Skip to content

Latest commit

 

History

History
76 lines (57 loc) · 4.01 KB

README.md

File metadata and controls

76 lines (57 loc) · 4.01 KB

DARt

DOI GitHub license Safe GitHub stars

DARt: Estimate Real-time Infection and the Time-varying Epidemiological Parameter Rt

Online platform: https://dsi-dart.shinyapps.io/dsi-covidv4/.

Installation

pip install DARtTool

Quick Start

DARtTool is designed to be used with simple function calls, the core functions of DARtToolare DARt() and car_r(). In the following section we give an overview of the simple use case forDARt()andcal_r().

DARt() and cal_r() are the two single-call functions can be used on its own to infer the underlying infection case curve from reported cases and estimate Rt.

Firstly we need to define a DARt class, and initialize with three parameters: gt (generation time), ds(report of incubation delay) and an inputfile in csv format.

import DARtTool
dart = DARtTool.DARt(filename='./inputfile.csv')

The function cal_r() represents the core functionality of the package aiming to:

  1. infer the underlying infection case curve from observations and estimate Rt.
  2. provides visualisations of results.
dart.cal_r()

Estimating the underlying infection cases and Rt curve via smoothing is substantially computationally demanding than using filter but can provide reliable estimate.

Input file example

Input file should contain date and corresponding observations, the date should be sorted from the oldest to the latest. The first observation number should be nonzero.

See 'Example_input.csv':

date	    newCasesByPublishDate
2021/1/1	53285
2021/1/2	57725
2021/1/3	54990
2021/1/4	58784
2021/1/5	60916
2021/1/6	62322
2021/1/7	52618
2021/1/8	68053
2021/1/9	59937
2021/1/10	54940
...	        ...

Citation

If you use this tool in your research, please cite our paper.

@misc{yang2020revealing,
      title={Revealing the Transmission Dynamics of COVID-19: A Bayesian Framework for $R_t$ Estimation}, 
      author={Xian Yang and Shuo Wang and Yuting Xing and Ling Li and Richard Yi Da Xu and Karl J. Friston and Yike Guo},
      year={2020},
      eprint={2101.01532},
      archivePrefix={arXiv},
      primaryClass={stat.AP}
}

License

This source code is licensed under the MIT license.