bmds 0.1.0

Software development kit for US EPA's Benchmark dose modeling software (BMDS)

This Python 2 package is designed to run the U.S. EPA BMDS software from a python interface. It is integrated into the HAWC software, but HAWC does not be installed on your computer to use this BMDS package. Example function calls are shown below:

import bmds

# get available BMDS versions
bmds.get_versions()

# get BMDS models for the specified version
bmds.get_models_for_version('2.40')

# create example datasets
ds1 = bmds.DichotomousDataset(
    doses=[0, 1.96, 5.69, 29.75],
    ns=[75, 49, 50, 49],
    incidences=[5, 1, 3, 14])

ds2 = bmds.ContinuousDataset(
    doses=[0, 10, 50, 150, 400],
    ns=[111, 142, 143, 93, 42],
    responses=[2.112, 2.095, 1.956, 1.587, 1.254],
    stdevs=[0.235, 0.209, 0.231, 0.263, 0.159])

# create new model runs
models = [
    bmds.Gamma_215(ds1),
    bmds.Power_217(ds2),
]

# execute each model and parse results
for model in models:
    model.execute()

Install a development version:

To install a development version, first checkout from git:

git clone https://github.com/shapiromatron/bmds

Change paths to the newly created bmds folder. Then, preferably in a python virtual environment, run the command:

pip install -r requirements.txt

Check for a successful installation by using the command:

py.test

This package should install fine in any platform (Windows, PC, Linux). However, the BMDS models are only compiled for use in Windows. Therefore, the execute method is only available in Windows.