Tyler Thomas

> I think it's best to have a function like _relu (it doesn't matter what it's called) so that this won't break if conventions change elsewhere. (No relu, no max....

I think the gradient is working. I also went through the paper again, and we are calculating it exactly how the authors did. I have attached an image of the...

Still trying to get the jacobian to work, but I have the initial forward pass ```julia using NNlib using LinearAlgebra using Zygote sparsemax(x; dims = 1) = sparsemax!(similar(x, (float ∘...

Ah, I assumed that mamba and micromamba would behave the exact same way. Thank you. Edit: This issue has been fixed in cuML v22.10

I have done some additional testing, and all other rapidsai libraries build. I only get errors when cuML is included. cuSpatial, cuDF, cuGraph, cuXFilter, cuSignal, cuCIM all install without issues....

> Could you let us know if you still see these errors with the current 22.10 nightly? I am able to build cuml with the current 22.10 nightly. Thank you...

I think that this is a fantastic change. As you said, the current `mach` returned by `evaluate!` is not useful. It may add some complications that `evaluate` and `evaluate` have...

> CUDA_Runtime_jll - 1.858 GiB Are you running on a machine with CUDA 11? If so, you might be able to eliminate the large CUDA artifacts by switching. The `CUDA_Runtime_jll`...

`EvoTrees.jl` is a pure julia version of gradient boosting. `EvoTrees.jl` has phenomenal cpu performance. This library is just a wrapper of the `xgboost` c library, so it wouldn't be possible...

It might be worth writing methods to convert EvoTrees.jl models to and from XGBoost. There's an issue for it on EvoTrees.jl https://github.com/Evovest/EvoTrees.jl/issues/179