Machine Learning

• 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…)

• Further Hack on the Multi-Dimensional Approximate Neighbour Search

  Traversing a tree structure often involves writing a recursive function. However, Python isn’t the best language for this purpose. Therefore I started flattening the tree into a key-value dictonary structure. Logically it is still a tree, but it is physically stored as a dictionary. Therefore it is now easier to write a simple loop to traverse it.

  (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…)

• Random notes on Pandas and Scikit-learn

  So I first heard about Panda probably a year ago when I was in my previous job. It looked nice, but I didn’t really get the chance to use it. So practically it is a library that makes data looks like a mix of relational database table and excel sheet. It is easy to do query with it, and provides a way to process it fast if you know how to do it properly (no, I don’t, so I cheated).

  (more…)

Random Posts

• Processing JSON with dask.bag

  Often times, I am dealing with JSONL files, though panda’s DataFrame is great (and blaze to certain extend), however it is offering too much for the job. Most of the received data is in the form of structured text and I do all sorts of work with them. For example checking for consistency, doing replace based on values of other columns, stripping whitespace etc.

  (more…)

• Weekend Code Quiz – Easy Template Parser

  A friend of mine recently posted a screenshot containing a code snippet for a fairly straight forward problem. So after reading the solution I suddenly had the itch to propose another solution that I initially thought would be better (SPOILER: Turns out it isn’t). Then mysteriously I stuck myself to my seat and started coding an alternative solution to it instead of playing Diablo 3 just now.

  (more…)

• Building a simple ASGI Websocket Application

  Getting comfortable to asyncio takes a bit of practice, so I revisited a practice project I did when I was working for my previous company. Suppose I want to build a very simple websocket application, without use of any web application library/framework. In order to keep it simple, I also opt to just build the frontend with minimal setup (just plain ES6 without webpack/vite).

  (more…)

• Statistical Analysis for Social Audit Project

  This is the formal draft of my statistical analysis report for the social audit project previously mentioned here. As the project is public by nature, I am cross-posting here for own reference.

  (more…)

• My evil evil form and database library

  Writing a usable form and database library has always been a painful experience. So why bother re-inventing the wheel when there are so many to choose from already? I am writing one mostly for learning purpose. After numerous attempts, I finally get my form and database library in shape. It is nowhere complete, but nor it is perfect, but it is currently the implementation that is closest to my original design. I will keep working on it so it can be used in my personal projects in the future.

  (more…)