jzicker
jzicker
the section on multi step prediction is very, very slow on mac os. Any tips or tricks to speed it up? ``` n_forecast
attempting example of graphjs and get this warning > g print(g) Warning message: In structure(x, class = unique(c("AsIs", oldClass(x)))) : Calling 'structure(NULL, *)' is deprecated, as NULL cannot have attributes....
Direct method for multistep forecasts is building a model for each time step into the future and then combining them. Modeltime has the potential to create multiple models for each...
Provide the functionality to incorporate xregs into recursive forecasting. One could envision this as a multi output one step forecast for all time series and then recursive multistep forecasting using...
The rectify package in R is an example of this. Use a recursive forecast first and then model the residuals using direct forecasting.
I'm not sure if this will be a question or request for additional functionality. In general support for machine learning multistep forecasts would be extremely beneficial. Currently the recursive functionality...
I need some help explaining this line of code in the example for the stemgnn model fcst.fit(df=Y_train_df, static_df=AirPassengersStatic, val_size=12) I don't understand what the purpose is of the static_df parameter
couldn't run colab tutorial for long horizon nhits This cell %%capture fcst = NeuralForecast( models=models, freq='15min') fcst_df = fcst.cross_validation(df=Y_df, val_size=val_size, test_size=test_size, n_windows=None) --------------------------------------------------------------------------- RuntimeError Traceback (most recent call last) [](https://localhost:8080/#)...
Anyone have an example/tutorial of a contextual bandit using Thompson sampling as mentioned in the comments in source files?
- PyTorch-Forecasting version: 1.0 - PyTorch version: - Python version: - Operating System: ### Expected behavior I added lags to my timeseriesdataset and added the variables to time_varying_unknown training =...