A library for generative modeling and evaluation of synthetic household-level electricity load timeseries. This package is still under active development.
Cents is a library built for generating contextual time series data. Cents supports several generative time series model architectures that can be used to train a time series data generator from scratch on a user-defined dataset. Additionally, Cents provides functionality for loading pre-trained model checkpoints that can be used to generate data instantly.
Cents was used to train the Watts model series.
Feel free to look at our tutorial notebooks to get started.
Cents has been developed and tested on Python 3.9, Python 3.10 and Python 3.11.
We recommend using Poetry for dependency management. Make sure you have poetry installed before following these setup instructions.
Poetry will automatically create a virtual environment and install all dependencies:
poetry installOnce installed, activate the virtual environment:
poetry shellThis gives you a clean, reproducible setup for development.
If you are only interested in using Cents functionality, we recommend using pip in order to install Cents:
pip install cents-mlThis will pull and install the latest stable release from PyPI.
If you want to reproduce the pretrained Watts model series from scratch, you will need to download the PecanStreet DataPort dataset and place it in an appropriate location specified in cents/config/dataset/pecanstreet.yaml. Specifically you will require the following files:
- 15minute_data_austin.csv
- 15minute_data_california.csv
- 15minute_data_newyork.csv
- metadata.csv
New models, new evaluation functionality and new datasets coming soon!