-
**Is your feature request related to a problem? Please describe.**
A clear and concise description of what the problem is.
`sktime` currently has a variety of forecasting error metrics (where a lo…
-
Hi,
If I've ran TabularPredictor.fit() and I've got a model that I like with all the HP optimsations and feature selections, how do I then refit that exact set of models and weighted ensemble with …
-
In renewables forecasting, we care about wind speed, solar irradiation at the location of single wind/solar parks. ECMWF IFS with its 0.1deg resolution gives us good forecast performance, but anything…
-
- PyTorch-Forecasting version: 0.9.0
- PyTorch version: 1.9.0
- Python version: 3.8.11
- Operating System: Ubuntu 18.04
### Expected behavior
I executed deepar example code
https://github.c…
-
**Is your feature request related to a problem? Please describe.**
Many forecasters require exogenous data for the forecasting horizon when generating predictions, however some models can make use of…
-
### Feature Summary
Predictive Analytics for Investment Strategies: Create a predictive analytics tool that helps users make better investment decisions based on historical data and trends.
### Desc…
-
Hello! I was trying to train DeepAR and NHiTS, the problem is that the two losses used in those models seem to not be available anymore in the library, where can I find them?
I tried both:
`from …
-
- PyTorch-Forecasting version: 0.10.3
- PyTorch version: 1.13.
- Python version: 3.9.13
- Operating System: macOS 12.6
### Expected behavior
Changing n_layers in NHiTS model to increase the n…
-
大佬，你好，这个代码是不是缺少主程序？这个是复现Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting这篇论文的吗？希望能得到源码，谢谢！
-
Hi Can you please check the MAPE values calculated in the "02-Forecasting_models.ipynb" should be multiplied by 100 or not? Since when using a standard MAPE formula the value, 100 is multiplied. Pleas…