Kevin M Jablonka

Results 390 comments of Kevin M Jablonka

a bit more far-fetched would be experiment tracking. I'm thinking of stuff like weights and biases which 1) has tools for hyperparameter sweeps 2) tools to visualize some chemistry

as framework, I really enjoy using Optuna

potentially one could even do something like a cookiecutter with something like https://app.crystaltoolkit.org/ or Python backend React frontend to bootstrap course projects

- Conformal predictions (e.g., https://arxiv.org/abs/1908.03569) are also quite interesting - Gives a nice bridge to the infinite width limit of NN and stuff like NNGP and NTK - The [cheminformaticians...

https://arxiv.org/pdf/2110.04629.pdf and the library

I'd like to look into this as some parameters that people need are based on the fit of an ellipse. I think the least-squares python implementation Luc linked looks reasonable.

depends on https://github.com/conda-forge/staged-recipes/pull/19798 and making a PyPi release for `moleculetda`

@a1k12 I added you as maintainer, hope it is fine! Of course, also fine for me to remove myself from this list!

currently blocked by https://github.com/sciris/sciris/issues/122

I'm also willing to be a maintainer.