Hi, is there an equivalent to Tensorflow’s stop_gradient/PyTorch’s detach in FluxML? Thanks! Rasmus

Hi, I don’t know if there is anything built in, but this issue has a good tip in the comments for how to do it yourself: https://github.com/FluxML/Zygote.jl/issues/175 Summary: stop_gradient(f) = f() Zygote.@nograd stop_gradient # Now you can use it like this (note: does not work in REPL due to …

Stop_gradient/detach equivalent in flux

Specific Domains Machine Learning

Rassibassi December 11, 2019, 8:34pm 3

Great! Thanks a lot, it made my code work example over here: Autoencoder for telecommunication (Constellation shaping)

Topic		Replies	Views
Which function in julia.Flux is similar to "detach()" in pytorch? Machine Learning	2	69	August 29, 2024
Stop gradients from propagating through some branches of computation tree in Flux General Usage flux , machine-learning	9	2579	February 4, 2020
Pullback for both Flux model's inputs and parameters Machine Learning question , flux , zygote , autodiff	9	788	June 26, 2022
Hessian Vector products in Flux with Zygote Machine Learning flux , zygote	4	1000	December 22, 2021
Mutate Zygote Gradients with a Custom Mask Before Update? Machine Learning flux , zygote	1	566	March 25, 2021

Stop_gradient/detach equivalent in flux

Related topics