With the boom of machine learning and Artificial intelligence, One of the questions I get frequently asked is
- How to get started with machine learning?
- What books do you suggest?
- Do you suggest any courses? etc.,
But the most common question is there are so many resources which one to follow. Worry not! I have curated resources that worked for so many people. The resources come with some projects for you to try out the concepts you learn.
Now you may get carried away by all the links and books here. Machine learning has a learning curve and usually takes six months of constant effort. Do not give up because you do not understand anything in the first few weeks. One can become a master only by continuous practice.
The resources focused on applied ML and not the core of machine learning. This is a starting point. Once you can build a few models, you can dig deeper to understand how they work.
Machine Learning mastery has an extensive set of roadmaps and blog posts that can guide you through the journey of transforming yourself into an ML Engineer.
I cannot thank this book enough because it takes you through every step of building an ML pipeline. With each chapter, you get a little closer to becoming an ML Engineer. By the time you finish the book, you will know how to tackle an ML problem.
Similar to the above book, this course has helped me simplify my understanding of machine learning. Until then, I thought it was a huge technology that takes years to master. The course suggested few mental models on how to approach each ML problem, which came in handy later when solving intensive problems.
Pretty much every exercise on Kaggle is a project worth building. While my focus is is mostly on text and NLP, you should focus on different kinds of problems like numerical, categorical, predictive, clustering, etc., The above resources should prepare you for the following projects.
- Twitter hate speech classifier
- Hot-dog not hot-dog
- Book/Movie recommendation system
- More projects
Write to us @thelearningdev