akali 0.3.1b0

AKALI library for language model augmentation and interfaces

AKALI: Aggressive Knowledge Augmenter and Language Interface

AKALI is a powerful library for language model augmentation and interfaces, designed to enhance the capabilities of AI models through strategic data augmentation and efficient task management.

Features

• Language Interface: Easily integrate and interact with various language models.
• Knowledge Augmentation: Improve model performance through intelligent data augmentation strategies.
• Task Management: Flexible system for handling different NLP tasks.
• CLI Tool: Command-line interface for easy usage and integration.
• Customizable: Extend functionality with custom tasks and augmentation strategies.

Installation

pip install akali

Quick Start

Using the CLI tool Run a language interface service directly

akali run --model alierenak/gemma-7b-akali --task EntitySentimentReasoner --host 0.0.0.0 --port 8000

Make a prediction

akali predict --model alierenak/gemma-7b-akali --task EntitySentimentReasoner --user-message "Turkcell hiç güzel çeken bir hat değil o yüzden Vodofone'u tercih ediyorum hem de daha ucuz"

Using as a Python Library

from akali import LanguageInterface
li = LanguageInterface.load_model("alierenak/gemma-7b-akali")

# Set the task
li.set_task("EntitySentimentReasoner")

# Make a prediction
result = li.predict(system_text=None, user_message="Turkcell hiç güzel çeken bir hat değil o yüzden Vodofone'u tercih ediyorum hem de daha ucuz")
print(result)

Advanced Usage

Custom Task Creation

from akali import LanguageInterface
from pydantic import BaseModel
from typing import List, Dict, Any

class CustomOutput(BaseModel):
    entity_list: List[str]
    results: List[Dict[str, str]]

def custom_processor(result: str) -> Dict[str, Any]:
    # Your custom processing logic here
    pass

li = LanguageInterface.load_model("your_model_id")
li.set_custom_task("CustomTask", custom_processor, CustomOutput)

Knowledge Augmentation

from akali import LanguageInterface, AggressiveKnowledgeAugmenter, AugmenterConfig

mentee = LanguageInterface.load_model("mentee_model_path")
mentor = LanguageInterface.load_model("mentor_model_path")

config = AugmenterConfig(
    # Configure your augmentation settings
)

augmenter = AggressiveKnowledgeAugmenter(mentee, mentor, "EntitySentimentReasoner", config)

# Evaluate current performance
augmenter.evaluate(test_data)

# Propose augmentation strategies
strategies = augmenter.propose()

# Augment the data
augmented_data = augmenter.augment(train_data)

Project Structure

akali/
│
├── src/
│   └── akali/
│       ├── __init__.py
│       ├── aka.py
│       ├── li.py
│       ├── utils.py
│       ├── task_manager.py
│       ├── models/
│       ├── processors/
│       └── ...
│
├── examples/
├── tests/
├── .gitignore
├── pyproject.toml
├── README.md
├── requirements.txt
└── setup.py

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is proprietary and confidential. Unauthorized copying, transferring or reproduction of this project, via any medium, is strictly prohibited.

Contact

For licensing inquiries, support or anything, here is my contact mail: [e-mail].