Welcome to my journey of learning PyTorch! This Blog documents my progress, insights, and projects as I delve into the world of deep learning using PyTorch.

Table of Contents

1. Introduction
2. Learning Goals
3. Progress Tracker
4. Projects
5. Resources
6. Future Plans

Introduction

I am currently exploring PyTorch, a powerful deep learning library developed by Facebook’s AI Research lab. My goal is to gain a solid understanding of its capabilities, build machine learning models, and eventually contribute to real-world AI projects.

Learning Goals

• Understand the basics of PyTorch, including tensors, autograd, and neural networks.
• Implement basic machine learning algorithms from scratch.
• Explore and use pre-trained models for various tasks.
• Gain hands-on experience with PyTorch’s neural network modules.
• Experiment with custom models for specific tasks, including handwriting recognition.

Progress Tracker

Date	Topic	Notes
2024-01-01	Introduction to Tensors	Learned about tensors and basic operations.
2024-09-05	Autograd and Backpropagation	Explored how PyTorch handles gradients and automatic differentiation.
2024-09-10	Neural Networks	Built a simple neural network for classification tasks.
2024-09-15	CNNs and Image Classification	Implemented a CNN for image classification using the CIFAR-10 dataset.

Source: Link

Projects

• Image to Text Converter: A Flask application that converts images to text. I’m working on integrating a custom model trained in PyTorch to recognize handwritten text.

Resources

• Official PyTorch Documentation
• Deep Learning with PyTorch: A 60 Minute Blitz
• CS50P - For foundational knowledge.

Future Plans

• Advanced Model Training: Delve deeper into training custom models with larger datasets.
• PyTorch Lightning: Explore this high-level wrapper to streamline the training process.
• Contribution: Contribute to open-source PyTorch projects and share my work with the community.

---