Skip to main content

Library to create and execute computation graphs

Project description

Bodkin is a Python library to create, evaluate, and visualize computation graphs.

Usage / terminology

bodkin.Node - an abstract base class for for a black box that updates its outputs based on its inputs

bodkin.Atom - a Node that is a fundamental building block. An Atom provides a function to update its outputs from its inputs. An Atom is what we would conventionally refer to as a node in a graph.

bodkin.DAG - a directed acyclic graph that encapsulates a set of Nodes and evaluates them in order of dependency. A DAG consists of Node objects, so it may contain Atoms and/or other embedded DAGs.

Any Node's inputs and outputs may be accessed through its properties i and o, resepectively. These properties behave like dictionaries but can be linked together so that setting a output on one Node also sets an input on some other Node.

To learn more see the examples/basic.py script for a basic usage example or pass an object into help(). Feel free to open an issue for further questions.

Installation

Dependencies

Basic functionality only relies on the standard library, but visualizing objects via their show method requires the dot command from graphviz.

From PyPI

Install using pip

python3 -m pip install bodkin

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

bodkin-0.1.1.tar.gz (10.2 kB view details)

Uploaded Source

Built Distribution

bodkin-0.1.1-py3-none-any.whl (10.6 kB view details)

Uploaded Python 3

File details

Details for the file bodkin-0.1.1.tar.gz.

File metadata

  • Download URL: bodkin-0.1.1.tar.gz
  • Upload date:
  • Size: 10.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.11.0

File hashes

Hashes for bodkin-0.1.1.tar.gz
Algorithm Hash digest
SHA256 82f540e745f01b1fb81d54ee0bfa5c55e01dcbafd9451ed845766f55ba2dd113
MD5 4333a7c6936ca8a2e3358e803d4c04ba
BLAKE2b-256 6d65c815bdc3b37ca730f223c7d1e36add1ee498d11f8261847befb3924ae36f

See more details on using hashes here.

File details

Details for the file bodkin-0.1.1-py3-none-any.whl.

File metadata

  • Download URL: bodkin-0.1.1-py3-none-any.whl
  • Upload date:
  • Size: 10.6 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.11.0

File hashes

Hashes for bodkin-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 e461a8f3e06cfeb24632cbf283cee63b7691bd8ffcf8b5899547c4ab395b6730
MD5 d322a9db417526abfe99445c6f6e48ca
BLAKE2b-256 13ec68c0a8d46423f713cf395adf0d2b723bea337afc484961423eeccdefe354

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