Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/elenacarino-max/Pildora4_ext_StarWars/llms.txt

Use this file to discover all available pages before exploring further.

The live coding notebooks in notebooks/01_live_coding/ are designed for instructor-led sessions that demonstrate MLflow experiment tracking from scratch, before or alongside the Streamlit app. They follow the same Digits dataset and Random Forest workflow as the six chambers and three missions, but present the concepts in a linear, cell-by-cell Jupyter format rather than a gamified sequence. This makes them well-suited for instructors who prefer a narrated walkthrough, for students who want to follow along on their own machines, or as a reliable fallback when the app is unavailable.

​
Available notebooks

Live_Coding_COMPLETO.ipynb

Complete reference version. Every cell is filled in with working code covering the full tracking cycle — experiment, run, parameters, metrics, artifacts, and model. Intended for instructor preparation and as a reference to consult after the session. Do not project this version live unless you want to reveal all solutions immediately.

Live_Coding_DEMO_SEGURA.ipynb

Recommended for live classroom demonstration. A carefully curated version of the complete notebook that is safe to project: it demonstrates the full tracking cycle step by step without spoiling chamber or mission solutions. Use this as your primary screen-share during the session.

Live_Coding_PARA_RELLENAR.ipynb

Student fill-in-the-blanks template. The same structure as the demo notebook, with key cells left blank for students to complete. Distribute this to students at the start of the session so they can practice alongside the live demonstration. Collecting completed copies provides a lightweight formative assessment.

​
How to use these notebooks

1

Prepare with the complete reference

Open Live_Coding_COMPLETO.ipynb before the session. Run all cells to verify your environment, review the narrative, and decide which cells to emphasise or skip based on the available time.
2

Demonstrate with the safe demo notebook

During the session, open Live_Coding_DEMO_SEGURA.ipynb as your shared screen. Walk through each concept in order — experiment naming, opening a run, logging parameters, logging metrics, saving artifacts, and logging the model — pausing at each cell to explain the MLflow call before executing it.
3

Students follow along with the fill-in notebook

Have students open Live_Coding_PARA_RELLENAR.ipynb on their own machines or in a shared cloud environment. As you complete each cell in the demo, students fill in the corresponding blank and run it themselves. This parallel practice reinforces the connection between the narrative explanation and the actual Python syntax.

​
Prerequisites

The notebooks require a Python environment with the packages listed in app-v2/requirements.txt. At a minimum you need:
  • mlflow
  • scikit-learn
  • pandas
  • matplotlib
  • jupyter (for JupyterLab or classic Notebook)
Install everything at once from the project root: 
cd app-v2
pip install -r requirements.txt
jupyter lab
No Docker is required for the notebooks — they connect to a local MLflow tracking server backed by a SQLite file that is created automatically on the first run.

​
Relationship to the Streamlit app

The live coding notebooks and the Streamlit app cover exactly the same six MLflow concepts:
App chamberNotebook cell topic
Chamber 1 — Experimentmlflow.set_experiment(...)
Chamber 2 — Runwith mlflow.start_run(run_name=...)
Chamber 3 — Parametersmlflow.log_params(parametros)
Chamber 4 — Metricsmlflow.log_metrics(metricas)
Chamber 5 — Artifactsmlflow.log_artifact(...)
Chamber 6 — Modelmlflow.sklearn.log_model(...)
The key difference is workflow: the notebooks are linear and instructor-paced, while the app is gamified and self-paced with automated validation. Using both in the same session (live coding first, app second) gives students two exposures to each concept in different contexts, which significantly improves retention.
The live coding notebooks are also the primary contingency plan if the Streamlit app is unavailable during a session. As noted in the GUIA_PROFESORADO: “Si la aplicación falla, utiliza los notebooks y documentos conservados en la raíz del repositorio. No los elimines: forman parte del plan de continuidad.” Keep the notebooks accessible at all times and verify they run in your environment before every session. See the Contingency page for the full fallback procedure.

Build docs developers (and LLMs) love

Get started for freeTalk to us