Developed Iris Flower Classification App with Python and Streamlit

🚀 Exploring Streamlit — The Easiest Way to Build Data Apps! 🔸 Recently, I started working with Streamlit, a Python framework that lets you turn your data science or machine learning scripts into interactive web apps — with just a few lines of code! ✨ What I learned: 🔸 It’s super beginner-friendly 🔸No need for HTML/CSS/JS 🔸 Perfect for quick ML model demos or dashboards 🔸 Integrates easily with pandas, matplotlib, and scikit-learn 🎯 If you want to learn a framework in just a few hours, Streamlit is the best one to start with! I’d be glad to help anyone who wants to explore it — feel free to reach out #Streamlit #Python #DataScience #MachineLearning #AI #LearningJourney

Just built a Movie Recommendation System! Excited to share my latest project — a Content-Based Movie Recommender built using the TMDB 5000 Movies Dataset. The app suggests top 5 similar movies based on user-selected titles using cosine similarity on processed metadata. 🔧 Tech Stack Python Pandas, NumPy Scikit-learn (similarity matrix) Streamlit (UI) Pickle (model + metadata storage) 🎯 What it does Reads and processes the TMDB dataset Extracts key features from movie metadata Builds a similarity matrix Uses it to recommend the 5 closest matches Provides a simple, clean UI for the user to choose any movie 🎬 Features Instant recommendations Fast lookup through a precomputed similarity matrix User-friendly web interface built with Streamlit Easily deployable 📂 GitHub Repository: https://lnkd.in/dB2nHSzW Feedback is always welcome 😊 #MachineLearning #Python #AI #RecommendationSystem #Streamlit #DataScience #Project

Just finished a fun project — an interactive digit recognition web app built using Streamlit and a Support Vector Machine (SVM) trained on the classic MNIST dataset. The app lets you upload any 28×28 handwritten digit image (even noisy ones), and it predicts the digit in real time 🧠✨ What makes this version special: ✅ Clean preprocessing pipeline using OpenCV (grayscale → threshold → crop → resize → pad → normalize) ✅ Adaptive thresholding for robust lighting correction ✅ Aspect-ratio preserving resize to match MNIST distribution ✅ Centered 28×28 output so the digit aligns just like MNIST samples ✅ Simple Streamlit UI with live preview of the processed image A small tweak — like fixing the image preprocessing — made a huge difference in accuracy and consistency. It’s always amazing how much impact proper data formatting has on model performance! 📦 Stack used: Python · Streamlit · OpenCV · NumPy · scikit-learn · Pandas 🧩 Next step: converting this into a CNN version for comparison. If anyone wants to explore or collaborate on extending it, I’d love to connect! 👇 #MachineLearning #Python #ComputerVision #Streamlit #SVM #OpenCV #AI #DataScience #MNIST #ProjectShowcase Gagan Puri 🙏 Mentor

🚀 Excited to share my latest project: BookRecs - An AI-Powered Book Recommendation System! 📚 Built a full-stack web application that leverages machine learning to help readers discover their next favorite book through intelligent recommendations. 💡 Key Features: ✅ Smart recommendation engine using collaborative filtering algorithm ✅ Curated collection of 50+ popular books with real user ratings ✅ Real-time similarity computation with cosine similarity ✅ Instant personalized suggestions based on user preferences 🛠️ Tech Stack: Python | Flask | NumPy | Scikit-learn | Machine Learning 📊 The system analyzes thousands of user ratings to identify patterns and suggest 4 highly relevant books instantly when users input their favorite title. 🎓 This project enhanced my expertise in machine learning pipelines, recommendation systems, data preprocessing, and production-ready web application development. 🔗 GitHub: https://lnkd.in/gua8TMtm #MachineLearning #Python #AI #DataScience #Flask 

🚭 Smoking Prediction Web Application – Built with Streamlit and Machine Learning I’m excited to share the modified version of my earlier project — a Smoking Prediction Web Application developed using Python, scikit-learn, and Streamlit. In this updated version, I’ve improved the model’s performance and refined the Streamlit interface for a smoother and more user-friendly experience. The application predicts whether an individual is likely to be a smoker based on health and lifestyle data, using insights from real-world datasets. Tools: Python | Pandas | scikit-learn | Streamlit GitHub Repository: https://lnkd.in/gHKUFiHD I welcome any feedback or professional suggestions for further enhancement. #machinelearning #datacollection #datapreprocessing #datavisualization #datamodeling #dataevaluation #deployment #python #pythonlibraries #streamlit

💰 Smart Expense Tracker Bot (AI + Firebase + Python Tkinter) 🚀 Built a modern desktop app that helps users track, visualize, and optimize expenses — now powered by Gemini AI for smart savings advice! ✨ Features: ✅ Beautiful Tkinter UI ✅ Firebase for real-time expense storage ✅ Interactive charts & CSV export ✅ AI coach that analyzes your spending patterns and gives personalized tips to save more 📽️ Included a short demo video inside the project folder. Check it out on GitHub 👇 🔗 https://lnkd.in/dUjtYmiA #Python #AI #Firebase #GeminiAI #Tkinter #ProjectShowcase #FinanceTech

👇 🚀 Introducing Smart Expense Splitter — Built with Python & Streamlit! 💸 Tired of the monthly struggle of splitting rent, electricity, and food bills? I built a simple yet powerful Streamlit web app that calculates and visualizes shared expenses with ease. Each person’s contribution is shown clearly with interactive charts and a clean, modern UI — making expense sharing fair and stress-free! 🧠 Tech Stack: Python | Streamlit | Plotly | Pandas 🔗 Try it live here: https://lnkd.in/dfvJxRiH 🎥 Check out the video below to see it in action — feedback and suggestions are always welcome! 💬 #Python #Streamlit #DataScience #AI #MachineLearning #WebApp #PortfolioProject #ExpenseTracker #Coding #Developers #Visualization

🏠 Project Showcase: House Price Prediction 📊 I’m excited to share my end-to-end machine learning project on predicting house prices! ✨ Project Overview: Built a regression model in Google Colab to predict house prices based on features like number of convenience stores, house area, and more. Applied data preprocessing, feature analysis, regression modeling, and model evaluation for accurate predictions. Exported the trained model as a pickle file and integrated it into a Python app (app.py) in VS Code. Ran the app in Anaconda Prompt, demonstrating real-time predictions. Recorded a video of the workflow and output to showcase the project in action. 💡 Tech Stack: Python | Pandas | NumPy | Scikit-learn | Google Colab | Pickle | VS Code | Anaconda | App Development | Regression Modeling GitHub Repository: [ https://lnkd.in/gQP5c7Qf ] 📈 Key Learnings: Understanding the impact of different features on house prices Debugging model predictions and improving accuracy Deploying a machine learning model into a runnable Python app #MachineLearning #DataScience #Python #Regression #HousePricePrediction #AI #MLDeployment #VSCode #Anaconda #Pickle

 Excited to share my latest Streamlit project — “Safura Iris Flower Classifier” 🌸 This interactive web app allows users to: 🔹 Explore and visualize the classic Iris dataset 🔹 Train a Logistic Regression model 🔹 Evaluate model performance with accuracy and F1-score metrics 🔹 Predict flower species with user-input features in real time 🌼 💻 Built using Python, Streamlit, scikit-learn, seaborn, and matplotlib, the app provides a simple and colorful interface for both data visualization and machine learning model interaction. #MachineLearning #DataScience #Streamlit #Python #IrisDataset #AI #ProjectShowcase #LogisticRegression

I’m excited to share one of my recent projects — a Movie Recommendation System built using Machine Learning, Python, and Streamlit! This project recommends movies similar to the one a user selects, based on their descriptions and features from The Movie Database (TMDb). It was an amazing opportunity to combine both my data science and web development skills. 💡 What I worked on: Built a content-based recommendation model using scikit-learn Processed and analyzed movie data with Pandas and NumPy Integrated the TMDb API to fetch real-time movie posters and details Designed an interactive and visually appealing Streamlit frontend to display recommendations horizontally 🎯 Key Learnings: This project helped me understand how machine learning models can be connected with real-world data and deployed through simple web apps. I also learned how to work with APIs, handle data preprocessing, and improve user interfaces. 🔗 Check out the project on GitHub: https://lnkd.in/gaTDvTJA #MachineLearning #Python #DataScience #Streamlit #RecommendationSystem #AI #ArtificialIntelligence #TMDB