Alexander März

Results 89 comments of Alexander März

The thing is that due to the multiple stacking of the data, the approach is infeasible if the original data is already big...

@kevindarby Thanks for your interest in the project. Right now, I mostly focus on getting XGBoostLSS and LightGBMLSS into a new and more efficient PyTorch environment.

@Cattes Hope all is well! May I ask you to remove the xgboostlss from TestPyPI so that I can continue brining it to PyPI. Thanks!

@Cattes Thanks again for your effort, very much appreciated. I am closing this since it is now on pypi.

@valeman Thanks for your comment. Yet I need to disagree with your comment, in particular with > these are all very old methods that do not work Unfortunately, this is...

> @StatMixedML Thanks for your feedbak. I am happy you like our work and find it useful to your research. I believe that what you propose is one of the...

Referring to https://github.com/awslabs/gluon-ts/issues/1013#issuecomment-687575025

@lostella Thanks for your reply and the hint to `pl.seed_everything`. Using the following helps to reduce the variability of the results from run to run, but does not completely make...

As the name GluonTS - Probabilistic Time Series Modeling in Python suggests, gluonts aims at producing probabilistic forecasts, rather than point forecasts. Hence, the default in DeepAR is a distribution...

@jfrank94 There is no explicit argument in the estimator that allows you to do that. Depending on which `distr_output` you choose, e.g., `GaussianOutput()`, you would need to create your own,...