https://github.com/ysenarath/octoflow

Streamlining machine learning tracking for seamless experiment management.
https://github.com/ysenarath/octoflow

Last synced: 2 days ago
JSON representation

Streamlining machine learning tracking for seamless experiment management.

• Host: GitHub
• URL: https://github.com/ysenarath/octoflow
• Owner: ysenarath
• License: mit
• Created: 2023-11-25T15:45:32.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2024-04-12T05:21:24.000Z (7 months ago)
• Last Synced: 2024-04-12T10:24:50.218Z (7 months ago)
• Language: Python
• Homepage: https://octoflow.readthedocs.io
• Size: 282 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE
  • Authors: AUTHORS.md

Awesome Lists containing this project

README

        
# OctoFlow

Streamlining machine learning tracking for seamless experiment management.

## Features

* Feature 1

* Feature 2

* ...

## Development

To set up [hatch] and [pre-commit] for the first time:

1. install [hatch] globally, e.g. with [pipx], i.e. `pipx install hatch`,

2. optionally run `hatch config set dirs.env.virtual .direnv` to let [VS Code] find your virtual environments,

3. make sure `pre-commit` is installed globally, e.g. with `pipx install pre-commit`,

4. run `pre-commit install` to install [pre-commit].

A special feature that makes hatch very different from other familiar tools is that you almost never

activate, or enter, an environment. Instead, you use `hatch run env_name:command` and the `default` environment

is assumed for a command if there is no colon found. Thus you must always define your environment in a declarative

way and hatch makes sure that the environment reflects your declaration by updating it whenever you issue

a `hatch run ...`. This helps with reproducability and avoids forgetting to specify dependencies since the

hatch workflow is to specify everything directly in [pyproject.toml](pyproject.toml). Only in rare cases, you

will use `hatch shell` to enter the `default` environment, which is similar to what you may know from other tools.

To get you started, use `hatch run cov` or `hatch run no-cov` to run the unitest with or without coverage reports,

respectively. Use `hatch run lint:all` to run all kinds of typing and linting checks. Try to automatically fix linting

problems with `hatch run lint:fix` and use `hatch run docs:serve` to build and serve your documentation.

You can also easily define your own environments and commands. Check out the environment setup of hatch

in [pyproject.toml](pyproject.toml) for more commands as well as the package, build and tool configuration.

## Credits

This package was created with [The Hatchlor] project template.

[The Hatchlor]: https://github.com/florianwilhelm/the-hatchlor

[pipx]: https://pypa.github.io/pipx/

[hatch]: https://hatch.pypa.io/

[pre-commit]: https://pre-commit.com/

[VS Code]: https://code.visualstudio.com/docs/python/environments#_where-the-extension-looks-for-environments