Time to Skip Tedious Steps – Spare Efforts with PyTorch Lightning

This repository demonstrates the usage of PyTorch Lightning, designed to support the talk "Time to Skip Tedious Steps – Spare Efforts with PyTorch Lightning" at PyCon Hong Kong 2024. The goal is to showcase how PyTorch Lightning can simplify and accelerate deep learning workflows, allowing developers to focus on innovation rather than boilerplate code.

Usage

git clone https://github.com/wyhwong/PyConHK2024-PyTorch-Lightning.git
cd PyConHK2024-PyTorch-Lightning

See README.md for more details.

Abstract of Talk

With the rapid advancement in deep learning, models become super large and consume significant resources, making efficiency and simplicity more critical than ever. In this talk, we introduce PyTorch Lightning, a deep learning framework that emerges as a powerful tool that streamlines the process of building, training, and scaling models, allowing researchers and practitioners to focus on what truly matters: innovation.

We will begin with an overview of PyTorch Lightning, discussing the key benefits it offers over traditional PyTorch. We will explore how PyTorch Lightning abstracts away the boilerplate code associated with model training, making it easier to implement and experiment with complex models. Then, we walk through the process of training a ResNet in PyTorch Lightning for image classification task and explore some advanced features in PyTorch Lightning.

Related Resources

• Powerpoint Slides: View on OneDrive

• Talk Homepage: Pretalx

• Studio Template: lightning.ai

• Talk Video: TBC

• PyTorch Lightning: https://lightning.ai/

• PyTorch: https://pytorch.org/

Author

@wyhwong