Notes on codes, projects and everything
So apparently Annoy is now splitting points by using the centroids of 2 means clustering. It is claimed that it provides better results for ANN search, however, how does this impact regression? Purely out of curiosity, I plugged a new point splitting function and generated a new set of points.
(more…)After a year and half, a lot of things changed, and annoy also changed the splitting strategy too. However, I always wanted to do a proper follow up to the original post, where I compared boosting to Annoy. I still remember the reason I started that (flawed) experiment was because I found boosting easy.
(more…)While following through the Statistical Learning course, I came across this part on doing regression with boosting. Then reading through the material, and going through it makes me wonder, the same method may be adapted to Erik Bernhardsson‘s annoy algorithm.
(more…)In the previous post, I re-implemented Annoy in 2D with some linear algebra maths. Then I spent some time going through some tutorial on vectors, and expanded the script to handle data in 3D and more. So instead of finding gradient, the perpendicular line in the middle of two points, I construct a plane, and find the distance between it and points to construct the tree.
Recently I switched my search code to Annoy because the input dataset is huge (7.5mil records with 20k dictionary count). It wasn’t without issues though, however I would probably talk about it next time. In order to figure out what each parameters meant, I spent some time watching through the talk given by the author @fulhack.
Implementing a Information Retrieval system is a fun thing to do. However, doing it efficiently is not (at least to me). So my first few attempts didn’t really end well (mostly uses just Go/golang with some bash tricks here and there, with or without a database). Then I jumped back to Python, which I am more familiar with and was very surprised with all the options available. So I started with Pandas and Scikit-learn combo.
Recently I am involved in developing some small modules for a enterprise class website using CodeIgniter (CI). There was no restriction given on which framework should I use for the development and I chose CI as I learned a bit on it (when I was considering whether to shift my personal development project). Of course there are other reasons why I chose to learn CI, for example the superior documentation and screencasts available.
While following through the Statistical Learning course, I came across this part on doing regression with boosting. Then reading through the material, and going through it makes me wonder, the same method may be adapted to Erik Bernhardsson‘s annoy algorithm.
(more…)I used to develop a bot, partly for work, that fetches current latest petrol retail price in Malaysia. The bot was really an experiment, but at the time it worked well. Then a few years later, out of boredom, I revisited the project after finding the telegram bot library is moving towards asyncio. It was great (at least a lot of people rave about it), but also at the same time intimidating, I learned about coroutines and used gevent in the past, but not asyncio itself.
(more…)I didn’t realize that I have been working for 3 weeks until the Labour Day which was a public holiday. Many things happened in these few weeks and I am still struggling to catch up with it. My superior and colleagues have been very helpful and offered me some helpful tutorials and books. I was instructed to build a event scheduler application using codeigniter in the first week and then work on a side project that extends a form using DOM methods and properties.
When one start writting Javascript in patterns like the module pattern, then sooner or later he would want to maintain the state when an event handler is called. The reason I am still using YUI to handle my event handling code is because I like how state can be maintained.