Skip to content

@cybertronai Cybertron AI

Change the repository type filter

All

    Repositories list

    15 repositories

    • imagenet18

      Public
      Train ImageNet in 18 minutes on AWS
      Python
      •
      The Unlicense
      •19•126•6•1•Updated Mar 20, 2024Mar 20, 2024

    • pytorch-sso

      Public
      PyTorch-SSO: Scalable Second-Order methods in PyTorch
      Python
      •
      MIT License
      •20•142•5•0•Updated Oct 1, 2023Oct 1, 2023

    • autograd-hacks

      Public
      Python
      •
      The Unlicense
      •32•155•7•1•Updated Jun 8, 2022Jun 8, 2022

    • autotune

      Public
      Python
      •0•0•0•0•Updated Dec 11, 2020Dec 11, 2020

    • pytorch-lamb

      Public
      Implementation of https://arxiv.org/abs/1904.00962
      Python
      •
      MIT License
      •49•370•4•0•Updated Dec 9, 2020Dec 9, 2020

    • transformer-xl

      Public
      Training Transformer-XL on 128 GPUs
      Python
      •
      Apache License 2.0
      •762•140•12•0•Updated Jun 11, 2020Jun 11, 2020

    • autograd-lib

      Public
      Python
      •
      MIT License
      •7•33•2•1•Updated May 21, 2020May 21, 2020

    • gradient-checkpointing

      Public
      Make huge neural nets fit in memory
      Python
      •
      MIT License
      •271•2.7k•27•1•Updated Apr 26, 2020Apr 26, 2020

    • imagenet18_old

      Public
      Code to reproduce "imagenet in 18 minutes" DAWN-benchmark entry
      Python
      •
      The Unlicense
      •160•846•10•0•Updated Nov 24, 2019Nov 24, 2019

    • ncluster

      Public
      Lightweight interface to AWS
      Python
      •
      MIT License
      •5•47•53•0•Updated Oct 15, 2019Oct 15, 2019

    • pytorch-aws

      Public
      Example code for "PyTorch on AWS made easy"
      Python
      •0•5•0•0•Updated Aug 27, 2019Aug 27, 2019

    • aws-network-benchmarks

      Public
      Tools to benchmark AWS network performance
      Jupyter Notebook
      •2•6•0•0•Updated Jul 31, 2019Jul 31, 2019

    • pytorch-fd

      Public
      Implementation of fluctuation dissipation relations for automatic learning rate annealing.
      pytorchlearning-rate-scheduling
      Python
      •
      MIT License
      •0•8•0•0•Updated Jul 24, 2019Jul 24, 2019

    • bflm

      Public
      Python
      •7•16•1•0•Updated Jun 8, 2019Jun 8, 2019

    • Megatron-LM

      Public
      Ongoing research training transformer language models at scale, including: BERT
      Python
      •
      Other
      •2.4k•15•0•0•Updated Apr 25, 2019Apr 25, 2019