support types in @non_differentiable #213
Labels
Comments
oxinabox
added
bug
bors bot
added a commit
to FluxML/Zygote.jl
that referenced
this issue
Sep 3, 2020
780: Move a bunch of no_grad to ChainRules r=oxinabox a=oxinabox this is the partner to JuliaDiff/ChainRules.jl#252 It will fail til that is merged and tagged What is left is: - Types (because JuliaDiff/ChainRulesCore.jl#213) (e.g. `Colon`, `OneTo` `Channel`) - Things to which the derivative is `Zero()` not `DoesNotExist()` (e.g. `one`, `ones`, `zero`, `zeros`) - Things that felt too magic: e.g. `Base.eval` Should I bump patch version and tag a release? Co-authored-by: Lyndon White <[email protected]> Co-authored-by: Lyndon White <[email protected]>
bors bot
added a commit
to FluxML/Zygote.jl
that referenced
this issue
Sep 4, 2020
780: Move a bunch of no_grad to ChainRules r=oxinabox a=oxinabox this is the partner to JuliaDiff/ChainRules.jl#252 It will fail til that is merged and tagged What is left is: - Types (because JuliaDiff/ChainRulesCore.jl#213) (e.g. `Colon`, `OneTo` `Channel`) - Things to which the derivative is `Zero()` not `DoesNotExist()` (e.g. `one`, `ones`, `zero`, `zeros`) - Things that felt too magic: e.g. `Base.eval` Should I bump patch version and tag a release? Co-authored-by: Lyndon White <[email protected]> Co-authored-by: Lyndon White <[email protected]>
nickrobinson251
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Follow up from #212
it doesn't seem to work write with types.
e.g.
@non_differentiable Channel(Any, Any)at least it doesn't seem to work right an play nice with Zygote.
I know one thing that needs to be changes is to use
Core.Typeof($primal_name)rathertypeof($primal_name)in the signatio.because
Core.typeof(Channel)isType{Channel}where astypeof(Channel)isDataType(for others it would be aUnionAllI think)buit I don't think that is the only thing that prevents it working nice with Zygote.
The text was updated successfully, but these errors were encountered: