I tried these two functions to get the l2 regularization term for a NN model: function ℓ₂(a) s = zero(Float32) for i in a s += sum(i.^2) end s end function l₂(a) s = zero(Float32) Lₐ = length(a) for i in 1:Lₐ @inbounds Lᵢ = length(a[i]) for j in …

This can be done with a one-liner :slight_smile: . Regularisation · Flux has some good examples.

Sum of Flux.params

ToucheSir February 4, 2021, 8:32pm 2

This can be done with a one-liner . Regularisation · Flux has some good examples.

Topic		Replies	Views
How to do L2 regularization with new Flux and Zygote Machine Learning flux	2	2164	December 29, 2019
Flux example 100x slower with L2 regularisation(?!) Performance question	5	723	November 10, 2020
Flux Regularisation in Julia at the layer level New to Julia	1	650	January 8, 2020
Summing matrix elements is >1000X slower than summing vector elements General Usage performance	8	1363	April 17, 2017
Julia equivalence tf.nn.l2_loss(x) in Flux General Usage flux	1	318	June 8, 2022