Intro to Data Science

Full overview

4-week intro to data science. You start from scratch (or near) and progress to deploying a supervised machine learning model. Each week combines lectures, guided labs and a personal project coached by the trainer.

Data science is no longer reserved for tech giants. Today, an SME that leverages its data — sales, inventory, customer behavior — gains a decisive competitive advantage. This 4-week training is designed for technical profiles (devs, sysadmins, analysts) who want to add data science to their toolkit, with no advanced math prerequisites.

Week one lays the foundations: Python for data (NumPy, Pandas), cleaning and preparing real datasets, exploratory data analysis (EDA) with visualizations (Matplotlib, Seaborn). Week two introduces supervised machine learning: linear regression, decision trees, random forests, cross-validation, and performance metrics (accuracy, precision, recall, F1-score).

Week three broadens the scope: clustering (K-means, DBSCAN) for customer segmentation, time series for forecasting, and an introduction to NLP for text analysis. The final week is dedicated to production: packaging a model, creating a prediction API with FastAPI, deploying to a cloud platform, and monitoring performance in production. You finish with a supervised personal project: from data acquisition to the prediction API.

Curriculum

1. 01Week 1 — Python data: Pandas, NumPy, Jupyter notebooks
2. 02Week 2 — Visualization and data storytelling
3. 03Week 3 — Statistics, data preparation, feature engineering
4. 04Week 4 — Supervised ML, evaluation, production deployment