FluxML / FluxML/NNlib.jl

remove internal usage of within_gradient

Open
#756 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Julia
Stars
258
Forks
138
Avg merge
2d 1h
Merged PRs (30d)
3

Description

`within_gradient` currently is not supported by Enzyme and Mooncake. Either remove internal usage (e.g. in softmax second order derivatives) or extend support to all AD frameworks.

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by locating internal uses of `within_gradient`, especially in softmax second-order derivatives, and review how Enzyme and Mooncake handle the operation. Determine whether the scope is removing those uses or extending support across all AD frameworks; done means the chosen approach works with both named frameworks and preserves the affected derivative behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia
Domain
machine-learning
Issue type
Refactor
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
38/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.