Pinned Repositories
AB-Testing-Marketing-Campaigns
This Jupyter Notebook demonstrates how to perform A/B testing to evaluate the effectiveness of two different marketing campaigns: a Control Campaign and a Test Campaign. The analysis focuses on various marketing metrics to determine which campaign performs better in terms of engagement, conversions, and overall effectiveness.
Article-Data-Analysis-Visualization
This project involves analyzing a dataset of articles to extract insights, visualize key aspects, and perform topic modeling. The analysis includes creating word clouds, performing sentiment analysis, extracting named entities, and modeling topics.
Aspect-Based-Sentiment-Analysis
This project provides an interactive web application for Aspect-Based Sentiment Analysis (ABSA) using Streamlit. The application allows users to input sentences and analyze them to extract aspects and their corresponding sentiments. The underlying model is powered by the pyabsa library.
Breast-Cancer-Detection-NaiveBayesClassifer
This project involves detecting breast cancer using the Naive Bayes classifier in Jupyter Notebook. Breast cancer detection is a crucial task in healthcare, as it aids in the early diagnosis and treatment of the disease. Through this project, we aim to explore and understand how the Naive Bayes classifier can be used for breast cancer detection.
Fine-tuning-GPT2-Medical-Data
This project involves fine-tuning a small version of the GPT-2 model (distilgpt2) on a dataset of diseases and symptoms. The goal is to train a language model that can generate text related to medical conditions and their symptoms.
LLM-Driven-Marketing-Campaign-Assistant
This Streamlit-based application leverages Language Learning Models (LLMs) like Google Palm to assist in generating customized marketing content such as sales copy, tweets, and product descriptions. The app tailors content for specific target audiences, including kids, adults, and senior citizens.
MultiPDF-Chatbot
A Streamlit application that allows users to interact with the content of multiple PDF files using Google's Gemini Model. This chatbot extracts text from PDF documents, processes it, and enables users to ask questions based on the content of the uploaded PDFs.
Multiple-Disease-Prediction-System
This project provides a streamlit web application for predicting multiple diseases, including diabetes, Parkinson's disease, and heart disease, using machine learning algorithms. The prediction models are deployed using Streamlit, a Python library for building interactive web applications.
Product-Recommendation-System
This is a simple Product Recommendation System built with Streamlit that recommends similar products based on a selected item. The system uses precomputed similarity scores and provides a user-friendly interface for browsing product recommendations.
Vehicle-Detection-Tracking-Counting
This project demonstrates how to combine Ultralytics YOLOv8, ByteTrack, and Supervision to perform object detection, tracking, and counting in a video stream. The setup allows real-time tracking of objects, counting objects that cross a defined line, and saving the results in an output video.
mshaadk's Repositories
mshaadk/LLM-Driven-Marketing-Campaign-Assistant
This Streamlit-based application leverages Language Learning Models (LLMs) like Google Palm to assist in generating customized marketing content such as sales copy, tweets, and product descriptions. The app tailors content for specific target audiences, including kids, adults, and senior citizens.
mshaadk/AB-Testing-Marketing-Campaigns
This Jupyter Notebook demonstrates how to perform A/B testing to evaluate the effectiveness of two different marketing campaigns: a Control Campaign and a Test Campaign. The analysis focuses on various marketing metrics to determine which campaign performs better in terms of engagement, conversions, and overall effectiveness.
mshaadk/Article-Data-Analysis-Visualization
This project involves analyzing a dataset of articles to extract insights, visualize key aspects, and perform topic modeling. The analysis includes creating word clouds, performing sentiment analysis, extracting named entities, and modeling topics.
mshaadk/Aspect-Based-Sentiment-Analysis
This project provides an interactive web application for Aspect-Based Sentiment Analysis (ABSA) using Streamlit. The application allows users to input sentences and analyze them to extract aspects and their corresponding sentiments. The underlying model is powered by the pyabsa library.
mshaadk/Currency-Exchange-Rate-Forecasting
This project focuses on forecasting the USD to INR exchange rate using time series analysis. The model uses SARIMA for predictions based on historical currency exchange data. The project also includes an analysis of seasonal trends and patterns in the exchange rate.
mshaadk/Document-QnA-Gemini-Groq
This application leverages advanced AI techniques to allow users to ask questions about documents and receive accurate answers based on the content.
mshaadk/Ed-Tech-QnA-Assistant
This project leverages LangChain, FAISS, and Groq to create an intelligent assistant capable of answering questions based on a knowledge base built from a CSV file.
mshaadk/Fine-tuning-GPT2-Medical-Data
This project involves fine-tuning a small version of the GPT-2 model (distilgpt2) on a dataset of diseases and symptoms. The goal is to train a language model that can generate text related to medical conditions and their symptoms.
mshaadk/Gemini-LangChain-Chatbot
This application leverages Google's Gemini LLM to create an interactive conversational agent using Streamlit. The chatbot maintains conversation context and offers creative responses.
mshaadk/Hybrid-Search-LangChain-Pinecone
This Google Colab Notebook demonstrates how to integrate Pinecone with LangChain and HuggingFace to create a hybrid search system that combines vector embeddings with sparse retrieval techniques. This setup can be used for efficient information retrieval and search tasks.
mshaadk/Image-Captioning-App
The Image Captioning App leverages the power of pre-trained deep learning models to generate descriptive captions for uploaded images. This web application is built using Streamlit and utilizes the VisionEncoderDecoderModel from the Hugging Face Transformers library to provide detailed captions based on image content.
mshaadk/Invoice-Extractor-Gemini
A web application that allows users to upload invoice images and ask questions about the content of those invoices. Powered by Google’s Gemini generative AI model, this tool is designed for financial advisors and anyone needing assistance with invoice interpretation.
mshaadk/Job-Fit-Analyzer
The Job Fit Analyzer is a web application designed to help job seekers analyze how well their resumes align with specific job descriptions. This tool acts as an intelligent Applicant Tracking System (ATS) and provides insightful feedback on resume match percentages, missing keywords, and a summary of how well the resume fits the job description.
mshaadk/mshaadk
mshaadk/Multi-Agent-Investment-Risk-Analysis
The project is a multi-agent system designed to monitor, analyze, and optimize trading strategies in real-time. Leveraging advanced statistical modeling, machine learning techniques, and risk assessment. This system aims to provide actionable insights and enhance trading decisions for various stocks.
mshaadk/MultiPDF-Chatbot
A Streamlit application that allows users to interact with the content of multiple PDF files using Google's Gemini Model. This chatbot extracts text from PDF documents, processes it, and enables users to ask questions based on the content of the uploaded PDFs.
mshaadk/News-Analysis-Chatbot
This Streamlit application allows users to input URLs of news articles, processes the content, and provides intelligent answers to questions based on the articles. It uses OpenAI's language models and FAISS for efficient information retrieval.
mshaadk/Next-Word-Prediction-LSTM
This project involves building a next-word prediction model using a Recurrent Neural Network (RNN) with LSTM layers. The model is trained on a text corpus to predict the next word given a sequence of words.
mshaadk/No-Code-Classification-Model-Trainer
This project enables users to train and evaluate classification machine learning models without writing a single line of code. Built with Streamlit, the application provides an interactive interface to load datasets, preprocess data, select machine learning models, and evaluate the model’s performance.
mshaadk/Retail-Store-Database-Assistant
The Retail Store Database Assistant is a Streamlit application designed to assist retail store managers and analysts in querying their database effortlessly. Users can input natural language questions, and the app translates them into SQL queries to fetch relevant data from the store's inventory.
mshaadk/Sentiment-Analysis-API
This project is a simple Sentiment Analysis API built with FastAPI and powered by TextBlob. The API analyzes the sentiment of input text, classifying it as positive, negative, or neutral, and provides additional sentiment metrics such as polarity and subjectivity.
mshaadk/Stable-Diffusion-Image-Generation
This project demonstrates how to generate images using a pre-trained Stable Diffusion model in a Jupyter Notebook environment. The notebook covers the installation of necessary libraries, configuration of model parameters, and the process of generating images from text prompts.
mshaadk/Statistics-Using-Python
This repository contains various statistical concepts and visualizations implemented using Python, ranging from basic descriptive statistics to advanced probability distributions and data visualization techniques.
mshaadk/Synthetic-Test-Data-Generation-Ragas
This project focuses on generating synthetic test data using Ragas with the help of pre-trained models like LLaMA3 for data generation and critique. The project is implemented in Google Colab, which makes it accessible and easy to use for running machine learning experiments in the cloud.
mshaadk/Tech-Stocks-Performance-Analysis
This Streamlit app provides an interactive interface to analyze and visualize the performance of various tech stocks over a specified date range. The app leverages historical stock data from Yahoo Finance and offers insights into stock prices, moving averages, and volatility.
mshaadk/Time-Series-Analsis-Stocks
The Time Series Analysis - Stocks application is a web-based tool for analyzing stock market data. Built using Python, Streamlit, and Plotly, this application allows users to visualize historical stock prices through different types of charts.
mshaadk/Vehicle-Detection-Tracking-Counting
This project demonstrates how to combine Ultralytics YOLOv8, ByteTrack, and Supervision to perform object detection, tracking, and counting in a video stream. The setup allows real-time tracking of objects, counting objects that cross a defined line, and saving the results in an output video.
mshaadk/Youtube-Assistant-LangChain
The YouTube Assistant is a powerful tool designed to help you extract valuable insights from YouTube videos. By leveraging advanced NLP models, it processes video transcripts and allows you to query them effectively. This tool integrates several libraries and APIs to provide detailed and contextually accurate responses based on video content.
mshaadk/Youtube-Video-Transcription-Whisper
This project provides a streamlined approach to downloading a YouTube video, extracting its audio, and transcribing the audio content using OpenAI's Whisper model.
mshaadk/Zomato-Dataset-Analysis
This project performs Exploratory Data Analysis (EDA) on a Zomato dataset, focusing on data cleaning, processing, and visualization. The aim is to derive meaningful insights about restaurant ratings, locations, cuisines, and more.