Examples

We provide several examples using ignite to display how it helps to write compact and full-featured training loops in several lines of code:

MNIST example

Basic neural network training on MNIST dataset with/without ignite.contrib module:

• MNIST with ignite.contrib TQDM/Tensorboard/Visdom loggers

• MNIST with native TQDM/Tensorboard/Visdom logging

These examples are ported from pytorch/examples.

Distributed examples

Training a ResNet on CIFAR10 in various configurations:

1. single gpu

2. single node multiple gpus

3. multiple nodes and multilple gpus

4. single or multiple TPUs

• CIFAR10 - This example displays usage of ignite.distributed helper module.

Other examples

• DCGAN - plain Deep Convolution Generative Adversarial Networks training

• Reinforcement Learning - Actor/Critic and Reinforce methods on Cart-Pole task

• Fast Neural Style - Artistic style transfer implementation.

These examples are ported from pytorch/examples.

Notebooks

• Text Classification using Convolutional Neural Networks

• Variational Auto Encoders

• Training Cycle-GAN on Horses to Zebras with Nvidia/Apex

• Another training Cycle-GAN on Horses to Zebras with Native Torch CUDA AMP

• Finetuning EfficientNet-B0 on CIFAR100

• Convolutional Neural Networks for Classifying Fashion-MNIST Dataset

• Hyperparameters tuning with Ax

• Basic example of LR finder on MNIST

• Benchmark mixed precision training on Cifar100: torch.cuda.amp vs nvidia/apex

• MNIST training on a single TPU

All notebooks can be opened on Google Colab with a link:

https://colab.research.google.com/github/pytorch/ignite/blob/master/examples/notebooks/<notebook-name>

Reproducible trainings

Inspired by torchvision/references, we provide several reproducible baselines for vision tasks:

• ImageNet

• Pascal VOC2012

Features:

• Distributed training with mixed precision by nvidia/apex

• Experiments tracking with MLflow or Polyaxon or TRAINS