LLM Section Release

Highlights

Cleaned all sections and included new Apptainer guide for HPC compute containerization and LLM introduction and hyperparameter tuning with Gemma 7b (Google) model.

Sections and Subsections

• Deep Learning and Deep Reinforcement Learning Tutorials (Libraries: Python, PyTorch, Gym, NumPy, Matplotlib and more)

  • Introduction
  • Course Progression
  • Practical Deep Learning with PyTorch
    • Matrices
    • Gradients
    • Linear Regression
    • Logistic Regression
    • Feedforward Neural Network (FNN)
    • Convolutional Neural Network (CNN)
    • Recurrent Neural Network (RNN)
    • Long Short-Term Memory Network (LSTM)
    • Autoencoders (AE)
    • Fully Connected Overcomplete Autoencoders
  • Improving Deep Learning with PyTorch
    • Derivative, Gradient and Jacobian
    • Forward- and Backward-propagation and Gradient Descent (From Scratch FNN Regression)
    • Learning Rate Scheduling
    • Optimization Algorithms
    • Weight Initialization and Activation Functions
  • Deep Reinforcement Learning with PyTorch
    • Supervised to Reinforcement Learning
    • Markov Decision Processes and Bellman Equations
    • Dynamic Programming
  • From Scratch Deep Learning with PyTorch/Python
    • From Scratch Logistic Regression Classification
  • Compute Optimization
    • Speed Optimization Basics Numba

• Language Models (Libraries: Python, Pytorch, Ollama, LlamaIndex, CUDA, Huggingface, Apptainer)

  • Intro
  • Containers
    • HPC Containers with Apptainer
  • Language Models
    • LLM Introduction & Hyperparameter Tuning

• Machine Learning Tutorials (Libraries: Python, cuDF RAPIDS, cuML RAPIDS, pandas, numpy, scikit-learn and more)

  • RAPIDS cuDF
    • Intro
    • GPU DataFrames
    • CPU/GPU Fractional Differencing

• Programming Tutorials (Libraries: C++, Python, Bash and more)

  • Intro
  • C++
  • Bash
  • Python
  • Javascript
  • Electron

• Data Engineering Tutorials (Libraries: Bash, Databricks, Delta Live Tables, Parquet, Python, Cassandra, and more)

  • Cassandra (NoSQL)
    • Introduction
    • Apache Cassandra Cluster Setup

Cleaned Stable Release

Highlights

Cleaned documentation, code, and website with a particular emphasis on sections and subsections to demarcate tutorials for ease of learning and ensure scalability in their respective sections.

Sections and Subsections

• Deep Learning and Deep Reinforcement Learning Tutorials (Libraries: Python, PyTorch, Gym, NumPy, Matplotlib and more)

  • Introduction
  • Course Progression
  • Practical Deep Learning with PyTorch
    • Matrices
    • Gradients
    • Linear Regression
    • Logistic Regression
    • Feedforward Neural Network (FNN)
    • Convolutional Neural Network (CNN)
    • Recurrent Neural Network (RNN)
    • Long Short-Term Memory Network (LSTM)
    • Autoencoders (AE)
    • Fully Connected Overcomplete Autoencoders
  • Improving Deep Learning with PyTorch
    • Derivative, Gradient and Jacobian
    • Forward- and Backward-propagation and Gradient Descent (From Scratch FNN Regression)
    • Learning Rate Scheduling
    • Optimization Algorithms
    • Weight Initialization and Activation Functions
  • Deep Reinforcement Learning with PyTorch
    • Supervised to Reinforcement Learning
    • Markov Decision Processes and Bellman Equations
    • Dynamic Programming
  • From Scratch Deep Learning with PyTorch/Python
    • From Scratch Logistic Regression Classification
  • Compute Optimization
    • Speed Optimization Basics Numba

• Machine Learning Tutorials (Libraries: Python, cuDF RAPIDS, cuML RAPIDS, pandas, numpy, scikit-learn and more)

  • RAPIDS cuDF
    • Intro
    • GPU DataFrames
    • CPU/GPU Fractional Differencing

• Programming Tutorials (Libraries: C++, Python, Bash and more)

  • Intro
  • C++
  • Bash
  • Python
  • Javascript
  • Electron

• Data Engineering Tutorials (Libraries: Bash, Databricks, Delta Live Tables, Parquet, Python, Cassandra, and more)

  • Cassandra (NoSQL)
    • Introduction
    • Apache Cassandra Cluster Setup

Stable Deep Learning Tutorials Release

Highlights

Stable releases for deep learning tutorials. And initial launch of deep reinforcement learning, scalable database and programming tutorials.

Stable Sections and Subsections

• Deep Learning and Deep Reinforcement Learning Tutorials (Libraries: Python, PyTorch, Gym, NumPy and Matplotlib)
  • Course Progression
  • Matrices
  • Gradients
  • Linear Regression
  • Logistic Regression
  • Feedforward Neural Network
  • Convolutional Neural Network (CNN)
  • Recurrent Neural Network (RNN)
  • Long Short-Term Memory (LSTM) network
  • Derivative, Gradient and Jacobian
  • Forwardpropagation, Backpropagation and Gradient Descent
  • Learning Rate Scheduling
  • Optimization Algorithms
  • Weight Initialization and Activation Functions
  • Supervised to Reinforcement Learning
  • Markov Decision Processes and Bellman Equations
  • Dynamic Programming
• Programming Tutorials (Libraries: C++, Python, Bash and more)
  • C++
• Scalable Database Tutorials (Libraries: Apache Cassandra, Bash and Python)
  • Apache Cassandra Cluster Setup

First Release of Deep Learning Wizard Materials

About This Release

After careful deliberation, I have decided to gradually open-source our written deep learning materials that I have used to teach more than 3000 students worldwide across 120 countries through my video course, Practical Deep Learning with Pytorch.

As this is a very new effort to open-source my materials, it's still a work-in-progress. Please bear with me while I clean it up.

This repository powers the main open-source Deep Learning Wizard website www.deeplearningwizard.com and contains the materials for the following topics.

Topics Covered

• PyTorch Fundamentals - Matrices
• PyTorch Fundamentals - Gradients
• PyTorch Fundamentals - Linear Regression
• PyTorch Fundamentals - Logistic Regression
• PyTorch Fundamentals - Feedforward Neural Network

Always Latest PyTorch Version

We always provide the latest PyTorch version (currently 0.4 or 1.0 pre-release) so that you will be learning up-to-date code!

About Deep Learning Wizard

We deploy a top-down approach that enables you to grasp deep learning theories and code easily and quickly. We have open-sourced all our materials through our Deep Learning Wizard Wikipedia. For visual learners, feel free to sign up for our video course and join over 2300 deep learning wizards.

To this date, we have taught thousands of students across more than 120+ countries from students as young as 15 to postgraduates and professionals in leading MNCs and research institutions around the world.

Contribution

We are openly calling people to contribute to this repository for errors. Feel free to create a pull request.