Providing reproducibility in deep learning frameworks
Project description
This package provides patches and tools related to determinism (bit-accurate, run-to-run reproducibility) in deep learning frameworks, with a focus on determinism when running on GPUs, and a tool (Seeder) for reducing variance in deep learning frameworks.
For further information, see the documentation in the associated open-source repository: GitHub/NVIDIA/framework-reproducibility
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
Built Distribution
Close
Hashes for framework-reproducibility-0.4.0.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | bd776444529e64d137dee0934974a96ce9a4afb1e36b008077217de102cf5be9 |
|
MD5 | 3720fb1b96fd73d84d063017a131ec8b |
|
BLAKE2b-256 | 0aa5cd245ff6ce5872a81288c59e17dd7c39361cc840a9eb7bcebd5e58d47dd4 |
Close
Hashes for framework_reproducibility-0.4.0-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 372f266700537804b0da673f4a329f968f2be0f1d524df2de2856ba9384ab3d2 |
|
MD5 | 8394010e13f13d5150e390492f21d6c6 |
|
BLAKE2b-256 | 471786be8fe469810807cecf9042efcf44612194395d397e2198eda5ffbb0695 |