Pinned Repositories
autograd-hacks
autograd-lib
bflm
gradient-checkpointing
Make huge neural nets fit in memory
imagenet18
Train ImageNet in 18 minutes on AWS
imagenet18_old
Code to reproduce "imagenet in 18 minutes" DAWN-benchmark entry
ncluster
Lightweight interface to AWS
pytorch-lamb
Implementation of https://arxiv.org/abs/1904.00962
pytorch-sso
PyTorch-SSO: Scalable Second-Order methods in PyTorch
transformer-xl
Training Transformer-XL on 128 GPUs
Cybertron AI's Repositories
cybertronai/gradient-checkpointing
Make huge neural nets fit in memory
cybertronai/imagenet18_old
Code to reproduce "imagenet in 18 minutes" DAWN-benchmark entry
cybertronai/pytorch-lamb
Implementation of https://arxiv.org/abs/1904.00962
cybertronai/autograd-hacks
cybertronai/pytorch-sso
PyTorch-SSO: Scalable Second-Order methods in PyTorch
cybertronai/transformer-xl
Training Transformer-XL on 128 GPUs
cybertronai/imagenet18
Train ImageNet in 18 minutes on AWS
cybertronai/ncluster
Lightweight interface to AWS
cybertronai/autograd-lib
cybertronai/bflm
cybertronai/Megatron-LM
Ongoing research training transformer language models at scale, including: BERT
cybertronai/pytorch-fd
Implementation of fluctuation dissipation relations for automatic learning rate annealing.
cybertronai/aws-network-benchmarks
Tools to benchmark AWS network performance
cybertronai/pytorch-aws
Example code for "PyTorch on AWS made easy"
cybertronai/autotune