Skip to main content

Feature Ordering Module from TabSeq (ICPR 2024)

Project description

TabSeq Feature Ordering

PyPI version Python Versions License: MIT ICPR 2024 Paper GitHub Stars


This module extracts and packages the feature ordering algorithm used in TabSeq (ICPR 2024) as a standalone utility, enabling integration into any tabular deep learning pipeline.

Key Features

  • Variance-based intra-cluster ordering
  • KMeans clustering for feature grouping
  • Weighted global ordering from local cluster orders
  • Minimal dependencies, flexible integration

Installation

pip install tabseq-feature-ordering

Usage

from tabseq_feature_ordering import reorder_features

# Inputs
X_train = ...  # pandas DataFrame of shape (n_samples, n_features)
cluster_size = 5
sort_order = 'descending'  # or 'ascending'

# Output
global_ordering, X_train_reordered = reorder_features(X_train, cluster_size, sort_order)

Parameters

  • X_train: Tabular training data as pd.DataFrame
  • cluster_size: Number of clusters (e.g., 5)
  • sort_order: Intra-cluster sorting order by variance ('ascending' or 'descending')

Output

  • global_ordering: List of column names in reordered order
  • X_train_reordered: DataFrame with reordered columns

Example

import pandas as pd
import numpy as np
from tabseq_feature_ordering import reorder_features

# Example input
X = pd.DataFrame(np.random.rand(40, 80), columns=[f"F{i}" for i in range(80)])

# Run feature ordering
order, X_reordered = reorder_features(X, cluster_size=5, sort_order='descending')

print(order[:10])  # First 10 features in the new order

License

MIT License © 2024 Zadid Habib

Citation

If you use this module, please cite our paper:

Habib, Al Zadid Sultan Bin, Kesheng Wang, Mary-Anne Hartley, Gianfranco Doretto, and Donald A. Adjeroh. "TabSeq: A Framework for Deep Learning on Tabular Data via Sequential Ordering." In International Conference on Pattern Recognition, pp. 418-434. Cham: Springer Nature Switzerland, 2024.


Bibtex

@inproceedings{habib2024tabseq,
  title={TabSeq: A Framework for Deep Learning on Tabular Data via Sequential Ordering},
  author={Habib, Al Zadid Sultan Bin and Wang, Kesheng and Hartley, Mary-Anne and Doretto, Gianfranco and A. Adjeroh, Donald},
  booktitle={International Conference on Pattern Recognition},
  pages={418--434},
  year={2024},
  organization={Springer}
}

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

tabseq_feature_ordering-0.1.4.tar.gz (6.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

tabseq_feature_ordering-0.1.4-py3-none-any.whl (5.2 kB view details)

Uploaded Python 3

File details

Details for the file tabseq_feature_ordering-0.1.4.tar.gz.

File metadata

  • Download URL: tabseq_feature_ordering-0.1.4.tar.gz
  • Upload date:
  • Size: 6.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.2

File hashes

Hashes for tabseq_feature_ordering-0.1.4.tar.gz
Algorithm Hash digest
SHA256 72275d39508661cb3b6f8d1cb6ce7e98694cde68e8747a62cc2e369931ccb54f
MD5 664e1fd55fbcd68195c5f28eabb2f21e
BLAKE2b-256 baceff495fdc1a18811c859aae5820291b38fc6ce29aeefa2ea157b89301e37f

See more details on using hashes here.

File details

Details for the file tabseq_feature_ordering-0.1.4-py3-none-any.whl.

File metadata

File hashes

Hashes for tabseq_feature_ordering-0.1.4-py3-none-any.whl
Algorithm Hash digest
SHA256 3c184d81869fc9fb4462de4a9ae5050972393b5a23398223613d90db0e032eb1
MD5 5f546bfc5f1183893dfed39e062b2b86
BLAKE2b-256 6707f0801dd955b8b1e49401c249dc0a8e8d6fa7c8dd61598f4949884a3afb9d

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page