Linear Algebra
Standard libraries and software packages like sciki-tlearn, xgboost are commonly used for prototyping different Machine Learning algorithms, because they provide the quick turnover when it comes to things like hyper parameter tuning and model evaluation. But when it comes to deploying models in production, the ability to code ML algorithms from scratch, gives the developers greater flexibility to customize models and have fewer dependencies on 3rd party applications (standard libraries) which are always subject to change. Performance issues could also be addressed through this use case, since the codebase would optimized to a specific problem statement.
Since linear algebra forms the basis of most of the machine learning problems, the idea behind this repository is 2-fold:
- Learn the concepts of linear algebra.
- Code them in Python.
I am also trying to learn how to write docstrings and unittests for error handling, which are standard programming practices through this exercise.