Skip to content

Easily launch PyTorch functions onto multiple machines & GPUs

License

Notifications You must be signed in to change notification settings

apoorvkh/torchrunx

Repository files navigation

torchrunx 🔥

PyPI - Python Version PyTorch Version PyPI - Version Tests Docs GitHub License

By Apoorv Khandelwal and Peter Curtin

Automatically distribute PyTorch functions onto multiple machines or GPUs

Installation

pip install torchrunx

Requires: Linux (with shared filesystem & SSH access if using multiple machines)

Demo

Here's a simple example where we "train" a model on two nodes (with 2 GPUs each).

Training code
import os
import torch

def train():
    rank = int(os.environ['RANK'])
    local_rank = int(os.environ['LOCAL_RANK'])

    model = torch.nn.Linear(10, 10).to(local_rank)
    ddp_model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[local_rank])
    optimizer = torch.optim.AdamW(ddp_model.parameters())

    optimizer.zero_grad()
    outputs = ddp_model(torch.randn(5, 10))
    labels = torch.randn(5, 10).to(local_rank)
    torch.nn.functional.mse_loss(outputs, labels).backward()
    optimizer.step()

    if rank == 0:
        return model

You could also use transformers.Trainer (or similar) to automatically handle all the multi-GPU / DDP code above.

import torchrunx as trx

if __name__ == "__main__":
    result = trx.launch(
        func=train,
        hostnames=["localhost", "other_node"],
        workers_per_host=2  # number of GPUs
    )

    trained_model = result.rank(0)
    torch.save(trained_model.state_dict(), "model.pth")

Why should I use this?

Whether you have 1 GPU, 8 GPUs, or 8 machines:

Features

  • Our launch() utility is super Pythonic
    • Return objects from your workers
    • Run python script.py instead of torchrun script.py
    • Launch multi-node functions, even from Python Notebooks
  • Fine-grained control over logging, environment variables, exception handling, etc.
  • Automatic integration with SLURM

Robustness

  • If you want to run a complex, modular workflow in one script
    • don't parallelize your entire script: just the functions you want!
    • no worries about memory leaks or OS failures

Convenience

  • If you don't want to:
    • set up dist.init_process_group yourself
    • manually SSH into every machine and torchrun --master-ip --master-port ..., babysit failed processes, etc.