lsds/Crossbow
Crossbow: A Multi-GPU Deep Learning System for Training with Small Batch Sizes
This system helps deep learning practitioners train complex models more efficiently on multi-GPU setups, especially when working with small datasets or batch sizes. It takes raw data (like ImageNet or MNIST) and deep learning model configurations, then outputs a trained model faster than traditional methods. Data scientists and AI researchers who develop and train neural networks will find this useful.
No commits in the last 6 months.
Use this if you are a deep learning engineer or researcher looking to speed up training of your neural network models on multiple GPUs, particularly when using small batch sizes that don't fully utilize GPU resources.
Not ideal if you primarily work with very large batch sizes or are looking for a simple, out-of-the-box solution for single-GPU training, as this system requires manual setup and configuration.
Stars
56
Forks
7
Language
Java
License
Apache-2.0
Category
Last pushed
Oct 05, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/lsds/Crossbow"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
deepspeedai/DeepSpeed
DeepSpeed is a deep learning optimization library that makes distributed training and inference...
helmholtz-analytics/heat
Distributed tensors and Machine Learning framework with GPU and MPI acceleration in Python
hpcaitech/ColossalAI
Making large AI models cheaper, faster and more accessible
horovod/horovod
Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.
bsc-wdc/dislib
The Distributed Computing library for python implemented using PyCOMPSs programming model for HPC.