I started off simply taking notes on the TensorFlow white paper, but as I worked I started putting more and more time into finding/linking to reference material in the TensorFlow documentation and resources. Additionally, I attempted to increase my understanding of the paper by re-phrasing certain sections as well as translating some of the algorithms from paragraph-form to ordered lists.
Reading the white paper has improved my comfort with the APIs dramatically, and I hope that others will benefit from my notes.
- Notes broken down section by section, as well as subsection by subsection
- Relevant links to documentation, resources, and references throughout
- SVG versions of figures/graphs
- So many bullet points!
- Create and utilize anchor tags throughout notes for self-navigating
Please feel free to submit pull requests for corrections, improved readability, consistency in terminology, additional graphs or whatever else you can think of.