Cattykit

Cattykit (in development) is a framework for implementing artificial intelligence models based on components from the cognitive models of the Fluid Analogies Research Group described in Hofstadter et al 1995. Their models are typically built with similar components, including: a shared workspace, slipnet, coderack, and codelets.

Models solve problems represented by graph-like structures in a workspace which is incrementally edited by micro-agents called codelets that propose, build, and destroy new structures. The slipnet keeps track of concepts relevant to the structures present in the workspace and encourages codelets that continue their construction. The coderack is a stochastic scheduler of codelets, which are selected according to urgency. All processes in each model occur with a degree of randomness mediated by its temperature which (unlike the related concept used in language model decoding and simulated annealing) is a measure of workspace quality constantly adjusted by the model itself.

This repository was initially motivated by the lack of a common software framework for FARG models, which were typically developed as isolated projects, often using different programming languages (some of which are now obsolete). This hinders further research into the models and, despite the promise they show as an approach to symbolic AI, acts as a barrier to their adoption in AI more widely.

Installation

Install with pip:

pip install cattykit

Quick start

Start by installing a model in your environment (this requires pip in the local environment).

import cattykit

# providing a name will install the model from the Cattykit repository
cattykit.install_model("copycat")

print(cattykit.available_models())
# ('cattykit',)

# providing a uri will install a model from your computer or elsewhere
cattykit.install_model("path/to/model")

Once you have installed a model, you can ask it to solve a problem.

import cattykit

copycat = cattykit.load_model("copycat", config={"seed": 1234})

result = copycat.solve("abc -> abd ==> ijk -> ?")

print(result)
# 'ijl'

References

  • Hofstadter, D. R., et al. (1995). Fluid Concepts and Creative Analogies: Computer Models of the Fundamental Mechanisms of Thought. Basic Books. doi:10.5555/218753
  • Mitchell, M (1993). Analogy-Making as Perception: A Computer Model. The MIT Press. doi:10.5555/152203

License

Cattykit is licensed under the GNU Affero General Public License v3.0 or later. See LICENSE for details.

Key dates

PyPI data
Data sourced directly from PyPI's database.
  • Released:
Latest release

1 maintainer

PyPI data
Data sourced directly from PyPI's database.
Avatar for georgeawright from gravatar.com georgeawright

Credits

Author: George Alexander Wright

License

GNU AFFERO GENERAL PUBLIC LICENSE

Requires

Python >=3.12

Tags

cognitive models AI

Classifiers

Report project as malware

Download files

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

Source Distribution

cattykit-0.1.1.tar.gz (50.1 kB view details)

Uploaded Source

Built Distribution

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

cattykit-0.1.1-py3-none-any.whl (40.3 kB view details)

Uploaded Python 3

File details

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

File metadata

  • Download URL: cattykit-0.1.1.tar.gz
  • Upload date:
  • Size: 50.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for cattykit-0.1.1.tar.gz
Algorithm Hash digest
SHA256 cf85c1926c8a6958e4a41a14cce666001584795e3d208e5f72c475aac010526c
MD5 2bf580d5bac8ac48655eb4d84cc4f198
BLAKE2b-256 ed3108d9c8fbb2b574945393b48a5df23f08cb94a8f4fa441ef2451180d7b57b

See more details on using hashes here.

Provenance

The following attestation bundles were made for cattykit-0.1.1.tar.gz:

Publisher: release.yml on georgeawright/cattykit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

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

File metadata

  • Download URL: cattykit-0.1.1-py3-none-any.whl
  • Upload date:
  • Size: 40.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for cattykit-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 98907e5bd66a4bd1f28b1ea513f5a00812250a61335e3eb7de47dbf0acc4798a
MD5 b54e0c08879982ad6f3366d213438180
BLAKE2b-256 1a783cb188ce2ac240a307b331e530775669df22a8437a3b47e5d553e2fa7759

See more details on using hashes here.

Provenance

The following attestation bundles were made for cattykit-0.1.1-py3-none-any.whl:

Publisher: release.yml on georgeawright/cattykit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

0.1.1 This release

2 files

0.1.0

2 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