Wine Quality Classification

Overview

Requirements

• Python 3.10, pip
• Poetry

Setup

1. Install dependencies: poetry install (This will create a poetry.lock file, which contains exact versions of the dependencies and should be committed to the repo.)
2. Activate environment: poetry shell
3. Update configs in config.yaml
4. Start working with the notebooks: run Jupyter lab: jupyter lab (or use your IDE to run the notebooks)

To add new dependencies run poetry add <pkg> update dependencies run poetry update.

Problem formulation

Data sources

Baseline

Model

Experiments

To change the default tracking location to another folder you can use put the following to your .env file: MLFLOW_TRACKING_URI="file:///home/<path>/mlruns"

To see the experiments:

1. (Optional) Load your environment variable from the .env file: source .env
2. Run mlflow ui --backend-store-uri $MLFLOW_TRCKING_URI to view the tracked experiments. (If you did not change the MLFLOW_TRCKING_URI, just run mlflow ui)

Evaluation

Next steps