# How compute \`gradient.(f, w)\` on GPU?

**URL:** <https://discourse.julialang.org/t/how-compute-gradient-f-w-on-gpu/30916>\
**Category:** Machine Learning\
**Tags:** first-steps\
**Created:** [November 10, 2019, 12:51pm UTC](https://discourse.julialang.org/t/how-compute-gradient-f-w-on-gpu/30916 "2019-11-10T12:51:25Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![xiaodai](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/xiaodai/32/15937_2.png) [@xiaodai](https://discourse.julialang.org/u/xiaodai)\
**Post date:** [November 16, 2019, 8:06am UTC](https://discourse.julialang.org/t/how-compute-gradient-f-w-on-gpu/30916/3 "2019-11-16T08:06:57Z")

</div>

See [This custom Zygote.jl adjoint is not giving me the speed up I expected and how to migrate to GPU?](https://discourse.julialang.org/t/this-custom-zygote-jl-adjoint-is-not-giving-me-the-speed-up-i-expected-and-how-to-migrate-to-gpu/31150)

Can’t quite get this to work on th GPU unless I differentiate the function myself.

---

_[View the full topic](https://discourse.julialang.org/t/how-compute-gradient-f-w-on-gpu/30916)._