Skip to main content

A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind.

Project description

TorchSnapshot (Beta Release)

build status pypi version conda version pypi nightly version codecov bsd license

A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind.

Install

Requires Python >= 3.7 and PyTorch >= 1.12

From pip:

# Stable
pip install torchsnapshot
# Or, using conda
conda install -c conda-forge torchsnapshot

# Nightly
pip install --pre torchsnapshot-nightly

From source:

git clone https://github.com/pytorch/torchsnapshot
cd torchsnapshot
pip install -r requirements.txt
python setup.py install

Why TorchSnapshot

Performance

  • TorchSnapshot provides a fast checkpointing implementation employing various optimizations, including zero-copy serialization for most tensor types, overlapped device-to-host copy and storage I/O, parallelized storage I/O.
  • TorchSnapshot greatly speeds up checkpointing for DistributedDataParallel workloads by distributing the write load across all ranks (benchmark).
  • When host memory is abundant, TorchSnapshot allows training to resume before all storage I/O completes, reducing the time blocked by checkpoint saving.

Memory Usage

  • TorchSnapshot's memory usage adapts to the host's available resources, greatly reducing the chance of out-of-memory issues when saving and loading checkpoints.
  • TorchSnapshot supports efficient random access to individual objects within a snapshot, even when the snapshot is stored in a cloud object storage.

Usability

  • Simple APIs that are consistent between distributed and non-distributed workloads.
  • Out of the box integration with commonly used cloud object storage systems.
  • Automatic resharding (elasticity) on world size change for supported workloads (more details).

Security

  • Secure tensor serialization without pickle dependency [WIP].

Getting Started

from torchsnapshot import Snapshot

# Taking a snapshot
app_state = {"model": model, "optimizer": optimizer}
snapshot = Snapshot.take(path="/path/to/snapshot", app_state=app_state)

# Restoring from a snapshot
snapshot.restore(app_state=app_state)

See the documentation for more details.

License

torchsnapshot is BSD licensed, as found in the LICENSE file.

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

torchsnapshot-nightly-2023.8.23.tar.gz (53.3 kB view details)

Uploaded Source

Built Distribution

torchsnapshot_nightly-2023.8.23-py3-none-any.whl (70.5 kB view details)

Uploaded Python 3

File details

Details for the file torchsnapshot-nightly-2023.8.23.tar.gz.

File metadata

File hashes

Hashes for torchsnapshot-nightly-2023.8.23.tar.gz
Algorithm Hash digest
SHA256 c98bb4d4211bdcb0c1b4527a611ab7465c1c80bbd5020c67a0f9d5933ee2cb48
MD5 f2789256279cb0844760d9c6114fc028
BLAKE2b-256 e342426e965fdc4efc16ef22c008c36e99be12f6d037f90eb3f3a50fc5a818a7

See more details on using hashes here.

Provenance

File details

Details for the file torchsnapshot_nightly-2023.8.23-py3-none-any.whl.

File metadata

File hashes

Hashes for torchsnapshot_nightly-2023.8.23-py3-none-any.whl
Algorithm Hash digest
SHA256 7feca13fb8409029ade946aad4a8dc861a2d1ad32f97abcee2027d02a4af246c
MD5 4353faa7c3db984d1e0475557d50aa0d
BLAKE2b-256 2088d52a09a81b02e8d7c036d924ed98c27c650b051d572e6de876179509580b

See more details on using hashes here.

Provenance

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page