MLP Network Configurator Files

📘 README – MLP Network Configurator 1.2

🎯 Overview
MLP Network Configurator is an interactive, educational tool designed 
to help users understand how artificial neural networks work.
It offers real-time training feedback, intuitive architecture visualization, 
and even camera-based data collection — ideal for students, teachers, 
and curious minds.

💻 System Requirements
- OS: Windows 11 (recommended)
- The program does not require an internet connection and does not 
  install any background services.

🚀 Installation
- During the installation, from the offered C:Users directory, 
  choose the subdirectory that has your own name.

🧠 Key Features
- Data Handling: Load, normalize, and validate datasets (JSON format).
- Model Configuration: Define hidden layers, activation functions, 
  loss and optimizer types.
- Training: Visualize loss curves, predictions, and PCA-based class separation.
- Testing: Evaluate accuracy, error rates, and misclassified samples.
- Camera Dataset Builder: Capture and label images live via webcam.
- Model Insight: Analyze weight and bias evolution across epochs.
- Save & Load: Export models and configurations for reuse and sharing.

📂 File Formats
- Training/Test data: JSON (x_data, y_data)
- Saved models: .pt (PyTorch weights), .txt (config summary)
- Camera exports: JSON format, ready for training

🔐 Source Code Notice
The core source code of the program is proprietary and not publicly available.
Modification, redistribution, or reverse engineering is prohibited without 
prior written permission from the author.

📜 License
This software may be used freely for educational and personal purposes.
Commercial use requires prior authorization.
See LICENSE.txt for full terms.