Welcome to AI Artisan! This is a repository where you can hone your skills as an "AI artisan" by crafting your own machine learning models from the ground up. If you are looking to learn machine learning by doing, and you want to build your own algorithms from scratch to understand how they work under the hood, then you've come to the right place!
" I do not understand what I cannot create" R.F.
You might be wondering why you should bother building machine learning algorithms from scratch when there are so many libraries and pre-built models available. The truth is, there's no better way to really understand how machine learning works than to build it yourself. By reinventing the wheel, so to speak, you gain a deeper understanding of the underlying principles and can better troubleshoot problems when they arise. Plus, it's just plain fun!
In this repository, you'll find the basic building blocks of machine learning, including:
- Decision Trees
- Bagging
- Boosting
- Support Vector Machines
This is just the beginning. As the repository grows, so too will the list of algorithms included.
Each algorithm is implemented from scratch in Python, with code and comments to help you understand how it works. You can explore the code and run the algorithms yourself to see how they perform on different datasets.
In addition to the code, this repository also includes sample datasets to use with each algorithm. You can find these in the data folder.
Here are some ideas for future additions to the repository:
- Rethink the module use ? -> This repo has been changed to a ** monorepo**
- Tree ML algo only? - -> No, monorepo with all the algorithms
- Refactor and Improve readeability of the decision tree class: **In progress **
- The bagging supports only classification for now: We need to add the regression part
- The accuracy / precision / f1 score measurements need to be implemented.
- Special visualizations need to be implemented: accuracy / error wrt to n_estimators for bagging etc ...
- More machine learning algorithms, such as neural networks and clustering algorithms.
- Tutorials and articles on specific topics in machine learning.
- A community forum or chat where you can discuss your ideas and get feedback from other "AI artisans."
We hope you enjoy using AI Artisan to build your machine learning skills from scratch. Happy coding!
AI Artisan is licensed under the MIT License. See the LICENSE file for more information.