow checkpoint
Build a mixed LinearRegression pipeline, compare a mean reference with five-fold CV, create OOF residuals, keep defaults and report final RMSE. No tree tuning is required.
25–35 min · 1 practices04 / Debug and demonstrate readiness
Repair a leaking workflow
Find the point where unavailable information enters a prediction.
25–40 min · 3 practicesWhen a good score is misleading
Choose validation evidence that answers the actual question.
25–40 min · 3 practicesReadiness · unfamiliar regression
Independently estimate a new repair’s completion time at intake.
35–50 min · 1 practicesReadiness · unfamiliar classification
Independently prioritise units that may develop a fault next week.
35–50 min · 1 practices