14 comments

[ 2.6 ms ] story [ 31.7 ms ] thread
Jacques Mattheij (hn:jacquesm) has a nice listing of free datasets that was posted here years ago.

http://jacquesmattheij.com/Free+Public+Data+Sets

If you know of any others, please post them.

Anyone know of a dataset of people, like front facing photos of people's upper bodies with no clothes on. I've been meaning to work on a cv method for bodyfat % estimation, given a series of progression photos of a person. I have a rough idea of how to do it, but I need a training set.
Well maybe you could do an Ask HN and request those images. And maybe you would get one such selfie from every HNer. And maybe eventually, with therapy and drugs, you would be able to put the horror behind you and get on with your life, but you might never be able to read HN again without flashbacks. Be careful what you wish for.
As a student @ UC Irvine I can tell you top notch talent leads most of their coursework
As a student @ UC San Diego with the ability to take a few courses at Irvine, are there any specific courses or prof. you recommend looking into?
I'm no longer a student there, but Shannon Alfaro was great at anything parser/language design related, and Gopi Meenakshisundaram was great for graphics. Eric Mjolsness is probably brilliant but everything I took from him was very dense.
I was viva'd in 1998 an yea! verily, I used Iris and Abalone (and 10 others) in my thesis. But why, young padawan? why?

Coz they worked.

16 years of industrial machine learning research later...

The world is not like the UCI repository

Getting the data into the form that the analysis algorithm can parse, asking the question... that is the work. Ok, also dealing with the answer - that too.

Data munging may be time consuming, but it is hardly difficult. The real challenge in ml is still the algorithms and theory.
No, I disagree.

Algorithms can be selected, it's not hard to learn which technique is suited to which situation.

What theory are you talking about? What do we have - Kearn's contributions on COLT and the descendant's there of (Michael Kearns is a great fella btw, and COLT is super - but so 1990's), or Vapnik and Structural Risk Minimisation?

It's not really physics is it?

All the data munging in the world will not make progress of any sort in improving the state of ML/AI. All advances come from better algorithms and understanding of associated mathematics.

Applying techniques to new problems involves data munging, and that is an important and useful task, but it is not difficult in the same way that doing new algorithmic work is, nor does it advance the state of the art.

I still think you are wrong.

There is no harm in developing a new algorithm, or coming up with new stories as to how it works, but to say that "data munging" will not advance the state of the art is to exclude the art! It's like saying that unless someone comes up with new physics it's impossible to build better spaceships.

And if Nasa were to build a better spaceship would that advance the state of the art in spaceship building?