Skip to main content

opake-dpsgd

DP-SGD mechanisms for Opake: Gaussian noise, clipping (fixed, AUTO-S, adaptive), and Poisson subsampling. Functional optimizers (including the universal adamw with optional DP bias correction) live in opake.optimizers.

Install

Install the root package as described in the repository installation guide. opake-dpsgd is included in the default opake package set.

Quick start

from opake.dpsgd.clipping import auto_clipped_grad, clipped_grad
from opake.random import key
from opake.dpsgd.clipping import adaptive_clipped_grad
from opake.dpsgd.noise import gaussian_noise
from opake.dpsgd.sampling import PoissonSampler

Layout

  • opake.dpsgd.noise — gaussian_noise
  • opake.dpsgd.clipping — clipped_grad, auto_clipped_grad, per_group, adaptive_clipped_grad, .types, .fun
  • opake.dpsgd.sampling — PoissonSampler (optional truncated_batch_size)

RNG keys, pytree helpers, distributed plumbing, and serialization live in opake-engine.

Metadata

Release files for opake-dpsgd 0.16.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distribution (wheel)

Table of built distributions (wheels) for opake-dpsgd 0.16.0
File Interpreter ABI Platform
opake_dpsgd-0.16.0-py3-none-any.whl Python 3 none any Details

Release files / opake_dpsgd-0.16.0-py3-none-any.whl

Download URL opake_dpsgd-0.16.0-py3-none-any.whl
Size 49.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
09d30279a8400070a5e73101c57cf762d6a608599ea8f99969ecd672b4513fef
BLAKE2b-256 checksum
How to use checksums
0fee22ce0293cfe58af468fe27537b8b77e4330b995f5de59a1ccac4f2a3fc36
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 23, 2026.

Transparency log

Release history Release notifications | RSS feed

0.16.1

1 release file

This release

0.16.0 This release

1 release file

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page