Tim Hargreaves
Tim Hargreaves
There would be considerable benefits to having the Jupyter Docker Stacks publish to [GitHub Package Repository](https://github.com/features/packages) as well as DockerHub. This allows [considerably faster caching](https://dev.to/dtinth/caching-docker-builds-in-github-actions-which-approach-is-the-fastest-a-research-18ei) than can be achieved via...
I have a (vector-to-scalar) function and corresponding derivative function written in Julia that I am unable to translate to Python. I would like to use these within PyTorch by defining...
## Reproducible Example [render_issue_mre.Rmd.txt](https://github.com/rstudio/flexdashboard/files/3636207/render_issue_mre.Rmd.txt) (Remove the `.txt` from the end - only included so GitHub would let me upload it) ## Actual Behaviour I have two UI elements being generated...
`lfact` has been replaced by `logfactorial` since at least version 1.2 of SpecialFunctions.jl, which is the earliest compatible version with Distributions.jl.
The Turing team have been developing [SSMProblems](https://github.com/TuringLang/SSMProblems.jl), a generic interface for defining state space model and their associated filters/smoothers. The goal of this project is to create a flexible and...
I was wondering whether there is/could be any way of detecting that a test item is being ran by selecting an individual test in the VSCode test explorer as opposed...
Small bug that can lead to massive slowdowns. Minimal example: ``` A = CUDA.rand(TA, 2, 2, 100) B = CUDA.rand(TB, 2, 2, 100) CUDA.@profile NNlib.batched_mul(A, B) ``` For TA =...
As shown in the video attached, when I launch Pluto with `Pluto.run()` and create a new notebook. I am unable to create the file by typing in the name and...
I thought I'd move this conversation out into its own issue since there are quite a few ideas to talk about. _Originally posted by @nsiccha in https://github.com/TuringLang/AdvancedHMC.jl/issues/439#issuecomment-3611468706_ > The trace...
When writing CUDA.jl kernels, one should favour Int32 indexing for performance. This currently isn't possible when using the adjoint wrapper since this has hard-coded `Int` types. This PR relaxes these...