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 >= 3.8 and PyTorch >= 2.0.0

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-2024.6.29.tar.gz (93.0 kB view details)

Uploaded Source

Built Distribution

torchsnapshot_nightly-2024.6.29-py3-none-any.whl (84.6 kB view details)

Uploaded Python 3

File details

Details for the file torchsnapshot_nightly-2024.6.29.tar.gz.

File metadata

File hashes

Hashes for torchsnapshot_nightly-2024.6.29.tar.gz
Algorithm Hash digest
SHA256 1e5099dfa8291dd6e111112d66b513731b3cb40549e19e631c2555e91c16c3e3
MD5 4285718203bad3a1e4d59ddd58062bea
BLAKE2b-256 6d84990a3ca88a2dedb25ca1056778d1656a2d2fbc7185ec94ebec27602af7be

See more details on using hashes here.

Provenance

File details

Details for the file torchsnapshot_nightly-2024.6.29-py3-none-any.whl.

File metadata

File hashes

Hashes for torchsnapshot_nightly-2024.6.29-py3-none-any.whl
Algorithm Hash digest
SHA256 6bdfd0c6abb2fc65e2fb97f6e8564995a461e7c7f7e6d9e02d3919bb609f6427
MD5 23eec8b8962420d9271fd7043401919a
BLAKE2b-256 9e865d7d09967796dd0be80d474d48e93ee18e84511d4f1d4956916745db03f2

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