Profile cover photo
Profile photo
Ivo Flipse
283 followers
283 followers
About
Ivo's interests
View all
Ivo's posts

Post has shared content
About time they updated Contacts!
Now to go fix those annoyances in my address book, such as my parents being merged into one...
Big launch for the team today!  We've hugely improved Google Contacts.  The once tedious task of "managing" your contacts on Google has been made a lot easier, more intelligent... and more beautiful...  The launch is still a 'preview', so go to contacts.google.com/preview, check it out, and let them know what you think!

Post has shared content
Very funny indeed! Bow to Jibbers!

Post has shared content
Looking forward to +Andrej Karpathy​​'s course! I probably fall right between the two target audiences so it will be interesting to see whether I'll be able to understand all of it.
I'm working on class notes / assignments for our upcoming ConvNet class CS231n (http://vision.stanford.edu/teaching/cs231n/). By far the hardest challenge I'm coming across is deciding on my target audience:

I want the notes to be broadly useful even to people who might not have a lot of ML background, but I'm now realizing that as a result they seem way too dumbed down, leaving a more experienced reader frustrated. Conversely I want to throw in a lot of nuggets of interesting subtle problems/issues and details, but many of them require quite a lot of background expertise and intuitions that sometime seem outside of the scope of the class.

For example, I spend 2 pages motivating the SVM loss and then gloss over the (more complex) Softmax loss that involves logs, exps and probabilities in 2 paragraphs. Or I discuss the k-Nearest Neighbor in 3 pages and then next section plunge into SGD and talk about convex problems and optimization dynamics. It's all inconsistent and giving me quite the headache.

What I'm most worried about is a problem that I sometimes see in other classes: they try to target everyone and end up targeting noone.

Post has shared content
We're looking for an experienced front-end/UX designer & developer at Clinical Graphics. If you're all about that interface, give us a call!

https://www.clinicalgraphics.com/file/185/VacatureFront-endDeveloper.pdf

Post has shared content
So, I made a short weekend trip to Netherlands last weekend in continuing my tradition of trying to explore a new country every weekend during my Finland work visit.

I'm pretty damn glad and excited that +Ivo Flipse came over to meet me from over 100 kilometers just to meet me and +Amber Prevoo.


We spoke about lots of diverse topics, from MOOCs to work stuff and yeah, a little bit about +Stack Exchange too.

Sadly, I had only couple of hours to catch up with him, but that was a well done Sunday morning. 

Cheers!
Photo
Photo
2014-10-18
2 Photos - View album

Post has attachment
I'll just drop this here. 

Very nifty solutions to display famous paintings using the color palette of another famous painting.

Post has shared content
This should make it a lot easier to share my analysis with others and let them run it against their own data.

It sounds like it will also bundle all the dependencies, like numpy and matplotlib, so it should be easy to get going for non-programmers. Hopefully it'll run more complicated libraries like OpenCV or ITK as well.

Now I just need to figure out how to use Google Drive as a storage layer or whether they'll offer some other way to give you access to your data without users having to install their own database.

I'm really liking where this is going!
Over the last year I've worked with an awesome folks (+Kester Tong  +Mark Sandler, +Corinna Cortes , +Matthew Turk,  +Gideon Mann  +Arnaud Sahuguet, +Adam Berenzweig) to understand how people collaborate on data analysis and to build better tools to support them. Yesterday, with the help of +Fernando Perez and +Wes McKinney we revealed this work at PyCon APAC.

We've created an interactive, collaborative analytics tool by integrating Google Docs, Chrome, and IPython. You can open a notebook from Drive. You can share notebooks like you would share a Google Doc. You can comment and edit collaboratively, in realtime. There is zero setup, because all the computation happens in Chrome. You can even quickly and easily package your analytics pipeline into a GUI for folks that don't want to program. In effect, you can go from zero to analytics with little impedance.  

What's even better is that you can build on our work. It will all be open source on top of public Google APIs. We'll have a larger Google Research blog post about this work when we release the code and the Chrome application.
Photo

Post has shared content
This seems useful, combined with Docker and Vagrant I could have a nice way to package a working version of my scientific stack

Post has shared content
Just what I needed! This bit me when I tried to reinstall my GFs laptop. Would have saved me quite some time downloading updates...

Post has shared content
I need to talk to that Google engineer that created the paw recognition algorithm!
Who says the underdog can’t be employee of the month? New Google technology proves to increase productivity by leveraging your company’s most underutilized asset.
Wait while more posts are being loaded