KerasCV and KerasNLP: Vision and Language Power-Ups

TL;DR

KerasCV and KerasNLP are open-source extensions of Keras that facilitate flexible, high-performance computer vision and NLP workflows across multiple frameworks, with pretrained models and efficient training capabilities.

Contribution

Introduction of modular, layered Keras domain packages for vision and language tasks, supporting multiple backends and pretrained models with optimized training pipelines.

Findings

Support for JAX, TensorFlow, PyTorch

Pretrained models for popular architectures

Efficient training with XLA and tf.data

Abstract

We present the Keras domain packages KerasCV and KerasNLP, extensions of the Keras API for Computer Vision and Natural Language Processing workflows, capable of running on either JAX, TensorFlow, or PyTorch. These domain packages are designed to enable fast experimentation, with a focus on ease-of-use and performance. We adopt a modular, layered design: at the library's lowest level of abstraction, we provide building blocks for creating models and data preprocessing pipelines, and at the library's highest level of abstraction, we provide pretrained ``task" models for popular architectures such as Stable Diffusion, YOLOv8, GPT2, BERT, Mistral, CLIP, Gemma, T5, etc. Task models have built-in preprocessing, pretrained weights, and can be fine-tuned on raw inputs. To enable efficient training, we support XLA compilation for all models, and run all preprocessing via a compiled graph of TensorFlow operations using the tf.data API. The libraries are fully open-source (Apache 2.0 license) and available on GitHub.