Naive Boyes - SafeStride - Botrush 3.0

Machine Learning Classification Interface + Pathfinding Intelligence

---

✨ A full-stack application for machine learning classification tasks with file upload capabilities, result visualization, and grid-based intelligent pathfinding — deployed on Vercel (frontend) and Render (backend).

---

🚀 Features

• 📁 Modern File Upload Interface

  • 🖱️ Drag & drop file upload via FileUpload.js
  • 📊 Progress tracking with Axios

• 🧠 Advanced ML Processing

  • 🤖 Trained ML models (binary.keras, multi.keras) for Ensembled Prediction
  • Achieved an impressive 98.2% accuracy
  • 👁️ Computer vision support via OpenCV (cv2)
  • 🔢 NumPy-based data processing

• 🧭 Shortest Path Finder (TSP + BFS Hybrid)

  • 🟩 Identifies all safe and supervisable blocks from the ML-classified image grid.
  • 🧠 Combines BFS (Breadth-First Search) for inter-point distance calculation and TSP (Traveling Salesman Problem) logic to compute the optimal route.
  • 🚫 Avoids dangerous or unsafe zones like fire or blocked areas, identified through the classifier.
  • 🔄 Ensures minimum cost traversal that starts from the bottom-left and covers all required safe points.

• ⚡ Real-time API

  • 🚀 FastAPI backend with Uvicorn server
  • 📚 Automatic docs at /docs

• 📱 Responsive UI

  • 🎨 Tailwind CSS styling
  • 📲 Mobile-optimized components

---

🛠️ Tech Stack

Area	Technologies
🌐 Frontend	Next.js 15, React, Tailwind CSS, Axios
🔙 Backend	FastAPI, Uvicorn, Python 3.10+
🤖 ML	TensorFlow/Keras, OpenCV, NumPy
🧭 Pathfinding	Custom TSP-BFS Hybrid Algorithm
☁️ Deployment	Vercel (Frontend), Render (Backend)

---

🌍 Live Demo

Try it Out!

---

🛠️ Installation

Frontend (Next.js)

cd nikhil-yadav15-botrush_3.0_naiveboyes/NextApp
npm install
npm run dev