A package for easy time series forecasting of electrical load
Project description
Loadforecast
A package for an easy time series forecasting of an electrical load based on facebook prophet. It inherits the base of prophet which is additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, with optional holiday effects. It is robust to missing data and outliers as well.
Package is build around child class of original prophet, replacing initial values of attributes by those tuned for electrical load forecasting using grid search method. Package contains functions that are mainly packing up original functions to more user-friendly ones.
We are sorry for inconvenience but currently it is necessary to install prophet
package manually as main dependency.
For more information about the installation procedure check https://github.com/facebook/prophet
Loadforecast is on PyPI, so you can use pip
to install it.
pip install loadforecast
API Demo:
# Initialize model on pandas.DataFrame containing columns 'DateTime' and 'Load'
m = LoadProphet(df)
# Make default one day prediction with sample period of 15 minutes.
forecast = m.prediction()
Changelog
License
Prophet is licensed under the MIT license.
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 loadforecast-0.0.4-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | da917445f99a00ef76ecbe4a842cc3cd8d28b721289a8aa0d8288d8c9ff2d2b1 |
|
MD5 | 373dc0b67e33249d62e7d937b05f7d45 |
|
BLAKE2b-256 | 016abc05d3bd53aa30f2ed85617151aebf7bb3ecb0f047e546877d568e45ed34 |