Graphcore/vit-base-ipu

Optimum Graphcore is a new open-source library and toolkit that enables developers to access IPU-optimized models certified by Hugging Face. It is an extension of Transformers, providing a set of performance optimization tools enabling maximum efficiency to train and run models on Graphcore’s IPUs - a completely new kind of massively parallel processor to accelerate machine intelligence. Learn more about how to take train Transformer models faster with IPUs at hf.co/hardware/graphcore.

Through HuggingFace Optimum, Graphcore released ready-to-use IPU-trained model checkpoints and IPU configuration files to make it easy to train models with maximum efficiency in the IPU. Optimum shortens the development lifecycle of your AI models by letting you plug-and-play any public dataset and allows a seamless integration to our State-of-the-art hardware giving you a quicker time-to-value for your AI project.

Model description

The Vision Transformer (ViT) is a model for image recognition that employs a Transformer-like architecture over patches of the image which was widely used for NLP pretraining.

It uses a standard Transformer encoder as used in NLP and simple, yet scalable, strategy works surprisingly well when coupled with pre-training on large amounts of dataset and tranferred to multiple size image recognition benchmarks while requiring substantially fewer computational resources to train.

Paper link : AN IMAGE IS WORTH 16X16 WORDS:TRANSFORMERS FOR IMAGE RECOGNITION AT SCALE

Intended uses & limitations

This model contains just the IPUConfig files for running the ViT base model (e.g. vit-base-patch16-224-in21k or deit-base-patch16-384) on Graphcore IPUs.

This model contains no model weights, only an IPUConfig.

Usage

from optimum.graphcore import IPUConfig
ipu_config = IPUConfig.from_pretrained("Graphcore/vit-base-ipu")