Carlo Lucibello
Carlo Lucibello
With Zygote v0.6.71 on julia v0.11 I get the following error: ```julia m = Chain(BatchNorm(3), sum) x = Float32[1 2; 3 4; 5 6] Zygote.hessian_reverse(m, x) ``` ``` ERROR: Compiling...
A complete rework of our recurrent layers, making them more similar to their pytorch counterpart. This is in line with the proposal in #1365 and should allow to hook into...
As a follow-up to #2492, this removes every residual implementation for data movement between devices in favor of relying entirely on MLDataUtils.jl. The 2 PRs combined save us ~700 lines...
Moving forward with the full deprecation of Flux.params and of the Optimise module, this removes them from tests. After this only explicit differentiation and the Optimisers are used in the...
cc @aurorarossi ```julia julia> using MLDatasets julia> TemporalBrains() ERROR: SystemError: opening file "/Users/carlo/.julia/scratchspaces/124859b0-ceae-595e-8997-d05f6a7a8dfe/datadeps/TemporalBrains/LabelledTBN/labels.txt": No such file or directory Stacktrace: [1] systemerror(p::String, errno::Int32; extrainfo::Nothing) @ Base ./error.jl:176 [2] systemerror @ ./error.jl:175...
- [x] create the package #460 - [ ] make it feature parity with GraphNeuralNetworks.jl (#469, #463, #475, #471) - [x] GNNChain - [x] GCNConv - [x] GraphConv - [x]...
It would be nice to add a small usage example to the readme
This would give us better [vscode integration](https://discourse.julialang.org/t/prerelease-of-new-testing-framework-and-test-run-ui-in-vs-code/86355) and is more like to have long-term support compared to e.g. ReTestItems.jl. As a part of this, should also document in the developer...
The next release is going to be tagged as v1.0.0 I created a milestone with the issues that need to be closed by then https://github.com/CarloLucibello/GraphNeuralNetworks.jl/milestone/1