-
Fix error in hidden nodes' simulation (in base.py); basically, all the simulations options were run all the time
-
Use residuals and Kernel Density Estimation (KDE) for predictive simulations
-
plot
method forMTS
(univariate and multivariate time series forecasting with uncertainty quantification, read this) objects -
this notebook contains some examples based on
nnetsauce
v0.13.0 news: https://github.com/Techtonique/nnetsauce/blob/master/nnetsauce/demo/thierrymoudiki_250923_nnetsauce_mts_plots.ipynb -
Multiple examples (not
MTS
) of supervised learning: https://github.com/Techtonique/nnetsauce/blob/master/nnetsauce/demo/thierrymoudiki_180823_all_examples.ipynb