A High Level Overview¶
Exploratory modeling framework¶
The core package contains the core functionality for setting up, designing, and performing series of computational experiments on one or more models simultaneously.
- Model (
ema_workbench.em_framework.model): an abstract base class for specifying the interface to the model on which you want to perform exploratory modeling.
- Samplers (
ema_workbench.em_framework.samplers): the various sampling techniques that are readily available in the workbench.
- Uncertainties (
ema_workbench.em_framework.parameters): various types of parameter classes that can be used to specify the uncertainties and/or levers on the model
- Outcomes (
ema_workbench.em_framework.outcomes): various types of outcome classes
- Evaluators (
ema_workbench.em_framework.evaluators): various evaluators for running experiments in sequence or in parallel.
The connectors package contains connectors to some existing simulation modeling environments. For each of these, a standard ModelStructureInterface class is provided that users can use as a starting point for specifying the interface to their own model.
- Vensim connector (
vensim): This enables controlling (e.g. setting parameters, simulation setup, run, get output, etc.) a simulation model that is built in Vensim software, and conducting an EMA study based on this model.
- Pysd connector (
- Excel connector (
excel): This enables controlling models build in Excel.
- NetLogo connector (
netlogo): This enables controlling (e.g. setting parameters, simulation setup, run, get output, etc.) a simulation model that is built in NetLogo software, and conducting an EMA study based on this model.
The analysis package contains a variety of analysis and visualization techniques for analyzing the results from the exploratory modeling. The analysis scripts are tailored for use in combination with the workbench, but they can also be used on their own with data generated in some other manner.
- Patient Rule Induction Method (
- Classification Trees (
- Logistic Regression (
- Dimensional Stacking (
- Feature Scoring (
- Regional Sensitivity Analysis (
- various plotting functions for time series data (
- pair wise plots (
- parallel coordinate plots (
- support for converting figures to black and white (