SunPower / packages / Carousel
pypi conda

Model Simulation Framework


pip install

To install this package with pip:
pip install -i carousel

conda install

  • source  v0.3.2
  • win-64  v0.3.2
To install this package with conda run:
conda install -c sunpower carousel


.. image:: :target:

Carousel - Model Simulation Framework

Carousel ia a framework for simulating mathematical models that decouples the models from the simulation implementation. It takes care of boilerplate routines such as loading data from various sources into a key store that can be used from any calculation, determining the correct order of calculations, stepping through dynamic simulations and generating output reports and visualizations, so that you can focus on developing models and don't have to worry about how to add new models or how to integrate changes.


  • Pint <>_
  • NumPy <>_
  • h5py <>_
  • xlrd <>_
  • UncertaintyWrapper <>_


Carousel releases are on PyPI <>_ and on GitHub <>_. You can use either pip, conda, or distutils to install Carousel.

pip <>_ ::

$ pip install Carousel

Extract the archive to use disutils <>_ ::

$ python install

SunPower conda channel <>_ ::

$ conda install -c sunpower Carousel


Carousel documentation <>_ is online. It's also included in the distribution and can be built using Sphinx <>_ by running the Makefile found in the docs folder of the Carousel package. Once built documentation will be found in the _build folder under the tree corresponding to the type of documentation built. EG: HTML documentation is in docs/_build/html.


Carousel source code <>_ is online. Fork it and report issues <>, make suggestions or create pull requests. Discuss the roadmap or download presentations on the wiki <>


The change log for all releases <>_ is on GitHub.

Quickstart Example

Define data, outputs, formulas, calculations, simulations and model::

#! python

from carousel.core.data_sources import DataSource, DataParameter
from carousel.core.outputs import Output, OutputParameter
from carousel.core.formulas import Formula, FormulaParameter
from carousel.core.calculations import Calc, CalcParameter
from carousel.core.simulations import Simulation, SimParameter
from carousel.core.models import Model, ModelParameter
from carousel.contrib.readers import ArgumentReader
from carousel.core import UREG
import numpy as np
import os

DATA = {'PythagoreanData': {'adjacent_side': 3.0, 'opposite_side': 4.0}}

class PythagoreanData(DataSource):
    adjacent_side = DataParameter(units='cm', uncertainty=1.0)
    opposite_side = DataParameter(units='cm', uncertainty=1.0)

    def __prepare_data__(self):
        for k, v in self.parameters.iteritems():
            self.uncertainty[k] = {k: v['uncertainty'] * UREG.percent}

    class Meta:
        data_cache_enabled = False
        data_reader = ArgumentReader

class PythagoreanOutput(Output):
    hypotenuse = OutputParameter(units='cm')

def f_pythagorean(a, b):
    a, b = np.atleast_1d(a), np.atleast_1d(b)
    return np.sqrt(a * a + b * b).reshape(1, -1)

class PythagoreanFormula(Formula):
    f_pythagorean = FormulaParameter(
        units=[('=A', ), ('=A', '=A')],

    class Meta:
        module = __name__

class PythagoreanCalc(Calc):
    pythagorean_thm = CalcParameter(
        args={'data': {'a': 'adjacent_side', 'b': 'opposite_side'}},

class PythagoreanSim(Simulation):
    settings = SimParameter(
        ID='Pythagorean Theorem',
        commands=['start', 'load', 'run'],
        sim_length=[0, 'hour'],
            'data': ['adjacent_side', 'opposite_side'],
            'outputs': ['hypotenuse']

class PythagoreanModel(Model):
    data = ModelParameter(sources=[PythagoreanData])
    outputs = ModelParameter(sources=[PythagoreanOutput])
    formulas = ModelParameter(sources=[PythagoreanFormula])
    calculations = ModelParameter(sources=[PythagoreanCalc])
    simulations = ModelParameter(sources=[PythagoreanSim])

    class Meta:
        modelpath = os.path.dirname(__file__)

if __name__ == '__main__':
    m = PythagoreanModel()
    m.command('run', data=DATA)
    out_reg = m.registries['outputs']
    fmt = {
        'output': out_reg['hypotenuse'],
        'uncertainty': out_reg.uncertainty['hypotenuse']['hypotenuse']
    print 'hypotenuse = %(output)s +/- %(uncertainty)s' % fmt

This is the MCVE <>_ of a Carousel model.

PRIVACY POLICY  |  EULA (Anaconda Cloud v2.33.29) © 2019 Anaconda, Inc. All Rights Reserved.