Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

[RFC] Deprecate the `unfreeze_milestones` finetuning strategy?

See original GitHub issue

Motivation

The unfreeze_milestones finetuning strategy is confusing:

how is the layer number interpretted? Does this include e.g. batch norm and non-linearity layers? Not documented
what’s the use case? Not aware of any time this would be recommended (also not documented)

Alternatives

At least document the answers to the above questions if there are any.

Issue Analytics

State:
Created a year ago
Comments:5 (4 by maintainers)

Top GitHub Comments

1reaction

karthikrangasaicommented, Mar 30, 2022

Great. Sounds good. I will try to get some work done on this then.

0reactions

stale[bot]commented, Jun 5, 2022

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

Top Results From Across the Web

Revisiting Fine-tuning Strategy for Few-shot Learning - arXiv

Abstract: The goal of few-shot learning is to learn a classifier that can recognize unseen classes from limited support data with labels.

RFC 8996: Deprecating TLS 1.0 and TLS 1.1 - » RFC Editor

Deprecating TLS 1.0 and TLS 1.1. Abstract. This document formally deprecates Transport Layer Security (TLS) versions 1.0 (RFC 2246) and.

Finetuning — Flash documentation

Finetuning (or transfer-learning) is the process of tweaking a model trained ... Choose a finetune strategy (example: “freeze”) and call flash.core.trainer.

Transfer learning and fine-tuning | TensorFlow Core

To learn how to use non-trainable weights in your own custom layers, see the guide to writing new layers from scratch.