Machine Learning & Deep Learning Projects with PyTorch

This repository contains a series of Jupyter Notebooks that explore fundamental concepts in machine learning and deep learning, from basic algorithms to advanced neural network architectures. Each notebook serves as a practical, hands-on implementation of a key topic using PyTorch.

---

Projects

The projects are ordered to build upon foundational concepts, creating a clear learning path.

#	Topic	Description
1	Perceptron	An implementation of the simplest single-neuron network for binary classification.
2	Gradient Descent	A deep dive into the foundational optimization algorithm used to train neural networks.
3	Universal Function Approximator	A demonstration of how neural networks can be used to approximate any continuous function.
4	Multi-Output & Batch Processing	Techniques for handling models with multiple outputs and processing data in batches for efficiency.
5	Classification in PyTorch	Building a complete multi-class image classification model using core PyTorch components.
6	Convolutional Networks (CNNs)	Implementing a CNN from scratch to solve image recognition tasks.
7	Transfer Learning	Leveraging a pre-trained model (like ResNet) and fine-tuning it for a new, custom dataset.
8	Open-Set Classification	Tackling the challenge of identifying samples that do not belong to any of the known training classes.
9	Convolutional Auto-Encoder	Building an unsupervised neural network for dimensionality reduction and feature extraction on images.
10	Learn to Write Like Shakespeare	Implementing an Elman network for character-level text generation to create new text in the style of Shakespeare.

---

Getting Started

To run these notebooks, you will need a Python environment with Jupyter Notebook or JupyterLab.

Installation

1. Clone the repository:

   git clone git@github.com:acd17sk/Deep-Learning-Projects.git
   cd Deep-Learning-Projects

2. Install the required libraries. A requirements.txt file is recommended, but you can also install the core packages manually:

   pip install torch torchvision numpy matplotlib scikit-learn pandas

Usage

Navigate to the repository directory and launch Jupyter:

jupyter notebook

Then, open any of the .ipynb files to explore the projects.

---

Libraries Used

• Python 3
• PyTorch
• NumPy
• Matplotlib
• Scikit-learn