Collection of tools for building diachronic/historical word vectors

Building an NLP model to translate English to Vietnamese

RAG-based Streamlit app that uses Langchain, OpenAI Embeddings, GPT, and Pinecone Vector Database to answer questions about a user-provided document

BERT, RNN from Scratch, Bigram, Azure AI, Sentiment Analysis, Word Embeddings, GloVe, LSTM.

This repository has a lot of Machine Learning projects done. It is the best place to start learning Machine Learning.

This is an experiment in learning langchain, pinecone and stuff, don't mind

Introductory labs on Graph embeddings and GNNs.

This project implements a search engine that combines traditional text-based search with vector-based semantic search for enhanced retrieval capabilities.

A powerful toolkit for text chunking and semantic search using OpenSearch. This toolkit provides various text chunking strategies and embedding capabilities for efficient document retrieval.

A model-based cleaner using Laser sentence embeddings to exploit embeddings to filter misaligned segment pairs. Product scaled by asynchronously building the Task Queues, dispatching the tasks in a Round Robin method and adding multiple workers on the RabbitMQ server for consumption.

Chat with the pdfs

demo of openai's embeddings api

AniSearchModel leverages Sentence-BERT (SBERT) models to generate embeddings for synopses, enabling the calculation of semantic similarities between descriptions. This allows users to find the most similar anime or manga based on a given description.

This library provides Discord.NET bots, including NDB, with easy embed functionality.

CS Bachelor Thesis. Open Domain Question Answering System that tries to answer general topic questions fetching from wikipedia.

This is repo for various deep learning projects

Open-source RAG API with FastAPI, FAISS, and OpenAI for LLM-driven document search and response generation.

An intelligent course recommendation system using Retrieval-Augmented Generation (RAG) and large language models (LLMs), designed to guide students through academic course selection by retrieving and reasoning over contextual information.

transformer build from scratch, referred the video of Umar Jamil for the transformer architecture.

A RAG system to upload product data (CSV) to a MongoDB database in XML form, generate multilingual embeddings using a Cohere model and create and store a FAISS index for fast similarity search. Lets users chat in real-time with a sales assistant personality using a Flask web interface.