Metadata-Version: 2.1
Name: predictably
Version: 0.1.2
Summary: Scikit-learn compatible Python forecasting module
Home-page: https://rnkuhns.github.io/predict-ably/
Author: Ryan Kuhns
Author-email: rnkuhns@gmail.com
License: BSD-3
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: Development Status :: 1 - Planning
Description-Content-Type: text/markdown
Requires-Dist: numpy (>=1.13.3)
Requires-Dist: scipy (>=0.19.1)
Requires-Dist: statsmodels (>=0.10)
Requires-Dist: scikit-learn (>=0.22.1)
Provides-Extra: dev
Requires-Dist: pip-tools ; extra == 'dev'
Requires-Dist: rpy2 ; extra == 'dev'

# predict-*ably*
A Python module for time-series forecasting that seeks to provide high-level functionality similar to R's fable and forecast packages

Where functionality already exists in Python the goal is to provide a consistent API wrapper that fits with scikit-learn and standard forecasting practice. 

To that extent, the project will leverage some of the code used in Alan-Turing Institute's sktime (https://github.com/alan-turing-institute/sktime). 

Plan to add hierarchical reconciliation along the lines of: 
1. https://github.com/carlomazzaferro/scikit-hts-examples (see: https://scikit-hts.readthedocs.io/en/latest/_modules/hts/model/ar.html#AutoArimaModel for docs)
2. https://github.com/CollinRooney12/htsprophet


