search

aws / random-cut-forest-by-aws

An implementation of the Random Cut Forest data structure for sketching streaming data, with support for anomaly detection, density estimation, imputation, and more.
https://github.com/aws/random-cut-forest-by-aws
Apache License 2.0
206 stars 33 forks source link

version is 3.6.0 #378

Closed sudiptoguha closed 1 year ago

sudiptoguha commented 1 year ago

Description of changes: Spurred by recent issue of floating point based exception, this version adds a number of checks and validations of state conversions. The code for a tree was spread across multiple classes since at some point the code supported multiple versions of trees (compact/non-compact/double precision) etc. The code is refactored. As a consequence it would be significantly easier to have an ensemble forest where each tree operates on a different transform of the input (possibly over different number of dimensions), or to have labels associated with points in a tree leading to a classification or semi-supervision capabilities.

The change to version 3.6 also resolves a regression in 3.5.1 as in issue #381

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

amitgalitz commented 1 year ago

Should we change the version here to 3.6.0-SNAPSHOT so its available on maven as a -SNAPSHOT before official 3.6.0 release.

sudiptoguha commented 1 year ago

Good idea. Let me push a change. We would have to also update the forecast horizon changes.