A list of 99 machine learning projects for anyone interested to learn machine learning from coding and building projects.
Our working philosophy is to provide a curated repo for anyone to contribute a cool/fun exercise and solution that is useful for anyone (including themselves) in their journey of learning machine learning.
The format is roughly the following:
-
Propose an exercise by creating an issue ticket and write what you think is an useful coding exercise for certain concepts.
-
If enough people are interested in that issue ticket, hopefully either you or someone else will write the exercise statement properly similar to the style of a lab exercise/homework question.
-
Then someone will fork the repo, write up their solution, with a bit of polish and documentation, submit a pull request. Please see general contribution guidelines for more details on how to contribute solutions.
-
Some of us will scrutinize the codes, review, make suggestions and eventually include (merge) them into the main project repo.
-
At anytime, someone can repeat suggest improvements/changes to 3-4 above for a particular exercise. This is done by creating an issue ticket for improvement/enhancement. One can then repeat 3-4.
-
Finally, repeat 1-5 indefinitely till we hit 99/99 projects.
Please abide by code of conduct guidelines to have an open and friendly open source collaboration.
Some of the libraries (and their versions) we are using:
- Python (>= 3.6)
- NumPy (>= 1.18.5)
- Pandas (>= 1.0.5)
- Matplotlib (>= 3.2.2)
- Seaborn (>= 0.10.1)
- Scikit-learn (>= 0.22.2)
- Tensorflow (>= 2.2.0)
- PyTorch (>= 1.5.1)
If you want to get in touch with us, say hi on our discord/gitter chatroom:
This project is inspired by Unnit Metaliya’s answer on quora: https://qr.ae/pNK0FW
For credits, these are the two repos (one for C and one for React) where I got the idea from:
- https://github.com/truedl/c-for-beginners
- https://github.com/UnnitMetaliya/99-reactjs-project-ideas
This repo is covered under The MIT License.