Skip to content

Aautonomous AI agent that generates neural network code and builds a CNN-based binary classifier to categorize uploaded data across two labels, streamlining machine learning development for researchers and developers.

Notifications You must be signed in to change notification settings

Magnet-AI/Anaa-Coder

Repository files navigation

🧠 Anna-Coder: Autonomous Neural Network Code Generator

Anna-Coder is an autonomous AI agent that automatically generates and trains neural network code. This project includes a CNN-based binary classifier designed to classify data across two labels from 16 uploaded files.


🚀 Features

  • Automated neural network code generation
  • CNN binary classifier for two categories
  • Easy-to-use data loading and preprocessing
  • Configurable model parameters and training options
  • Evaluation and inference scripts for quick testing

🗂️ Project Structure

anna_coder/
│
├── data/
│   ├── class_0/        # Files for label 0
│   └── class_1/        # Files for label 1
│
├── src/
│   ├── model.py        # CNN architecture
│   ├── dataset.py      # Data loading and preprocessing
│   ├── train.py        # Training script
│   ├── eval.py         # Evaluation script
│   └── utils.py        # Helper functions
│
├── configs/
│   └── default.yaml    # Model and training configuration
│
├── requirements.txt
└── README.md

⚙️ Installation

python -m venv .venv
source .venv/bin/activate   # or .venv\Scripts\activate on Windows
pip install -r requirements.txt

🧩 How to Run

1. Train the Model

python -m src.train --config configs/default.yaml

2. Evaluate the Model

python -m src.eval --weights outputs/best_model.pt --data_dir data

3. Run Inference

python -m src.infer --image path/to/image.jpg

👥 Collaborators

  • Manush Murali
  • Sovit Nayak

About

Aautonomous AI agent that generates neural network code and builds a CNN-based binary classifier to categorize uploaded data across two labels, streamlining machine learning development for researchers and developers.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •