The materials may include courses, tutorials, books, etc.<p>The topics can include:<p>1. Data pre-processing<p>2. Library specific tutorials (e.g. PyTorch, MxNet, scikit-learn, etc)<p>3. Building things from scratch (using numpy, scipy, vanilla python, etc)<p>4. Foundations and theory behind the popular algorithms<p>5. Applications like Computer Vision, NLP, etc.<p>6. How to read and implement research papers.<p>7. The math of ML/DL<p>The difficulty of the materials can be anything ranging from someone with programming experience starting out or someone who is a practitioner and wants to look at more deeper explanations.
CMU has two open courses on deep learning which are very good.<p>1. Deep Learning - <a href="https://deeplearning.cs.cmu.edu/F23/index.html" rel="nofollow noreferrer">https://deeplearning.cs.cmu.edu/F23/index.html</a><p>2. Deep Learning Systems - <a href="https://dlsyscourse.org/" rel="nofollow noreferrer">https://dlsyscourse.org/</a><p>The second course dives much deeper into the internals of the libraries and all.