azev77
azev77
@juliangehring & @colintbowers I think it's great for the Julia ecosystem that you guys are working on combining your packages! I'm slowly moving my work away from R/STATA into Julia....
I think I found the culprit ```julia julia> Pkg.add("NaiveBayes") Resolving package versions... Installed GPUCompiler ─ v0.9.2 Installed CUDA ──────── v2.6.0 Updating `~/.julia/environments/v1.6/Project.toml` [f6006082] ↓ EvoTrees v0.7.0 ⇒ v0.5.3 [add582a8] ↓...
But do you know why installing a new pkg downgraded MLJ automatically & can MLJ make this not happen w/o consent?
You're prob correct, I wanted to make sure there is no MLJ setting that is behind this. These kinds of things can create a lot of issues for newcomers esp....
Would it be a big hassle to run it in CI?
That would be great!
Just tried loading all on my work computer. Still have same problem w/ `NaiveBayes`
It works now, but there are some instabilities. The first time I load all packages I get an error: ```julia @inline function load_m(model_list) model_types = [] @inbounds for model in...
Just tried again. Got the same error message the 1st time, not 2nd time. Same as before...
The following compromise seems to work for my blog post. I can add the latest versions of all packages except: Conda/PyCall/DataFrames/MLJBase. I add the immediate previous version for those four...