Models

Olympus provides easy access to different models that can be trained to emulate an experiment. When loading an Emulator with a standard dataset and model, one of the models provided will be loaded from file. However, if you would like to customize the model used by an Emulator, you can do this with one of the model classes. Similar to Planners, you can either import the desired class directly:

from olympus.models import BayesNeuralNet
model = BayesNeuralNet(...)

Or you can take advantage of the Model function if you want to use the class with default arguments:

from olympus.models import Model
model = Model(kind='BayesNeuralNet')

The models available are the following:

• Bayesian Neural Network
• Neural Network