AI

ESIA

School of Artificial Intelligence

Worldwide cohort
Students across time zones
HomeMaster’sTrainingsProjectsResearchBlogAboutContact
← Back to Trainings

Machine Learning Foundations (Python)

A practical, end-to-end introduction to supervised ML: data prep, modeling, evaluation, and production-ready baselines.

MLPythonscikit-learnMetricsPipelines

Duration

3 days

Format

Remote

Level

Beginner

Key outcomes

  • Frame problems correctly (regression vs classification) and avoid leakage
  • Build preprocessing + modeling pipelines with scikit-learn
  • Evaluate models with the right metrics and error analysis
  • Deliver a reproducible baseline ready for API integration

Syllabus

Day 1 — From data to features

  • Problem framing and dataset split strategy
  • Leakage pitfalls and validation mindset
  • Missing values, encoding, scaling
  • Pipelines: ColumnTransformer + Pipeline

Day 2 — Models & evaluation

  • Baselines: linear/logistic, trees, random forest
  • Cross-validation and learning curves
  • Metrics: RMSE, ROC-AUC, F1, precision/recall
  • Error analysis and model debugging

Day 3 — Baseline to production

  • Hyperparameter search (Grid/Random)
  • Saving/loading models, versioning patterns
  • Input validation (schema) and inference contract
  • Packaging the predictor for FastAPI integration