Machine Learning Practice

This repository contains a comprehensive collection of Jupyter notebooks and Python scripts for implementing and understanding key concepts in machine learning, deep learning, and PyTorch. It serves as a hands-on guide for learners and students for the course Applied Machine Learning, focusing on practical implementations and a final project example.

---

Table of Contents

1. Getting Started
2. Project Structure
3. Key Notebooks and Concepts
4. Additional Scripts
5. Installation
6. Usage
7. License

---

Getting Started

This repository provides an applied approach to understanding machine learning concepts, using Python and the PyTorch deep learning framework.

---

Project Structure

The repository is organized as follows:

Machine-Learning-Practice/
|
├── Basics of PyTorch and OOP.ipynb
├── Basics to Intermediate Concepts with Python.ipynb
├── Basics to Intermediate NumPy Concepts with Python.ipynb
├── Batch Norm_Learning Rate_Normal vs Gradient Descent.ipynb
├── Convolutional Neural Networks.ipynb
├── Data Augmentation.ipynb
├── Introduction to Python.ipynb
├── Introduction to Random Variables, Probability, and Likelihood.ipynb
├── Linear Regression.ipynb
├── Logistic Regression and Binary Classification.ipynb
├── MNIST Classification using PyTorch Lightning.ipynb
├── Multi-Layer Perceptron (MLP) Model with PyTorch.ipynb
├── NumPy with Python.ipynb
├── Recurrent Neural Networks.ipynb
├── Regression Model With Pytorch.ipynb
├── Regression With Pytorch.ipynb
├── Transformers with Pytorch.ipynb
├── F107 Solar Radio Flux CNN-Transformer.py
└── F107 Solar Radio Flux Plotter.py

---

Key Notebooks and Concepts

Here are the highlights of each notebook:

1. Basics and Python Foundations

• Introduction to Python.ipynb: Introduction to Python programming for beginners.
• NumPy with Python.ipynb: Fundamental concepts and operations using NumPy.
• Basics to Intermediate Concepts with Python.ipynb: Python programming exercises for ML foundations.
• Basics to Intermediate NumPy Concepts with Python.ipynb: Essential and advanced operations with NumPy.

2. Regression Models

• Linear Regression.ipynb: Implementation of linear regression from scratch.
• Logistic Regression and Binary Classification.ipynb: Logistic regression for binary classification tasks.
• Regression Model With Pytorch.ipynb: Building regression models using PyTorch.
• Regression With Pytorch.ipynb: Advanced PyTorch-based regression models.

3. Deep Learning with PyTorch

• Basics of PyTorch and OOP.ipynb: Introduction to PyTorch and object-oriented programming.
• Batch Norm_Learning Rate_Normal vs Gradient Descent.ipynb: Understanding batch normalization and learning rates.
• Multi-Layer Perceptron (MLP) Model with PyTorch.ipynb: Implementing MLPs for classification.

4. Convolutional Neural Networks (CNNs)

• Convolutional Neural Networks.ipynb: Introduction and implementation of CNNs.
• Data Augmentation.ipynb: Techniques to augment data for improving CNN performance.
• MNIST Classification using PyTorch Lightning.ipynb: Using PyTorch Lightning to classify handwritten digits.

5. Sequence Models

• Recurrent Neural Networks.ipynb: Implementing RNNs for sequential data.
• Transformers with Pytorch.ipynb: Understanding and applying Transformers using PyTorch.

6. Probability and Statistics

• Introduction to Random Variables, Probability, and Likelihood.ipynb: Core statistical concepts for ML.

---

Additional Scripts

These Python scripts focus on advanced applications and visualizations for time series data:

F107 Solar Radio Flux Prediction

1. F107 Solar Radio Flux CNN-Transformer.py

• Purpose: Predicts the 10.7-cm solar radio flux using a hybrid CNN-Transformer model.
• Key Features:
  • Preprocesses time series data for solar flux measurements.
  • Implements a hybrid model combining CNN (for local features) and Transformer Encoder (for global dependencies).
  • Provides visualizations such as residual plots, loss curves, and prediction scatter plots.
• Input: solar_flux.txt (Solar flux measurements dataset).
• Output: Predictions, residual plots, and performance metrics for Obs, Adj, and URSI-D columns.

2. F107 Solar Radio Flux Plotter.py

• Purpose: Visualizes solar flux measurements.
• Key Features:
  • Cleans and preprocesses solar flux data.
  • Generates combined and separate time series plots for Obs, Adj, and URSI-D columns.
  • Ensures data consistency by handling missing and invalid values.
• Output: Plots showcasing trends in solar flux measurements.

---

Installation

To use the notebooks and scripts in this repository, follow these steps:

1. Clone the repository:

   git clone https://github.com/Keeby-Astro/Machine-Learning-Practice.git
   cd Machine-Learning-Practice

2. Install the required Python libraries:

   pip install -r requirements.txt

---

Usage

1. Open Jupyter Notebook:

   jupyter notebook

2. Navigate to the desired notebook or script and execute cells to explore the concepts and run code.
3. For standalone Python scripts, run them directly using:

   python [name_of_specific_script].py

License

This project is licensed under the MIT License.

---

Acknowledgments

Special thanks to:

• PyTorch developers and the open-source community.
• Dr. Iman Ghamarian, University of Oklahoma professor for Applied Machine Learning