Skip to content

Introduction to Machine Learning - Environment Setup

Anaconda Environment

We choose to use Anaconda to set up the environment for machine learning. On one hand, it comes with many commonly used data science packages such as Numpy, TensorFlow, etc., and includes the modules they depend on; on the other hand, it can easily manage and switch runtime environments. In short, Anaconda is an all-in-one data science programming environment.

Download Anaconda from: https://www.anaconda.com/download

After installation, how do we get started? You can follow this official course step by step: Get Started with Anaconda, or you can directly refer to the following concise steps.

Once the installation is complete, you can verify the installation status and the list of built-in packages by entering the command conda list on the Anaconda Prompt command-line interface.

It is recommended by the official documentation to create a virtual environment. This way, if something goes wrong, you can easily switch to another one without having to uninstall and reinstall:

conda create --name NEW_ENV_NAME
conda activate NEW_ENV_NAME

The default channel for conda is defaults, but this channel does not have all the packages, so it is suggested to change it to conda-forge:

conda config --add channels conda-forge

Next, install some commonly used packages in this virtual environment:

conda install jupyterlab rich faker chime schedule pandas scikit-learn

Finally, start JupyterLab:

jupyter lab

scikit-learn Machine Learning Toolkit

In the upcoming articles, we will dive into machine learning with scikit-learn. Scikit-learn (sklearn) contains fundamental and commonly used machine learning algorithms for classification, regression, dimensionality reduction, clustering, as well as modules for feature extraction, data processing, and model evaluation, making it perfect for beginners.

References and Acknowledgments

This post is translated using ChatGPT, please feedback if any omissions.