model_class.py
--------------

This example shows defining a custom model class by subclassing.


.. code-block:: python

    import numpy as np
    
    from modelrunner import ModelBase
    
    
    class MyModel(ModelBase):
        parameters_default = {"a": 1, "b": 2}
    
        def __call__(self):
            self.storage.write_array("arr", np.arange(4))
            self.storage.write_attrs("arr", {"key": "value"})  # write extra information
            return self.parameters["a"] * self.parameters["b"]
    
    
    # create an instance of the model
    model = MyModel({"a": 3}, output="test.yaml")
    # run the instance
    print(model())