exercise5_machine_learning

Exercise 5: Machine Learning

In this exercise you will explore the data and build a simple classifier.

Development

Pre-requirements

This exercise requires (at least) the following services to be up & running:

• elasticsearch

Setup

First, launch the jupyter service

docker compose start jupyter

• Open Jupyter: http://localhost:8888
• You need the token to start, and this can be retrieved in two different ways:
  • Look at the jupyter container logs (using lazydocker, Visual Studio Code or Docker Dashboard)
  • Connecting to the node:
    • Run on command line: docker exec -it docker_jupyter_1 bash
    • On docker container get token: jupyter notebook list
    • Copy the token
• Now that you have the token, just use it in the login screen

Part 1: Load and test the datasets

• Load notebook. This is done by clicking on "Upload" and load the MLExercise.ipynb
• Follow the instructions there

Part 2: Play around

Now that you can access the data and it is included in the dataframe, feel free to play around with the data and continue with any analysis of your choice.

Resources

• Jupyter Documentation
• Python Documentation
• Pandas Documentation
• scikit-learn Documentation