Skip to main content

Close-k Classifier

This repository contains code accompanying

Minimizing Close-k Aggregate Loss Improves Classification

Bryan He, James Zou.

We provide a Python 3 implementation using the scikit-learn API, and provide code to reproduce the figures and tables from the paper.

Installation

Our package is available on PyPy, and can be installed using

pip install -i https://pypi.org/project/ closek

You can also install this package by cloning the Github repository, and running

pip install closek

If you want directly use the implementation in your package, you can also copy closek/closek.py into your code.

Usage

An example of how to use our package is shown in test.py.

Generating Results from Paper

The code used for the paper is in experiments. See the README there for more details.

Metadata

Release files for closek 1.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for closek 1.1
File Size Uploaded
closek-1.1.tar.gz 3.4 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for closek 1.1
File Interpreter ABI Platform
closek-1.1-py3-none-any.whl Python 3 none any Details

Total release size: 8.2 kB

Release files / closek-1.1.tar.gz

Download URL closek-1.1.tar.gz
Size 3.4 kB
Tags Source
SHA-256 checksum
How to use checksums
189e564af01c4025676e0192e26e080cd1b39b74c6a71ad62d9c02b67798bbf7
BLAKE2b-256 checksum
How to use checksums
4cf54aba3e2889057bea529344a4ad040050c86942f6af2f15e106dedd6e0d54
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/1.12.1 pkginfo/1.4.2 requests/2.19.1 setuptools/40.4.3 requests-toolbelt/0.8.0 tqdm/4.26.0 CPython/3.7.0

Release files / closek-1.1-py3-none-any.whl

Download URL closek-1.1-py3-none-any.whl
Size 4.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
a63a05cc37b0e68df25fed129f0de7b95c1e1f31f879b022cb64242403eceab5
BLAKE2b-256 checksum
How to use checksums
84164ad8019d008e790ef543ace24dde9b773efa36908dbd4f379e0644b31618
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/1.12.1 pkginfo/1.4.2 requests/2.19.1 setuptools/40.4.3 requests-toolbelt/0.8.0 tqdm/4.26.0 CPython/3.7.0

Release history Release notifications | RSS feed

This release

1.1 This release

2 release files

1.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page