Schedule
The course runs for 6 weeks: 5 content sessions and one session for final presentations. Both sections cover exactly the same material each week.
| Week | Casa Central Wednesday 16:05–18:40 |
Vitacura Friday 11:05–13:40 |
Session | Sprint |
|---|---|---|---|---|
| 1 | Oct 28 | Oct 30 | Python in Google Colab | Sprint 1 starts |
| 2 | Nov 4 | Nov 6 | Data analysis with pandas | Sprint 1 review · Sprint 2 starts |
| 3 | Nov 11 | Nov 13 | Linear regression and model evaluation | Sprint 2 review · Sprint 3 starts |
| 4 | Nov 18 | Nov 20 | Classification: logistic regression and metrics | Modeling clinic |
| 5 | Nov 25 | Nov 27 | Trees, random forests, and communicating results | Sprint 3 review |
| 6 | Dec 2 | Dec 4 | Final presentations | Final assessment |
Key project dates
- Team and case registration: at the end of session 1.
- Notebook and slides due: Tuesday, December 1, 23:59 (both sections).
- Presentations: during each section's class time, December 2 (Casa Central) and December 4 (Vitacura).
Each session lasts 2 hours and 35 minutes and has two parts: a first block of concepts with live coding and, after a break, a workshop where teams apply what they just learned to their project.
Session 1 · Python in Google Colab
Casa Central: Wednesday, October 28 · Vitacura: Friday, October 30
By the end you can: open and run notebooks in Colab; work with variables, data types, lists, dictionaries, conditionals, and functions; and operate on NumPy arrays.
- Colab guide: open, save a copy, run, and share.
- Class 1: Introduction to Python
- Class 2: NumPy
Workshop: form teams, choose a case, and load the data into a shared notebook. → Sprint 1
Session 2 · Data analysis with pandas
Casa Central: Wednesday, November 4 · Vitacura: Friday, November 6
By the end you can: read CSV and Excel files; select, filter, group, and sort data; spot missing values, duplicates, and inconsistent categories; and build exploratory charts with seaborn.
Workshop: Sprint 1 review; cleaning and exploratory analysis of your case. → Sprint 2
Session 3 · Linear regression and model evaluation
Casa Central: Wednesday, November 11 · Vitacura: Friday, November 13
By the end you can: split data into training and test sets, encode categorical variables, train a linear regression, interpret its coefficients, measure error (MAE, RMSE, R²), and recognize overfitting.
Workshop: Sprint 2 review; prepare the data for modeling and train your first model. → Sprint 3
Session 4 · Classification: logistic regression and metrics
Casa Central: Wednesday, November 18 · Vitacura: Friday, November 20
By the end you can: train a logistic regression, read a confusion matrix, choose between accuracy, precision, recall, and F1 based on the cost of each error, and adjust the decision threshold.
Workshop: modeling clinic. Each team brings its baseline model and we review it together.
Session 5 · Trees, random forests, and communicating results
Casa Central: Wednesday, November 25 · Vitacura: Friday, November 27
By the end you can: train decision trees and random forests, control their complexity, read feature importance, compare models with a ROC curve, and build a presentation using the pyramid principle.
Workshop: Sprint 3 review; compare models and structure the final presentation.
Session 6 · Final presentations
Casa Central: Wednesday, December 2 · Vitacura: Friday, December 4
Each team presents its project: 10 minutes of presentation and 5 of questions. The presentation order is published on Monday, November 30. Details are in the project brief.