-
For the API design proposal, see [this wiki entry](https://github.com/alan-turing-institute/sktime/wiki/Forecasting-API-proposal).
## Forecasters
### Atomic
- [x] NaiveForecaster (strategies={"l…
-
### Describe the feature you want to add to this project
Hi,
There are already good tutorials for time series forecasting, but they mainly focus on statistical model.
Is it possible to create one…
-
I have around 19,800 values, AutoARIMA, AutoTheta and AutoRegressive work fine, but AutoCES gives this error
)]), 'pipeline_y__forecaster__strategy': 'last'}`…
-
**Is your feature request related to a problem? Please describe.**
I'm always frustrated when trying to train forecasting models because it is slow and not using all my CPU cores.
**Describe the s…
-
### Description
Proposal:
A transfer learning functionality in NeuralForecast core which handles a transfer learning process automatically to perform transfer learning and/or potential change to c…
-
- 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…
-
Wondering if it is possible to add a min and max encoder length instead of a fixed `input_chunk_length` for the TFT model. One of the benefits of the Transformer model is that it allows different inpu…
-
I think there is no reason transformer models are limited to past covariates.
As RNNs does, I think transformer also can treat future covariates and hstoric future covariates.
Is there any reas…