ARC Core
Adaptive Recursive Consciousness (ARC) Core is a framework for building continual learning AI systems that can learn and reason over time. It features:
- Continual Learning: Real-time learning with LoRA adapters
- Reasoning Engine: Graph-based reasoning and pattern recognition
- Biological Learning: Implements biological learning mechanisms
- Model Agnostic: Works with various transformer architectures
Installation
pip install arc-core
Quick Start
from arc_core import LearningARCConsciousness, ARCCore, ARCTrainer # All aliases for the same class
# Initialize with a base model
arc = LearningARCConsciousness(model_name="gpt2")
# Learn from interactions
arc.learn_from_experience("The sky appears blue due to Rayleigh scattering")
# Generate responses
response = arc.generate("Why is the sky blue?")
print(response)
Features
- Dynamic LoRA Adapters: Automatically adapts to different model architectures
- Reasoning Graph: Maintains a knowledge graph of learned concepts
- Biological Learning: Implements contextual gating and cognitive inhibition
- Persistence: Saves learning progress between sessions
Documentation
For detailed documentation, see ARC Core Documentation.
License
MIT
1 maintainer
Release files for metisos-arc-core 1.2.8
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| metisos_arc_core-1.2.8.tar.gz | 46.8 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| metisos_arc_core-1.2.8-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 91.6 kB
Release files / metisos_arc_core-1.2.8.tar.gz
| Download URL | metisos_arc_core-1.2.8.tar.gz |
|---|---|
| Size | 46.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
fccdc06da902541e7d98702d118488bd9bd34c9326b9d932f6751c83167e3e81
|
|
BLAKE2b-256 checksum How to use checksums |
132448571081977c73e0e744785be6511710939e9f6e47dd489385e2c54f7980
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.12.3
|
Release files / metisos_arc_core-1.2.8-py3-none-any.whl
| Download URL | metisos_arc_core-1.2.8-py3-none-any.whl |
|---|---|
| Size | 44.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
8c625084f6ba3f9fdc7cb6c5947cca700db79c7d9450ead7820ace080525a243
|
|
BLAKE2b-256 checksum How to use checksums |
a9b6f96287b33d596f82617e816fd90b59ff18101c4e5cdcb6233b47a2a999c5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.12.3
|
