DynamicalODE Changes #1067
DynamicalODE Changes #1067
Conversation
|
No that's backwards in compat. |
…into DynamicalODE
|
@ChrisRackauckas any changes needed further ?? |
|
|
|
|
I think this is already present here : https://github.com/SciML/SciMLBenchmarks.jl/blob/master/benchmarks/DynamicalODE/single_pendulums.jmd Where can I make an update if this needs a fix ?? |
|
The point of the project is to get that benchmark set working on the latest versions 😅 . That's in the benchmark script of the folder. |
|
Oh, Got it 😅 |
|
But my question was why did we get a undef error if that's already defined there ?? |
|
Previous bad bump maybe? |
|
@ChrisRackauckas How can I see the errors that have occurred ??
Like here |
|
@ParamThakkar123 Check the README of this repo: https://github.com/SciML/SciMLBenchmarks.jl?tab=readme-ov-file#inspecting-benchmark-results, and the built benchmarks can be checked in the corresponding md files |
|
Seems the errors are caused by PyPlot.jl You may consider figuring out what's wrong with PyPlot.jl here or change to Plots.jl(recommend) |
The errors says that the figure function is not defined. Though it's there in PyPlot |
|
You can try to update the compat and the Manifest of the packages in this benchmark to the latest version, many of them are outdated |
|
It would probably be best to just update the plots to use Plots.jl |
Noted. I will make the changes |
| git-tree-sha1 = "9c304562909ab2bab0262639bd4f444d7bc2be37" | ||
| uuid = "d8fb68d0-12a3-5cfd-a85a-d49703b185fd" | ||
| version = "1.4.1+1" | ||
| version = "1.4.1+1" |
There was a problem hiding this comment.
this manifest bump is not correct!
|
The updates to the plots are correct, but the manifest was not bumped so this isn't actually testing the newer versions. Can you bump the manifest? Otherwise we need to revert. |
|
Yes @ChrisRackauckas , I will create a new pr by today which bumps the manifest with the latest versions for both DynamicalODE and BayesianInference. |
Reverts part of #1067 because the manifest never got bumped. Keeping the update to the plots in single pendulums and the formatting change, but removing the package bumps for now since they were not used in the latest manifest and this will let CompatHelper track the real current state
3 participants
Checklist
contributor guidelines, in particular the SciML Style Guide and
COLPRAC.
Additional context
Add any other context about the problem here.