machine learning framework / deep learning
TensorFlow is an end-to-end open source machine learning framework for building and deploying ML models, with Python and C++ APIs.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
Full deep learning framework with similar scope to TF; choose for a China-supported stack with integrated distributed training and inference.
PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
JAX-based network modules; choose if you're already on JAX and want a lightweight TF-free alternative to tf.Module.
JAX-based neural network library
High-level training library on PyTorch; choose for a concise API if you prefer the PyTorch ecosystem.
The fastai deep learning library
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
Older distributed-training framework for extremely large models; choose only if you need its specific fault-tolerant orchestration and can tolerate stale code.
Fault-tolerant, highly scalable GPU orchestration, and a machine learning framework designed for training models with billions to trillions of parameters
These projects were analysed and named tensorflow among their alternatives. The relationship is not symmetric — how tensorflow rates them is a separate judgement, made when tensorflow is analysed in its own right.
calls tensorflow “A mature, widely adopted deep learning framework with a massive ecosystem; choose it for broad community support and production deployments.”
PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）
calls tensorflow “choose when you need a mature production ecosystem with TPU support and extensive deployment tools”
OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.
calls tensorflow “Choose when you need a mature, low-level framework with a full ecosystem and production tooling; heavier than Keras.”
Deep Learning for humans
calls tensorflow “when you need the full TensorFlow framework in Python/C++ for server/mobile, not browser.”
A WebGL accelerated JavaScript library for training and deploying ML models.
calls tensorflow “When you need TensorFlow's production deployment ecosystem rather than CoreNet's PyTorch research stack.”
CoreNet: A library for training deep neural networks