diogo364 / mlbookcamp_project

ML Bookcamp Project

Table of Contents

• ML Bookcamp Project
  • Table of Contents
  • About
  • Repo Structure
    • Environment
    • Conda
    • Docker
  • References

About

This repo contains all the code and projects developed by following the MLBookCamp-code Course offered by DataTalks.Club and Alexey Grigorev.

Repo Structure

All homework questions are solved in Jupyter Notebook format at the root of the repository. Further custom code might be found inside a few packages, such as model.

Environment

Choose one of the following environment instructions below to build all dependencies to the project.

Conda

All requiments needed to run the project might be found within the docker/resources/conda.yml file, which might be used to build a conda environment easily through the command:

conda env create -f docker/resources/conda.yml

Obs: Conda or Miniconda are required for this setup.

Docker

It is also possible to run the project through a local Jupyter Notebook by using a Docker Container in a few steps.

1. Access the .env file to customise the environment variables.

   Note: Map the USER_UID variable to your User ID.

2. Build the Docker Image by running the command:

   docker-compose build
   

3. Run the Docker container in detached mode by running:

   docker-compose up -d
   

4. Verify if your container is running through the command below:

   docker ps
   

   Note: You should see an output like this:

   CONTAINER ID   IMAGE                     COMMAND                  CREATED         STATUS         PORTS                                       NAMES
   03ca4cdb7893   miniconda-docker:latest   "jupyter notebook --…"   3 minutes ago   Up 3 minutes   0.0.0.0:8000->8000/tcp, :::8000->8000/tcp   mlbookcamp_project_myenv_1
   

After this last step you should be able to access a local instance of Jupyter Notebook in the following link: http://localhost:8000/

Important: The default authentication token is DOCKERNOTEBOOK, but it can be changed in the .env file.

Obs: Docker and Docker-Compose are required to this setup.

References

• mlbookcamp-code: https://github.com/alexeygrigorev/mlbookcamp-code/
• mlbookcamp-code 2022 Cohort: https://github.com/alexeygrigorev/mlbookcamp-code/blob/master/course-zoomcamp/cohorts/2022