Adala: Autonomous Data Labeling Agent Framework
Adala is an innovative Autonomous Data Labeling Agent framework that aims to transform data processing and labeling tasks for AI professionals. This advanced framework allows for the creation of specialized agents capable of autonomous learning, which is crucial for AI engineers, machine learning researchers, data scientists, and educators.
Key Features
- Reliable Agents: Adala's agents are built on accurate ground truth data to provide consistent and reliable results.
- Controllable Output: Users can configure outputs and constraints for each skill to meet specific project requirements.
- Specialized in Data Processing: Adala agents excel in various data labeling tasks and can be customized for diverse data processing needs.
- Autonomous Learning: Intelligent agents in Adala learn autonomously through iterative processes.
- Flexible and Extensible Runtime: Adala offers an adaptable runtime environment, allowing deployment across multiple scenarios and encouraging community extensions.
- Easily Customizable: Users can quickly customize agents without extensive learning, catering to both beginners and experts.
Who Can Benefit from Adala?
- AI Engineers: Architect and design AI agent systems with interconnected skills.
- Machine Learning Researchers: Experiment with complex problem decomposition and causal reasoning.
- Data Scientists: Utilize agents for data preprocessing and postprocessing.
- Educators and Students: Use Adala for teaching or advanced projects and research.
Getting Started
Installation: pip install adala
For the latest updates from GitHub: pip install git+https://github.com/HumanSignal/Adala.git
Prerequisites: Set your OpenAI API key with export OPENAI_API_KEY='your-openai-api-key'
Quickstart Example
import pandas as pd
from adala.agents import Agent
from adala.environments import StaticEnvironment
from adala.skills import ClassificationSkill
from adala.runtimes import OpenAIChatRuntime
agent = Agent(
environment=StaticEnvironment(df=train_df),
skills=ClassificationSkill(
name='sentiment',
instructions="Label text as positive, negative or neutral.",
labels={'sentiment': ["Positive", "Negative", "Neutral"]},
input_template="Text: {text}",
output_template="Sentiment: {sentiment}"
),
runtimes = {
'openai': OpenAIChatRuntime(model='gpt-3.5-turbo'),
},
default_runtime='openai'
)
agent.learn(learning_iterations=3, accuracy_threshold=0.95)
predictions = agent.run(test_df)
Available Skills
Adala offers various pre-built skills such as Classification, Summarization, Question Answering, Translation, Text Generation, Ontology Creator, and Math Reasoning. Each skill includes example notebooks and Colab integration for easy experimentation.
Conclusion
Adala represents the future of autonomous data processing and labeling, empowering AI professionals to tackle complex challenges efficiently. Whether you're in cutting-edge research, AI system development, or data science education, Adala provides the necessary tools and capabilities to advance your projects.
