Documenting my learning progress in machine learning (ML) and artificial intelligence (AI)
At university I've learned relevant theory in subjects such as calculus, linear algebra, or probability and statistics. I've also taken a more focused course on optimization algorithms and an introductory ML course on Coursera. My baseline is that I understand how optimization problems are formulated and what are some of the typical regression and classification algorithms used in ML. I can navigate implementation details, but I lack fluency in more recent concepts and terminology, especially in deep learning. For example, it is hard to understand the model implementation diagrams in scientific literature or keep up with all the model variants.
The programming language of my choice is Python. I have approximately 1.5 years of professional software developer experience and have spent even a longer time programming in a scientific environment back at university. I am very comfortable with the language, its best practices, and part of the scientific library suite relevant for ML - numpy, pandas, matplotlib.
Because the field is moving rapidly, the goal is to focus on applications and learn through hands-on projects. Although this can be more frustrating, it allows me to focus on skills more relevant for the industry. Moreover, more time is invested learning skills that will be directly useful to me, and if there is confusion at a particular step, I can sidetrack and dive into the theory.
-
Preliminary skills
1.1 Data exploration and preparation
1.2 SQL and databases
1.3 Regression project
1.4 Classification project
-
Deep learning skills
2.1 Keras API
2.2 Tensorflow
2.3 ?
-
Cloud skills
3.1 Spark
3.2 ?
-
CI/CD, MLOps skills
-
Capstone project ?