README.md

Template codes for Deep learning with Pytorch

This repo provides different pytorch implementation for training a deep learning model. It uses a simple classification task example for CIFAR10 Dataset to show:

1. A simple vanilla, single-GPU implementation
2. A multi-GPU, distributed data parallel implementation
3. A Pytorch-ligtning implementation along with tracking and visualization in TensorBoard
4. A Pytorch-ligtning Hydra implementation for rapid experimentation and prototyping using new models/datasets

Single-GPU implementation

Multi-GPU distributed data parallel implementation

Pytorch-lightning implementation

Pytorch-lightning Hydra implementation