Skip to content
/ orbit Public

A Python package for Bayesian forecasting with object-oriented design and probabilistic models under the hood.

License

Notifications You must be signed in to change notification settings

uber/orbit

Repository files navigation

Join Slack   |   Documentation   |   Blog - Intro   |   Blog - v1.1

Orbit banner


GitHub release (latest SemVer) PyPI Build and Test Documentation Status PyPI - Python Version Downloads Conda Recipe Conda - Platform Conda (channel only) PyPI - License

User Notice

The default page of the repo is on dev branch. To install the dev version, please check the section Installing from Dev Branch. If you are looking for a stable version, please refer to the master branch here.

Disclaimer

This project

  • is stable and being incubated for long-term support. It may contain new experimental code, for which APIs are subject to change.
  • requires cmdstanpy as one of the core dependencies for Bayesian sampling.

Orbit: A Python Package for Bayesian Forecasting

Orbit is a Python package for Bayesian time series forecasting and inference. It provides a familiar and intuitive initialize-fit-predict interface for time series tasks, while utilizing probabilistic programming languages under the hood.

For details, check out our documentation and tutorials:

Currently, it supports concrete implementations for the following models:

  • Exponential Smoothing (ETS)
  • Local Global Trend (LGT)
  • Damped Local Trend (DLT)
  • Kernel Time-based Regression (KTR)

It also supports the following sampling/optimization methods for model estimation/inferences:

  • Markov-Chain Monte Carlo (MCMC) as a full sampling method
  • Maximum a Posteriori (MAP) as a point estimate method
  • Variational Inference (VI) as a hybrid-sampling method on approximate distribution

Installation

Installing Stable Release

Install the library either from PyPi or from the source with pip. Alternatively, you can also install it from Anaconda with conda:

With pip

  1. Installing from PyPI

    $ pip install orbit-ml
  2. Install from source

    $ git clone https://github.com/uber/orbit.git
    $ cd orbit
    $ pip install -r requirements.txt
    $ pip install .

With conda

The library can be installed from the conda-forge channel using conda.

$ conda install -c conda-forge orbit-ml

Installing from Dev Branch

$ pip install git+https://github.com/uber/orbit.git@dev

Quick Start with Damped-Local-Trend (DLT) Model

FULL Bayesian Prediction

from orbit.utils.dataset import load_iclaims
from orbit.models import DLT
from orbit.diagnostics.plot import plot_predicted_data

# log-transformed data
df = load_iclaims()
# train-test split
test_size = 52
train_df = df[:-test_size]
test_df = df[-test_size:]

dlt = DLT(
  response_col='claims', date_col='week',
  regressor_col=['trend.unemploy', 'trend.filling', 'trend.job'],
  seasonality=52,
)
dlt.fit(df=train_df)

# outcomes data frame
predicted_df = dlt.predict(df=test_df)

plot_predicted_data(
  training_actual_df=train_df, predicted_df=predicted_df,
  date_col=dlt.date_col, actual_col=dlt.response_col,
  test_actual_df=test_df
)

full-pred

Demo

Nowcasting with Regression in DLT:

Open All Collab

Backtest on M3 Data:

Open All Collab

More examples can be found under tutorials and examples.

Contributing

We welcome community contributors to the project. Before you start, please read our code of conduct and check out contributing guidelines first.

Versioning

We document versions and changes in our changelog.

References

Presentations

Check out the ongoing deck for scope and roadmap of the project. An older deck used in the meet-up during July 2021 can also be found here.

Citation

To cite Orbit in publications, refer to the following whitepaper:

Orbit: Probabilistic Forecast with Exponential Smoothing

Bibtex:

@misc{
    ng2020orbit,
    title={Orbit: Probabilistic Forecast with Exponential Smoothing},
    author={Edwin Ng,
        Zhishi Wang,
        Huigang Chen,
        Steve Yang,
        Slawek Smyl},
    year={2020}, eprint={2004.08492}, archivePrefix={arXiv}, primaryClass={stat.CO}
}

Papers

  • Bingham, E., Chen, J. P., Jankowiak, M., Obermeyer, F., Pradhan, N., Karaletsos, T., Singh, R., Szerlip, P., Horsfall, P., and Goodman, N. D. Pyro: Deep universal probabilistic programming. The Journal of Machine Learning Research, 20(1):973–978, 2019.
  • Hoffman, M.D. and Gelman, A. The No-U-Turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo. J. Mach. Learn. Res., 15(1), pp.1593-1623, 2014.
  • Hyndman, R., Koehler, A. B., Ord, J. K., and Snyder, R. D. Forecasting with exponential smoothing: the state space approach. Springer Science & Business Media, 2008.
  • Smyl, S. Zhang, Q. Fitting and Extending Exponential Smoothing Models with Stan. International Symposium on Forecasting, 2015.

Related projects