orbit
orbit copied to clipboard
uber
A Python package for Bayesian forecasting with object-oriented design and probabilistic models under the hood.
Sometimes it requires to forecast values that can be only more or equal zero (for instance, views). What model should I use or what parameters should I add to model...
**Describe the bug** `test_plot.py::test_plot_predicted_data` test hangs on M1 macOS. **To Reproduce** Steps to reproduce the behavior: Run `pytest tests/orbit/diagnostics/test_plot.py::test_plot_predicted_data` on M1 macOS **Expected behavior** Test **Screenshots** If applicable, add screenshots...
## Description A small PR adding some utils working with multiple time-series data. Please delete options that are not relevant. - [ ] Bug fix - [x] New feature -...
I'm using ubuntu 22.04 with python=3.10 and numpy 1.23.0 There was an issue installing pystan(I've managed to install it only from source (2.19.1.1). The only model I can use is...
Looks like somehow the [documentation](https://orbit-ml.readthedocs.io/en/latest/tutorials/ktr1.html) page mistakenly use `seasonality=[7]` instead of `seasonality=[7, 365.25]`. *Solution* - [ ] revert/refresh the notebook
TBATS: Trigonometric seasonality, Box-Cox transformation, Trend and Seasonal components. It would be a gret improvement for ORBIT.
I need to do a rolling prediction, that is, for t = T0, T0+1, ..., T1 I need the model to accept the time series up to t and predict...
Hi! I'm trying to use the [Orbit](https://orbit-ml.readthedocs.io/en/stable/) package that has the torch package as a dependency (view https://github.com/uber/orbit/blob/dev/requirements.txt). However, when trying to import the package, the following error message appears...
**Is your feature request related to a problem? Please describe.** Right now, our tqdm outlook is not optimized within the notebook. **Describe the solution you'd like** See if we can...