JoinBoost: In-Database Tree-Models over Many Tables
Project description
JoinBoost: In-Database Factorized Tree-Models
JoinBoost the first In-DB factorized learning system for tree-based models.
The technical report is in the /technical.
Reproducibility
We note that some feataures discussed in the paper (e.g., inter-query parallelism, DP) are not implemented in the main codes for reliability concerns. To reproduce the experiment results from the paper, we include the prototype codes for JoinBoost under /proto folder, which includes the JoinBoost codes and Jupyer Notebook to train models over Favorita. The Favorita dataset is too large to store in Github. They can be found in dropbox.
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
joinboost-0.0.11.tar.gz
(16.3 kB
view hashes)
Built Distribution
joinboost-0.0.11-py3-none-any.whl
(18.3 kB
view hashes)
Close
Hashes for joinboost-0.0.11-py3-none-any.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 0637484a11f139e77d9ff364c8ca067c7d79ebd9c1c5d63182c03938e7f4acba |
|
| MD5 | b8144f8e7dbffcde5b86238d754ead19 |
|
| BLAKE2b-256 | 9155853a15df015e493d5437b4b792fc7aec3c724da44b73b98a3ed80c64f844 |
