We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Philip Meier - From RAGs to riches: Build an AI document interrogation app | PyData Global 2023
Learn how to build an AI-powered document interrogation app using RAG (retrieval augmented generation) technique and the RAGNA framework, featuring fine-tuning, customizable, and open-source capabilities.
-
RAGstands for “retrieval augmented generation”, a technique used to generate answers to questions by retrieving relevant documents and then using a language model to generate the answer. - The technique is used to improve the accuracy and efficiency of language models, particularly in cases where the model has a limited context size or is asked a question that requires specific knowledge.
- RAGNA is a framework that provides a set of tools and components for building RAG applications, including a Python API, REST API, and graphical web UI.
- The framework includes a vector database that allows for fast and efficient retrieval of documents based on similarity to a query or prompt.
- RAGNA provides support for multiple language models and assistants, including GPT 3.5 and GPT 4, and can be used to extend the capabilities of existing models.
- The framework includes a feature for fine-tuning models using a user’s own data, which can be used to improve the accuracy and relevance of the answers generated by the model.
- RAGNA is designed to be highly configurable, allowing users to customize the behavior of the framework and tailor it to their specific needs.
- The framework includes a feature for embedding documents in a vector space, which allows for fast and efficient retrieval of relevant documents based on similarity to a query or prompt.
- RAGNA provides support for multiple data sources and can be used to integrate with a variety of external APIs and services.
- The framework includes a feature for visualizing the performance of the model, which allows users to evaluate the accuracy and relevance of the answers generated by the model.
- RAGNA is open-source and is available for download on the GitHub repository.