Sane handling of time series data for forecast modelling - with production usage in mind.
Project description
Sane handling of time series data for forecast modelling - with production usage in mind. While modelling time series data with data science libraries like Pandas, statsmodels, sklearn etc., dealing with time series data is cumbersome - timetomodel takes some of that over. Loading data, making train/test data, feeding data into rolling forecasts… Also, the context and assumptions under which a model was made and used should not be in notebooks, they should have a readable and reproducible spec. Timetomodel is hopefully useful while doing data & model exploration as well as when integrating or replacing models in production environments.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for timetomodel-0.5.4-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | e4b1cb97df814d0880ed5a06c3bb5b03883d40e16933d7dfa62e7d9fca4aef4f |
|
MD5 | d400c198bd00338192664164aac07b0b |
|
BLAKE2b-256 | 8ec26c28621bfffd717dfc20d355f0c7a7a680203be90d0ab4d9866f8929f290 |