Skip to content

SayantanML-IO/pc-builder-pro

Repository files navigation

PC Builder Pro 🖥️

An intelligent, AI-powered PC building assistant that uses Machine Learning to predict performance, a Genetic Algorithm to optimize budgets, and Large Language Models (LLM) to explain compatibility issues.

🚀 Key Features

  • 🧠 AI Compatibility Engine: Hybrid system using rule-based logic and Groq LLM to explain why parts are incompatible in plain English.
  • 📈 ML Performance Prediction: A Random Forest model (trained on 3DMark data) predicts the gaming performance of any build.
  • 🧬 Genetic Algorithm Optimizer: Generates the mathematically optimal build for any specific budget using evolutionary algorithms (DEAP).
  • 🔍 Semantic Search: Vector-based search allows users to find parts using natural language (e.g., "fast white GPU for 4k gaming").
  • 📊 Interactive UI: Modern Streamlit dashboard with interactive network graphs and real-time feedback.

🛠️ Tech Stack

  • Backend: FastAPI, SQLAlchemy, Pydantic
  • Frontend: Streamlit, Plotly (for visualizations)
  • Machine Learning: PyTorch, Scikit-learn, FAISS (Vector Store)
  • AI/LLM: Groq API (Llama-3 70B)
  • Database: SQLite (with Alembic migrations)

⚙️ Setup & Installation

  1. Clone the repository:

    git clone [https://github.com/SayantanML-IO/pc-builder-pro.git](https://github.com/SayantanML-IO/pc-builder-pro.git)
    cd pc-builder-pro
  2. Install dependencies:

    python -m venv venv
    source venv/bin/activate  # On Windows: venv\Scripts\activate
    pip install -r requirements.txt
  3. Set up environment variables: Create a .env file and add your keys:

    GROQ_API_KEY=your_groq_api_key_here
  4. Run the App:

    • Backend: uvicorn app.main:app --reload
    • Frontend: streamlit run ui/app.py

About

An AI-powered PC building assistant featuring ML performance prediction, Genetic Algorithm budget optimisation, and LLM-based compatibility analysis.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors