jghrng

It gets unethical pretty quickly as it is produced at industrial scale.

I'm using Ctrl+S and Ctrl+Q for interrupting tail -f, but that's missing the search feature of course.

Wow, thanks a bunch for that hint!

Don't forget the missing zoom buttons that make it almost impossible to use the app with one hand only...

On a related historical note, this can happen if you don't pay the chiselers building your church: http://www.holidaycheck.de/data/urlaubsbilder/images/41/1163... (part of http://en.wikipedia.org/wiki/Freiburg_Minster).

I did exactly the same and I unfortunately also really got used to the "go" prefix by now.

Here's the link from TFA: http://erratasec.blogspot.de/2012/08/who-will-fight-for-me.h...

Wow, I never heard of doing annual mammograms, or even "base-line screening" (!). Wouldn't that actually count as personal injury?

EDIT: Continued to read the actual article -- I didn't know about the mammography madness in the US. AFAIK doing X-Rays or CTs without reasonable suspicion is considered personal injury here (in Germany).

> As an aside it's worth noting that as this library is GPL3 it means you can't use this code in any non-GPL product.

Well you can always simply ask the author -- if he has got the copyright, he can grant you permission.

All that being said, anyone care to weigh in on the pros/cons of RBMs vs something like a contractive autoencoder?

For some problems, it may be nice to have a generative model as offered by RBMS (although Rifai et al. published a sampling method for contractive auto-encoders recently: http://icml.cc/2012/papers/910.pdf). I feel like with RBMs, you can design models which incorporate prior knowledge more "easily" (you may end up with pretty complex models...), e.g. the conditional RBM, the mean-covariance RBM or the spike & slab RBM. Additionally, there's the deep boltzmann machine that consists of multiple layers that are jointly trained in an RBM-like fashion.

Auto-encoders are straightforward to understand and implement. With contractive terms or denoising, the are powerful feature extractors as well.

But as you already noted, if you "just" want to have a good classifier, I think it pretty much boils down to personal preference since you're going to spend some effort on making these techniques work well on your problem anyway.

I'd recommend Theano and the accompanying Deep Learning Tutorials at http://deeplearning.net/tutorial/

I think you'll need the female voice. No idea how to change that, though.

EDIT: Try the US version: http://translate.google.com.mx/?hl=en#auto/en/filled%20with

While looking cool, it seems to me like this catches a pedestrian's attention more than the one of the drivers. If pedestrians are distracted from traffic right before crossing, this would make crosswalks less safe.

It should be more useful (but more expensive) to lighten up the actual crosswalk signs on the road as soon as pedestrians are approaching.

I like ikiwiki a lot because it plays nicely with git and there are many useful plugins: http://ikiwiki.info/

Cool, I just found out it works for beamer, too. Having instant preview for presentations is great, as I often find myself moving figures around all the time.

Great work!