This notebook mainly contains digests about machine learning theories, applications and code snippets.
I learn best with toy code that I can play with. http://iamtrask.github.io/2015/07/12/basic-python-network/
I'm always trying to rebuild the models from memory, which is a killer exercise.
All my external knowledge is stored in EverNote with Markdown & LaTeX support. After editing, I port those highly structured ones into GitBook.
I tried to build the whole notetaking system with Jupyter notebooks, but failed due to the lack of hierarchy and global search functionality.