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.

MLflow Jedi can be up and running on any machine in under five minutes through either of two paths: a local Python environment using Python 3.12, or Docker Compose for a fully containerised deployment that persists data across restarts. Both options expose the app on http://localhost:8501 and require only a copied .env file to configure. Choose the option that matches your setup and follow the steps below.
Change MLFLOW_JEDI_TEACHER_PASSWORD in your .env file before sharing the app with anyone. The default value 1234 in .env.example is for local testing only and must not be used in any shared or networked environment.

​
Option A: Local Python

This path requires Python 3.12 installed on your machine. All commands below use PowerShell syntax; adapt to your shell as needed.
1

Clone the repository and enter the app directory

git clone https://github.com/elenacarino-max/Pildora4_ext_StarWars.git
cd Pildora4_ext_StarWars/app-v2
2

Create and activate a virtual environment

py -3.12 -m venv .venv
.\.venv\Scripts\Activate.ps1
On macOS/Linux, activate with source .venv/bin/activate instead.
3

Install dependencies

python -m pip install -r requirements.txt
This installs Streamlit 1.49.1, MLflow 3.3.2, scikit-learn 1.7.1, pandas, Matplotlib, and python-dotenv.
4

Configure environment variables

Copy the example file and open it in any text editor:
Copy-Item .env.example .env
Set a strong value for MLFLOW_JEDI_TEACHER_PASSWORD before continuing. The other two variables have sensible defaults and do not need to change for local use:
MLFLOW_JEDI_TEACHER_PASSWORD=your-strong-password-here
MLFLOW_JEDI_DATA_DIR=./data
MLFLOW_JEDI_MAX_TEAMS=10
5

Start the application

streamlit run app.py
Streamlit will print a local URL in the terminal once it is ready.
6

Open the app in your browser

Navigate to http://localhost:8501. The MLflow Jedi home screen — La Cámara de los Experimentos Perdidos — will load immediately.

​
Option B: Docker Compose

This path requires Docker with Docker Compose installed. It builds a self-contained image and mounts a named volume so that sessions, scores, and MLflow runs survive container restarts and rebuilds.
1

Configure environment variables

From the app-v2/ directory, copy the example env file:
Copy-Item .env.example .env
Edit .env and replace the placeholder teacher password before building the image.
2

Build and start the container

docker compose up --build
Docker will download the python:3.12-slim base image, install all dependencies, and start Streamlit on port 8501. The first build typically takes two to three minutes.
3

Open the app in your browser

Navigate to http://localhost:8501. The application is ready when the Streamlit home screen appears. To stop the app without losing any data, press Ctrl+C in the terminal and run docker compose down.

​
First Session Flow

Once the app is running, follow these steps to create a session and get teams participating.
1

Create a session in the Profesorado panel

Click Soy profesora on the home screen (or select Profesorado in the sidebar), enter your teacher password, and press Nueva sesión. The app generates a unique session code in the format JEDI-XXXX.
2

Share the JEDI-XXXX code with your teams

Display or dictate the JEDI-XXXX code to the students. Each team will need this exact code to join. You can project it from the teacher panel or write it on a whiteboard.
3

Teams join via the Alumnado portal

Students click Soy un equipo on the home screen (or select Alumnado in the sidebar) and enter the JEDI-XXXX session code they were given.
4

Each team picks an available name

After entering the code, a dropdown shows the team names that are still available for this session. Each team selects one name and confirms to enter the activity. Up to ten teams can join a single session simultaneously.

​
Available Team Names

MLflow Jedi ships with ten fixed Star Wars team names. Each name can be claimed by only one team per session:
#Team Name
1Guardianes de Naboo
2Centinelas de Coruscant
3Exploradores de Tatooine
4Cronistas de Alderaan
5Custodios de Dagobah
6Vigías de Endor
7Navegantes de Hoth
8Forjadores de Mustafar
9Archiveros de Jedha
10Protectores de Lothal
If you just want to explore MLflow Jedi without any installation, the live public instance is available at https://pildora4extstarwars-14.streamlit.app/. Create a session from the Profesorado portal and join it from a second browser tab as a team — no setup required.

Build docs developers (and LLMs) love

Get started for freeTalk to us