Lightning-nc

Latest version: v0.0.3

Safety actively analyzes 623909 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

0.0.3

Only applies pruners' `on_before_optimizer_steps` and `on_after_optimizer_step` after their `start_step`.

Those two hooks compute a score, which might depend on compute the gradient of some parameters as it is the case for SNIP-based methods. Applying them before the `start_step` causes issues for example in finetuning paradigms, where some parameters might see their parameters frozen before any pruning strategy.

**Full Changelog**: https://github.com/clementpoiret/lightning-nc/compare/v0.0.2...v0.0.3

0.0.2

Introduced a new finetuning callback to use Intel(R) Neural Compressor inside a finetuning paradigm.

**Full Changelog**: https://github.com/clementpoiret/lightning-nc/compare/v0.0.1...v0.0.2

0.0.1

Links

Releases

Has known vulnerabilities

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.