LLM Building on our own data | Low Internet Availability Space | PDF - Video Summarisation using OpenAI
Shivam Mittal (~shivam0) |
Description:
In This talk, I would like to discuss about two concepts which I have worked in Microsoft Research. 1. Building Summarisation module using OpenAI on our own PDFs and Video content. 2. Building our own chatbot on our own data
In this talk, I will be taking Class-9th-Class 12th Social Science content and end-goal would be we can ask any doubt to our AI chatbot like when French revolution happened and what are the reasons behind it etc? And users can access specific chapter summarisation, so that users who want to revise a particular chapter, can do so.
Presentation Link: https://docs.google.com/presentation/d/1GGPbQEy6ZNQ8vhY7TGNc1HSNH1jnyKikLrzEvvGAgoc/edit?usp=sharing
Prerequisites:
No experience needed. I will start from basic OpenAI chatgpt apis and sharing how we can build our own specific chatbot.
Content URLs:
Content resources links: Tutorials
https://blog.langchain.dev/tutorial-chatgpt-over-your-data/
https://python.langchain.com/en/latest/modules/models/llms/examples/llm_caching.html
https://github.com/zilliztech/GPTCache
https://zilliz.com/learn/what-is-vector-database
https://github.com/towhee-io/examples/tree/main/nlp/question_answering
https://docs.pinecone.io/docs/semantic-text-search
Speaker Info:
I am Shivam, working at Microsoft, having past experience at Microsoft Research, Cred and other ML startups. I am Kaggle expert been into top-300 in kaggle competitions category. I have started experimenting upon Gpt-4 apis and how we can build our own chatgpt and get our data specific response. So, In this workshop, I would like to share my experience and learnings.
Speaker Links:
Linkedin: https://www.linkedin.com/in/shivam274 Kaggle: https://www.kaggle.com/shivammittal274