141–150 of 194 posts
Mobile photo scavenger hunt: http://www.playlookit.com One topic a day, one pic per day.
This is a cool project, throw in an iPhone app and improve the design/UI a bit and I can see it getting even more traction.
Masters project on using bandit algorithms for optimising CTRs on website content. Also involves some search engine / text mining / dimensionality reduction stuff. A little bit of messing around with Android SDK too...
My Django hosting service. I looked at Heroku and thought, "I want that for Python." Yesterday, I got most of the http request path finished. There is some node.js work left, and I still need to do some more work on Varnish. I'm hoping to get to the Postgresql stuff next week, then on to the website and API...
http://search.cpan.org/~michael/Class-Declarative-0.06/lib/C... A declarative framework for Perl that lets you set up complex data structures and code in an easy-to-read format. A complete working example GUI program, using a not-yet-published set of semantic classes: use Class::Declarative qw(Wx::Declarative); dialog (xsize=250, ysize=110) "Wx::Declarative dialog sample" field celsius (size=100, x=20, y=20) "0" butt…
I did something similar to that once -- a program that compiled GUI descriptions into a runable Python class, which could then be subclassed to provide functionality. The syntax looked like this: window @MyBigWindow "My Big Window" { menuBar { menu "File" { menuItem @New "New" menuItem @Open "Open..." menuItem @Save "Save" menuItem @Exit "Exit" } menu "Edit" { menuItem @Copy "Copy" menuItem @Cut "Cut" menuItem @Paste…
I've got big plans, and since it's not expected to pay any bills for a while, I can afford to think things through sufficiently. I just hope I won't drop it entirely.
It's been fun so far, though.
Earlier quoted context omitted.
I did something similar to that once -- a program that compiled GUI descriptions into a runable Python class, which could then be subclassed to provide functionality. The syntax looked like this: window @MyBigWindow "My Big Window" { menuBar { menu "File" { menuItem @New "New" menuItem @Open "Open..." menuItem @Save "Save" menuItem @Exit "Exit" } menu "Edit" { menuItem @Copy "Copy" menuItem @Cut "Cut" menuItem @Paste…
There's some similar stuff for wx in Perl and Python, too - I've just been taking the time to (try to) be systematic about making things fast and easy to specify. I really, really get tired of coding all that stuff by hand every damn time - it's one of the major things that made me get out of GUI work in the first place, back in the Stone Age. I've got big plans, and since it's not expected to pay any bills for a whi…
Absolutely. Having sensible and configurable defaults for everything is the way to go.
Started yesterday, want to finish it tonight.
I used to not want to leave my desk for lunch because I'd need to come back if the market moved. I wrote a script that would tell me if the market moved past a certain threshold and would text me to come back.
I was sort of shocked that it didn't exist in the wild (or at least wasn't easy to find) so I decided to see if I could make a web app of it.
Hoping to "launch" in the next 2 weeks. Would love testers.
Earlier quoted context omitted.
That sounds fascinating. How did you come up with the algorithms to use?
I experimented with existing text classification algorithms for an author identification project I was doing for fun. What I'm currently using is somewhere between KNN and SVM, but I'm not done tweaking it yet. I'm also working on boosting results using different feature sets.
Masters project on using bandit algorithms for optimising CTRs on website content. Also involves some search engine / text mining / dimensionality reduction stuff. A little bit of messing around with Android SDK too...
You're doing better than 90 percent of everybody else just by expressing this as a bandit problem rather than an AB testing problem.
If anyone is interested in a technical intro to the setting there is a set of slides from John Langford at Yahoo Research (many good and standard reference papers cited in it): http://hunch.net/~exploration_learning/
A/B testing could be thought of as a sort of epsilon-Greedy strategy (particularly if such testing is carried out at regular intervals initially). While not enjoying the optimality characteristics of other algorithms, such an approach can in fact outperform in many practical cases :)