epysurv is a Pythonic wrapper around the R surveillance package
that strives to implement a scikit-learn like API for epidemiological surveillance in Python.
from epysurv import data as epidata
from epysurv.models.timepoint import FarringtonFlexible
train, test = epidata.salmonella()
train.head()
# n_cases n_outbreak_cases outbreak
# 2004-01-05 0 0 False
# 2004-01-12 0 0 False
# 2004-01-19 2 0 False
# 2004-01-26 2 0 False
# 2004-02-02 1 0 False
model = FarringtonFlexible()
model.fit(train)
model.predict(test)
# n_cases n_outbreak_cases outbreak alarm
# 2011-01-03 1 0 False 0.0
# 2011-01-10 0 0 False 0.0
# 2011-01-17 3 0 False 0.0
# 2011-01-24 3 0 False 0.0
# 2011-01-31 3 0 False 0.0As epysurv requires both Python and R it can only be conveniently installed through conda:
conda install -c conda-forge epysurv
You can read the documentation or try an interactive demo on binder.
To set up a local development environment, run
conda env create -f env.yml
conda activate epysurv-dev
pip install -r requirements-dev.txt
pre-commit install
To run all tests, simply run
pytest
in the project's root directory.
If you want to contribute to the documentation, run
pip install -r requirements-doc.txt
to install the necessary packages for building the sphinx documentation.
Pull requests are highly welcome!