Watching the universe and being fascinated by it is the one thing that all humans, that have ever lived, have in common. Even today, knowing as much as we do, it still is an extremely humbling and wonderful experience.
Large Synoptic Survey Telescope
Unfortunately, until now we could only watch a tiny part of the night sky at once, which was kind of limiting. Fortunately, those times are about to change - with the introduction of the new Large Synoptic Survey Telescope, located in north-central Chile.
Please do not get too hyped yet, as it will take another 3 to 4 years before the telescope is fully operational. The goal of the project is to perform a 10-year survey, collecting 200 petabytes (!) of data.
That will address some of the most pressing questions about the structure and evolution of the universe and the objects in it.
Not to get into too many details and technical stuff, the main difference between LSST and the standard telescopes is the field of view. Having an extremely wide field of view, the LSST will be able to survey the entire available sky in only three nights!
LSST will be discovering up to 100 times more astronomical sources that vary in the night sky than we've ever known.
You might wonder, what does it have to do with TEONITE?
So, collecting all this data is only the first step. The second step is actually understanding WHAT we’re seeing and this is where it gets interesting. But before we get into this, I need to say a little bit about KAGGLE.COM (you will understand why in just a moment).
What is Kaggle?
As stated in the Getting Started section, Kaggle:
is an AirBnB for Data Scientists – this is where they spend their nights and weekends. It’s a crowd-sourced platform to attract, nurture, train and challenge data scientists from all around the world to solve data science, machine learning and predictive analytics problems. It has over 536,000 active members from 194 countries and it receives close to 150,000 submissions per month.
Kaggle enables data scientists and other developers to engage in running machine learning contests, write and share code, and to host datasets.
In short, you can either join or host a competition in any data science matter you find interesting. Cool stuff right?
TEONITE and PLAsTiCC Astronomical Classification
PLAsTiCC Astronomical Classification, hosted by the University of Toronto, it is one of the numerous competitions you can find on Kaggle.com, but it has a special meaning for us, as TEONITE is a part of the event!
In theory, all sounds simple:
- You join, receive access to the training data set and your task is to help prepare to classify the data from this new survey.
- You submit your algorithm, and its being put to test on a very large test set.
Unfortunately, it is not as simple as it sounds...
Currently there are 659 teams competing, with TEONITE on 83 place (we used to be on 12th position, but the competition is really fierce and it is only a side project for us).
Competition will be concluded in about 3 weeks, so be on the lookout for the new blog post, with the summary of the competition and a bit more details about the methods we used to classify data. In summary, more technical stuff coming soon!