framework-internalsdevelopment
pytorch-distributed
Distributed training strategies including DistributedDataParallel (DDP) and Fully Sharded Data Parallel (FSDP). Covers multi-node setup, checkpointing, and process management using torchrun. (ddp, fsdp, distributeddataparallel, torchrun, nccl, rank, process-group)
maintainer
cuba6112
Обновлено 12/26/2025
Звёзды
0
Форки
0
quick start
Installation and usage
Distributed training strategies including DistributedDataParallel (DDP) and Fully Sharded Data Parallel (FSDP). Covers multi-node setup, checkpointing, and process management using torchrun. (ddp, fsdp, distributeddataparallel, torchrun, nccl, rank, process-group)
Установка
$ install --globalskills.sh
Использование
После установки вы можете использовать этот skill, выполнив следующую команду в терминале:
skills use pytorch-distributed