"Open Source Models with Hugging Face" course empowers you with the skills to leverage open-source models from the Hugging Face Hub for various tasks in NLP, audio, image, and multimodal domains.

A Python-based REST API for PDF OCR using AI models with PyTorch and Transformers that runs in a Docker container.

llm-newsletter-generator transforms a valid RSS feed into a "Newsletter" using AI models via PyTorch and Transformers; this is experimental.

Spam Detector is a Data Science Project built using Pytorch and Hugging Face library. Used BERT model based on Transformer Architecture and got 99.97% accuracy on train set and 98.76% accuracy on test set.

A FastAPI-powered REST API offering a comprehensive suite of natural language processing services using machine learning models with PyTorch and Transformers, packaged in a Docker container to run efficiently.

A real-time voice-to-text and text-to-speech AI pipeline using Whisper, an LLM, and Edge-TTS with tunable parameters for low-latency audio processing and response generation.

A web-based utility for fetching, categorizing, summarizing and managing global news and articles using the GDELT 2.0 API. Designed for content creators, news aggregators, and researchers, this tool simplifies access to up-to-date articles with an intuitive UI and customizable configurations.

With the use of AI, summarise your movies and bring back the colour in older films.

Build a sentiment analysis tool that processes user reviews from various platforms (like Amazon or Yelp) and provides insights on sentiment trends over time. Use advanced NLP techniques like Transformers (BERT, GPT).

Explore and implement Hugging Face Transformers and Pipelines for leveraging powerful pretrained AI models in NLP and more.

Successfully developed a fine-tuned DistilBERT transformer model which can accurately predict the overall sentiment of a piece of financial news up to an accuracy of nearly 81.5%.

Ce projet propose une application pratique de FinBERT, un modèle de traitement du langage naturel spécialisé dans l'analyse des textes financiers.

Summarize, , NSP answer questions in dockerised environment

Successfully established a Seq2Seq with attention model which can perform English to Spanish language translation up to an accuracy of almost 97%.

Google-Palm powered web aplication allowing you to query your own PDF file. Uses streamlit for UI, ChromaDB to store embeddings and langchain.

Deployed an interactive web platform for exploring and utilizing language models. Features include real-time text analysis and translation, built with Django for robust performance and scalability

Source codes and materials of Advanced Spelling Error Correction project.

This project contains codes and paperwork based on the course CSI5386 at University of Ottawa (delivered by Professor Dr. Diana Inkpen).

Plataforma desenvolvida em Python que visa automatizar e agilizar o processo de avaliação de projetos de inovação tecnológica, utilizando inteligência artificial e critérios padronizados com base na Lei do Bem.

This project leverages knowledge distillation to create a lightweight yet powerful sentiment analysis model, tailored specifically for financial news data. Using a teacher-student approach, the project distills knowledge from a large FinBERT model into a compact DistilBERT-based student model, balancing performance and efficiency.