annoy

• Regression with 2-means clustering Annoy (non-scientific)

  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…)

• Regression with Annoy (cont’d)

  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…)

• Quick experiment, boosting vs annoy

  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…)

• Approximate Neighbour Search in Multiple Dimensions

  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.

  (more…)

• Re-implementing Approximate Nearest Neighbour Search

  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.

  (more…)

• Information Retrieving with ….. a lot of libraries

  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.

  (more…)

Random Posts

• Picking Up Category Theory (and eventually Haskell)

  I don’t quite remember when did I first heard about Category Theory, but the term stuck in my head for quite a while. Eventually I attempted to start looking for tutorials on the topic, but it is hard to find one that I actually understand. Most of them are either leaning too much to the Mathematics side, or too much to the Programming side.

  (more…)

• Setting Up Virtuoso for Redland Under Ubuntu

  Just a quick update to the previous post, the virtuoso storage engine works with redland provided the required packages are properly installed (yes, yes, yes, I know I haven’t release my PHP OO wrapper for Redland). Now that the package is installed, we need to do some configuration so that Redland can use it.

  (more…)

• Resource Definition Framework

  As the name implies, Resource Definition Framework, or RDF in short, is a language to represent information about resources in world wide web. Information that can be represented is mostly metadata like title (assuming the resource is a web-page), author, last modified date etc. Besides representing resource that is network-accessible, it can be used to represent things that cannot be accessed through the network, as long as it can be identified using a URI.

  (more…)

• Spiraling Number

  Back then in college, we were given a lot of programming practices. These questions usually shows a desired output format, and we were required to write a program to print out the exact thing. Usually it involves printing a matrix of numbers, or symbols etc. For these problems, usually a loop structure or two should solve the problem.

  (more…)

• Blending ACL and FSM-based Workflow

  After reading through the documentation, I find that the role based ACL and work flow can be more tightly integrated. Therefore I made all the transaction into many FSMs and my work flow component now consists of one work flow library and one work flow management model. As I am going a more normalized design (I use denormalized design in work as it deals with a lot of documents, however for a small project like mine, a denormalized design should do well).

  (more…)