Using q in Machine Learning with Neural Network and Clustering Examples

4 Apr 2017 | , , ,
Share on:

 

Expert kdb+ programmer, and algorithmic quantitative analyst, Mark Lefevre, who is based in Tokyo, recently gave a couple of talks about using high-performance machine learning with kdb+ at the Kx Community Tokyo Meetup.

His first talk was called  “Using Q to Read Japanese.” It focused on utilizing neural networks and how supervised learning can be used in q to teach a machine to recognize Japanese characters from handwritten images. Neural networks are used in a variety of real-world applications including character recognition, object recognition, image compression, asset path prediction, medicine and self-driving vehicles.

Here is Mark’s presentation onUsing Q to Read Japanese:”  Machine Learning in q

His second and most recent talk, “Kx for Wine Tasting,”  focused on utilizing the k-means clustering algorithm and how unsupervised learning can be used in q to teach a machine to appreciate, well, at least recognize, different types of wine! K-means clustering enjoys broad applications in computer vision, vector quantization, marketing, finance and as a pre-processing step for subsequent machine learning algorithms.

Here is Mark’s presentation on “Kx for Wine Tasting: Kx for Wine Tasting

SUGGESTED ARTICLES

ML and kdb+

Machine Learning Toolkit Update: Cross-Validation and ML Workflow in kdb+

23 Jul 2019 | , ,

The Kx machine learning team has an ongoing project of periodically releasing useful machine learning libraries and notebooks for kdb+. This release relates to the areas of cross-validation and standardized code distribution procedures for incorporating both Python and q distribution. Such procedures are used in feature creation through the FRESH algorithm and cross-validation within kdb+/q.

kx and machine learning

Machine Learning Toolkit Update: Multi-parameter FRESH and updated utilities

25 Apr 2019 | , ,

This latest toolkit release, is the first in a series of planned releases in 2019 that will add updates to the functionality of the FRESH (Feature Extraction based on Scalable Hypothesis tests) algorithm and the addition of a number of accuracy metrics, preprocessing functions and utilities. In conjunction with code changes, modifications to the namespace structure of the toolkit have been made to streamline the code and improve user experience.