Skip to main content

More Faster Object Classification Training Framework

Project description

Language: 🇺🇸 🇨🇳

«ZCls2» is a more faster classification model training framework



Documentation Status

Table of Contents

Background

After nearly one and a half years of development, ZCls has integrated many training features, includes configuration module, register module, training module, and many model implementations (resnet/mobilenet/senet-sknet-resnest/acbnet-repvgg-dbbnet/ghostnet/gcnet...) and so on. In the development process, it is found that compared with the current excellent classification training framework, such as apex, the training speed of ZCls is not outstanding.

In order to better improve the training speed, we decided to develop a new training framework ZCls2, which is implemented based on apex and provides more friendly and powerful functions. In the preliminary implementation, it can be found that ZCls2 improves the training speed by at least 50% compared with ZCls. More functions are being added.

Installation

See Install

Usage

See Get started

Maintainers

  • zhujian - Initial work - zjykzj

Thanks

Contributing

Anyone's participation is welcome! Open an issue or submit PRs.

Small note:

License

Apache License 2.0 © 2022 zjykzj

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

zcls2-0.5.0.tar.gz (52.8 kB view details)

Uploaded Source

Built Distribution

zcls2-0.5.0-py2.py3-none-any.whl (79.1 kB view details)

Uploaded Python 2 Python 3

File details

Details for the file zcls2-0.5.0.tar.gz.

File metadata

  • Download URL: zcls2-0.5.0.tar.gz
  • Upload date:
  • Size: 52.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.8.3 requests/2.28.1 setuptools/63.2.0 requests-toolbelt/0.9.1 tqdm/4.64.0 CPython/3.7.4

File hashes

Hashes for zcls2-0.5.0.tar.gz
Algorithm Hash digest
SHA256 1a79969f4ce1246dee6da1bc70dc2c730b94f5706d2fbb54317077ad907cae92
MD5 5578166ffd292b83a6c39edda956e9d0
BLAKE2b-256 94ab8256614dca7e36e77145884bab27b61b82784542a04d26eab6c9dfbdfceb

See more details on using hashes here.

File details

Details for the file zcls2-0.5.0-py2.py3-none-any.whl.

File metadata

  • Download URL: zcls2-0.5.0-py2.py3-none-any.whl
  • Upload date:
  • Size: 79.1 kB
  • Tags: Python 2, Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.8.3 requests/2.28.1 setuptools/63.2.0 requests-toolbelt/0.9.1 tqdm/4.64.0 CPython/3.7.4

File hashes

Hashes for zcls2-0.5.0-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 31106819b5a9324e18724a36421d88a96d8e1c199c9d709634bf9c22b658bbe1
MD5 cb5f5d27287a12cdfee4ee69600b9408
BLAKE2b-256 6ac1667fb1e8efe8a089faacb6e782f6e3e27cf65720d92fae04872f6b7edb4d

See more details on using hashes here.

Supported by

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