Algorithms

You need to find a good detector, or maybe an OCR or a speech2text tool. Where do you go ?

There are two entryway into the world of algorithms: - Academic: searching for research paper. Very efficient when you are up to date with your state of the art, gives access to the best performances available. Implementation can be an issue. - Code driven: search directly for githubs, or browse your favorite framework's model zoo. A fast implementation is much easier than going the academic way, but performances might not be the best.

arXiv.org [Academic]

This website hosts a number of published and unpublished research papers, if your building your state of the art, you will most likely stumble upon it.

Medium [Code]

Medium hosts a good number of blog-like posts, covering a wide variety of subjects, among which is deep learning. These posts usually take between 7 and 10 minutes to read, and can help you implement an entire algorithm. The strength of Medium lies in its vulgarization power, and its very up to date posts which cover the major problem of code-driven search: keeping up with the state of the art.

Siraj Raval [Academic and Code]

One of the best youtuber in the field, very good at vulgarizing and explaining the last research papers. At the same time, teaches you how to code, and provides you with working algorithms for interesting projects. If you enjoy video content, don't forget to subscribe to his channel.

Two Minute Papers [Academic]

Another youtube channel that sums a research paper in 2 minutes (duh). Good alternative to stay updated on the state of the art, without bothering with implementation at all.

Papers With Code [Academic and Code]

This website is the ultimate link between the two worlds: for each post, you have the link to the original paper, as well as a list of implementation available on Github. Fully up to date with the state of the art. Use without moderation...