ujju0906's Projects
Code from presentations
An API built using flask deployed on aws lambda which uses a model that in-turn classifies hotel reviews into good and bad . If its a bad review it lists out all the issues faced in the review so that the hotel management can quickly solve these issues in an automated manner
🚀 List of distributed system projects for inspiration and learning to build distributed services from real world examples
BERT score for text generation
Blood donation database management
-> Built “Books Recommendation System - Item: Item Based” for amazon by analyzing more than 1 Million products (metadata) -> Executed graph analysis, developed weighted edge graph and finalized an ego network with depth=1 based on product similarity -> Devised a formula for the best recommendations combining factors such as degree of centrality
LLM powered retrieval engine designed to process a ton of sources to collect a comprehensive list of entities.
An python application built for converting all the new fir documents stored as a pdf into a digitised text format using google vision api. Extracting relevant information from it using regex in python followed by storing into an sql database
FAISS-youtube-dataloader-LLM enhances FAISS integration with RAG models, providing a you tube video transcript data loader for efficient handling of large text datasets. This facilitates seamless use of FAISS for similarity search tasks in RAG applications, improving performance in natural language processing projects.
A guide to making your first open-source contribution !
Using Langchain and OpenAI to analyse resume
This is a guide for how one can prepare for GRE within a month's duration.
Using a deep learning framework to segment tumors and glimas in the brain for tumor detection, followed by deployment of the same on google cloud platform.
code for different Text Generation LLM's
Only for "Mad" people
Python code for common Machine Learning Algorithms
This project consist MERN boiler plate code.
A mini-kafka setup done using socket programming in python consisting of dynamic producers, consumers, 3 brokers and a zookeeper for broker election and heartbeat monitoring. The topics are hashed and stored in corresponding directories
A recommendation system for classical music performers using graph embeddings.
Demo app to test and get used to the demo envrionment