ML crash course

These notebooks provide a high-level overview of ML, along with curated material for those who want to learn more about ML and ML applications. Thus, many of the materials used in notebooks are from the internet and other sources, and I give credit and references accordingly.

ML is a fast-moving area in terms of models, algorithms, and software frameworks and libraries, so I try to present the most recent versions of the implementation; nevertheless, they may be outdated by the time this material is reviewed.

In part because there are many SW tools and methods used in ML industry, infrastructure, and MLOps, I will only cover a few representative methods. However, there could be a variety of different tools and methods employed in ML pipelines.

Colab is a recommended way to run the notebooks for easy testing; however, notebooks can be run on other types of Jupyter notebook environments.

Contents

0. Review
   • Linear Algebra
1. ML overview
2. Linear Regression (predicting numbers)
3. Classification
4. Neural Network, simple neural network code
5. Convolutional Neural Network(CNN), sample CNN code
6. Sequential problem with Recurrent neural network(RNN)
7. ML development process
8. (TBD) MLOps and Deep Learning Computation

Credits

Note: Credits for slides are described in each slide.

1. Numpy tutorial.
2. Chapter 1 from Book Hands-on Machine Learning with Scikit-Learn, Keras and TensorFlow
3. ML overview
4. ML linear regression

Further readings