intura-ai 0.0.3.1

Intelligent Research and Experimentation AI for LLM experimentation production.

Intura-AI: Intelligent Research and Experimentation AI

intura-ai is a Python package designed to streamline LLM experimentation and production. It provides tools for logging LLM usage and managing experiment predictions, with seamless LangChain compatibility.

Dashboard: dashboard.intura.co

Features

• Callbacks:
  • UsageTrackCallback: Log LLM usage details for analysis and monitoring.
• Experiment Prediction:
  • ChatModelExperiment: Facilitates the selection and execution of LangChain models based on experiment configurations.
• LangChain Compatibility:
  • Designed to integrate smoothly with LangChain workflows.

Installation

pip install intura-ai

Usage

Initialization

Before using intura-ai, you need to initialize the client with your API key.

from intura_ai.client import intura_initialization

INTURA_API_KEY = "..."
intura_initialization(INTURA_API_KEY)

Experiment Prediction

Use ChatModelExperiment to fetch and execute pre-configured LangChain models.

from intura_ai.experiments import ChatModelExperiment

EXPERIMENT_ID = "..."
client = ChatModelExperiment(EXPERIMENT_ID
llm, messages = client.build()
messages.append(('human', 'give me today quote for programmer'))
llm.invoke(messages)

Usage Tracking Callback

Integrate UsageTrackCallback to log LLM usage during execution.

from intura_ai.callbacks import UsageTrackCallback
from langchain_google_genai import ChatGoogleGenerativeAI
from langchain.schema import HumanMessage

EXPERIMENT_ID = "..."
llm = ChatGoogleGenerativeAI(
    model="gemini-1.5-pro",
    max_tokens=300,
    timeout=None,
    max_retries=2,
    callbacks=[
        UsageTrackCallback(EXPERIMENT_ID)
    ]
)

messages = [HumanMessage(content="What is the capital of France?")]
llm.invoke(messages)

Contributing

Contributions are welcome! Please feel free to submit pull requests or open issues for bug reports or feature requests.

License

This project is licensed under the MIT License.