Metadata-Version: 2.4
Name: cognitive-cores
Version: 1.0.3
Summary: Universal Cognitive Architecture Framework for AI Models
Home-page: https://github.com/Volgat/nexus-standardisation
Author: Mike Amega
Author-email: Mike Amega <contact@amewebstudio.com>
License: Proprietary
Project-URL: Homepage, https://github.com/Volgat/nexus-standardisation
Project-URL: HuggingFace, https://huggingface.co/amewebstudio/cognitive-core
Project-URL: Documentation, https://github.com/Volgat/nexus-standardisation#readme
Project-URL: Issues, https://github.com/Volgat/nexus-standardisation/issues
Keywords: cognitive-ai,neural-network,transformer,llm,world-model,multimodal,huggingface,pytorch
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: Other/Proprietary License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: torch>=2.0.0
Requires-Dist: transformers>=4.35.0
Requires-Dist: datasets>=2.14.0
Requires-Dist: huggingface_hub>=0.19.0
Requires-Dist: accelerate>=0.24.0
Provides-Extra: dev
Requires-Dist: pytest>=7.0.0; extra == "dev"
Requires-Dist: black>=23.0.0; extra == "dev"
Requires-Dist: ruff>=0.1.0; extra == "dev"
Provides-Extra: training
Requires-Dist: wandb>=0.15.0; extra == "training"
Requires-Dist: tensorboard>=2.14.0; extra == "training"
Provides-Extra: vision
Requires-Dist: torchvision>=0.15.0; extra == "vision"
Requires-Dist: pillow>=9.0.0; extra == "vision"
Provides-Extra: audio
Requires-Dist: torchaudio>=2.0.0; extra == "audio"
Requires-Dist: librosa>=0.10.0; extra == "audio"
Provides-Extra: all
Requires-Dist: wandb>=0.15.0; extra == "all"
Requires-Dist: tensorboard>=2.14.0; extra == "all"
Requires-Dist: torchvision>=0.15.0; extra == "all"
Requires-Dist: pillow>=9.0.0; extra == "all"
Requires-Dist: torchaudio>=2.0.0; extra == "all"
Requires-Dist: librosa>=0.10.0; extra == "all"
Dynamic: author
Dynamic: home-page
Dynamic: requires-python

# COGNITIVE-CORES Framework

> 🧠 Universal Standard for Cognitive Architectures by Ame Web Studio

**Cognitive-Cores** is a robust, agnostic framework designed for building advanced cognitive AI models. It provides a standardized interface for integrating Vision, Language, Audio, World Modeling, and Multimodal capabilities into a unified system.

## 🚀 Installation

### Option 1: Via Pip (From PyPI)

```bash
pip install cognitive-cores
```

### Option 2: Via Pip (From GitHub)

```bash
pip install git+https://github.com/Volgat/nexus-standardisation.git@cognitive-core
```

### Option 3: Via HuggingFace

```bash
pip install git+https://huggingface.co/amewebstudio/cognitive-core
```

### Optional Dependencies

```bash
pip install "cognitive-cores[vision]"    # For Vision Models
pip install "cognitive-cores[audio]"     # For Audio Models
pip install "cognitive-cores[training]"  # For Training Tools (WandB, etc.)
pip install "cognitive-cores[all]"       # Full Installation
```

## 🛠️ Usage

### Loading Models regarding Cognitive Finetuning

To finetune a model built with **Cognitive-Cores** (like NEXUS-LPOL) from HuggingFace, use the standard `AutoModel` interface with `trust_remote_code=True`.

```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
from cognitive_core import CognitiveTrainer, CognitiveTrainingConfig, prepare_dataset

# 1. Configuration
model_id = "amewebstudio/nexus-lpol-v3"  # Example Model

# 2. Load Tokenizer & Model
# trust_remote_code=True is essential to load the custom cognitive architecture
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    trust_remote_code=True, 
    torch_dtype=torch.float16,
    device_map="auto"
)

# 3. Training Setup
config = CognitiveTrainingConfig(
    output_dir="./nexus-finetuned",
    num_train_epochs=3,
    per_device_train_batch_size=4
)

# 4. Initialize Trainer
trainer = CognitiveTrainer(
    model=model,
    args=config,
    train_dataset=my_dataset, # Prepare your dataset using prepare_dataset helper
)

# 5. Start Finetuning
trainer.train()
```

## 🧩 Core Capabilities

The framework provides a suite of standardized, reusable modules designed for high-performance cognitive modeling.

*   **Advanced Normalization & Encoding**: Optimized implementations for stability and long-context handling.
*   **Attention Mechanisms**: Efficient attention layers supporting extensive context windows and multimodal fusion.
*   **Memory Systems**: sophisticated short-term, long-term, and episodic memory modules.
*   **World Modeling**: Components for simulating and predicting states across physical, social, and abstract domains.
*   **Internal State Management**: Modules for handling agentic internal states, drives, and cohesion.
*   **Multimodal Integration**: Universal latent space mapping for seamless alignment of text, audio, and visual data.
*   **Neurogenesis**: Dynamic architectural adaptation capabilities.

## 📄 License

**PROPRIETARY - ALL RIGHTS RESERVED**
Copyright © 2026 Mike Amega - Ame Web Studio
