📚 Project Index

This repository covers:

• Parallel Computing Basics using HPC concepts (HPC & Parallelism)
• Deep Learning – Getting started with basics
• Artificial Intelligence – Foundations
• Convolutional Neural Networks - Deep Learning with Python

---

🎯 Project Goal

• The ultimate goal of this project is to build and run a Deep Learning network on a High-Performance Computing (HPC) platform from scratch.
• To successfully follow this project, advanced knowledge of C++ and Python is mandatory.
• Note: This document is intended as a collection of concise reference points or memory aids. It does not cover detailed explanations or theoretical foundations. Readers are expected to have prior knowledge of the underlying concepts.

🛠 Requirements

• A powerful PC and an Edge Computing Platform (e.g., NVIDIA Jetson REStudio J4011).
• Familiarity with C++, Python, and parallel programming concepts.

---

📦 Project Structure

The project is divided into Modules, Sub-chapters, and Exercises.

---

Module 1: HPC & Parallelism

This module includes:

• ✅ Portable HPC applications using ISO C++
• ✅ GPU acceleration using the C++ Standard Library
• ✅ Fundamentals of ISO C++ parallelism
• (Lessons are not in order* only for this module)

Files Included:

• DAXPY
• ISO C++ Algorithms HPC Documentation
• Indexing in Parallel Computing
• NVIDIA Grace Hopper Coherent HW
• Parallel Algorithms in C++
• Deep Learning Exercise
• Exercises
• Exercise 1
• Exercise 2
• Exercise 3
• Extra Exercise 1
• Introduction
• Main Program

---

Module 2: Deeplearning Introduction

• Deeplearning Introduction
• neurel network architecture
• neural network example structured data
• neural network example unstructured text data

---

Module 3: Advanced Neural Network Topics

This module includes:

• Training a Neural Network
• Classify News Article (Code)
• Multilayer Perceptron (Notes)
• Multilayer Perceptron (Code)
• Convolutional Neural Networks
• Recurrent Neural Network
• Transformer Architecture
• Types of Neural Networks

---

Module 4: Convolutional Neural Networks

This module covers CNN fundamentals, preprocessing, augmentation, and practical implementations.

📂 Contents

• Convolutional Neural Networks
• Preprocessing Image Dataset
• Image Augmentation
• CNN with Python (Notes)
• CNN CIFAR-10 Full Implementation
• CNN Architecture
• Computer Vision Basics ( Revised )

---

Module 5: Build Neural Network with PyTorch

This module focuses on building a feed‑forward neural network in pure PyTorch and then demonstrates how PyTorch Lightning simplifies training loops, logging, and device handling. It includes hands‑on demos for both classification and regression using synthetic datasets (no external downloads required).

📂 Contents

• Build Neural Network with PyTorch
• PyTorch Lightning
• PyTorch Lightning — Classification (Code)
• PyTorch Lightning — Regression (Code)

✅ Notes

• All links are relative to the repository root.
• If you keep per-module folders (e.g., module4/), update links like:
  ./module4/convolutional_neural_networks.md, etc.

👤 Current Status

Currently, all project activities are done by myself with GPT-4 support.
Once the base version of this project is ready, I will invite collaborators to contribute and expand the work.

---

---

• ✅Python Example:

  pip install torch pandas scikit-learn
  python wine_preprocessing_example.py
  python classify_news_article.py
  

✅ How to Run C++ Code

To run the C++ examples, set up Visual Studio Code in a Linux environment.

---