-
It is often convenient to extract a part of a Sample based on its indices.
However, the following script fails.
```Python
import openturns as ot
# Case #1 : a part of a Sample
distribution…
-
The library does not provide Field arithmetic.
Hence, if we subtract two Fields defined on the same domain with the same "output" dimension, the following exception is generated:
```
TypeError: un…
-
Depending on the scale of a covariance model and the domain size, the Karhunen Loeve modes are recovered from a large sample of a Gaussian process or not.
It is illustrated by the following script:…
-
Sub-issue of https://github.com/h2oai/h2o-3/issues/6516
This issue is meant to track papers and frameworks we might want to research. (In random order)
# Papers
- [ ] A Tutorial on Bayesian Opt…
-
Is it possible to easily implement a non-zero mean function with an (improper) vague prior on coefficients of explicit basis functions, cf. Section 2.7 in the GPML book. Specifically, predictions to b…
oteym updated
3 years ago
-
Return `mean_and_cov(::PosteriorGP)` instead of `mean_and_var(::PosteriorGP)` in the `model_posterior(::GaussianProcess)` to enable extracting multivariate predictive distributions.
-
# 🐛 Bug
When using the MultitaskGaussianLikelihood, I've found the model often does not converge to good fits.
The simplest occurence of this phenomena is when fitting the same 1D regression prob…
-
Since moe expects numpy arrays as input and gives numpy arrays as output, it'd be nice if colander supported this directly.
Currently (as a hack), I (eliu) call numpy.array() on the output of params.…
-
Dear Alex,
I personally appreciate your contribution of this work for inversing the ice thickness of surging glaciers.
I have been working on this code for a few days. However, i noticed that ma…
-
Training new neural networks currently involves making decisions given the following parameters [1]:
- training window
- learning rate
- number of trainingsteps
- batch size
- train/test ratio
…