Finding non-deterministic functions in pytorch code
Project description
torchdetscan: rooting out non-determinism in pytorch code
A linter for deep learning non-determinism, randomness and reproducibility with applications to scientific computing and simulation.
Usage
usage: torchdetscan [-h] [--verbose] [--pytorch-version {1.6.0,1.7.0,1.7.1,1.8.0,1.8.1,1.9.0,1.9.1,1.10,1.11,1.12,1.13,2.0,2.1,2.2,2.3}] path
torchdetscan is a linter for finding non-deterministic functions in pytorch code.
positional arguments:
path Path to the file or directory in which to recursively lint
options:
-h, --help show this help message and exit
--verbose, -v Enable chatty output
--pytorch-version {1.6.0,1.7.0,1.7.1,1.8.0,1.8.1,1.9.0,1.9.1,1.10,1.11,1.12,1.13,2.0,2.1,2.2,2.3}, -ptv {1.6.0,1.7.0,1.7.1,1.8.0,1.8.1,1.9.0,1.9.1,1.10,1.11,1.12,1.13,2.0,2.1,2.2,2.3}
Version of Pytorch to use for checking
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
torchdetscan-0.1.tar.gz
(6.6 kB
view hashes)
Built Distribution
Close
Hashes for torchdetscan-0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | d181f259215b2700b8a11ca69385e91b5c953a23fdab03227d7d33f440116a9b |
|
MD5 | 884371f65f86f45f2e6f3ca1a6da7455 |
|
BLAKE2b-256 | 1735d5050bbc9c2d42c18a5cdb4a27218106dfed1b2d1e2f0d6b59e4d1b5c84f |