Deep Learning in Clojure from Scratch to GPU: Learning a Regression

dragan.rocks

120 points·dragandj··10 comments

The table of content of the series is at:

https://dragan.rocks/articles/19/Deep-Learning-in-Clojure-Fr...

If you want to be notified about the book, please subscribe to the newsletter at the bottom of the page.

Are you planning on including some of your other content besides the Deep Learning sections? I personally would like to see the 'Linear Algebra Refresher' posts included.

That will be another book "Vector, Matrices, and Linear Algebra for Programmers". Expect early draft releases in May/June.

Thank you for the series!

Done!

... did you say there's a book coming? :-)

Very nice work. Clojure's REPL support is orders of magnitude better than Python's. Hopefully Clojure will be able to compete with Python in the near future for data science/ML work.

Incanter[1] has been around for a long time and I recently discovered kixi.stats[2]. Links to some other resources can also be found on the Skymind wiki.[3]

[1] http://incanter.org/

[2] https://cljdoc.org/d/kixi/stats/0.5.0/doc/readme

[3] https://skymind.ai/wiki/clojure-ai

kixi.stats is very nice, but Incanter has been unmaintained for a few years.

Interesting. Hope it will run on mac (should be). Got a ti1080 and a vega 56 to compare.

It does run on a mac, but for this particular release of jcuda 10.0 that I'm using, the mac binaries are not bundled, so you'll have to build them yourself. I managed to build it on a mac, so it will be included in the next release of JCuda that matches CUDA 10.1, we are just waiting JCuda author to update the bindings. You can upvote that issue at https://github.com/jcuda/jcuda-main/issues/29

Reply on news.ycombinator.com