Uncertainty Methods¶
The uncertainty methods translate input parameter uncertainty information provided in the scenario config file into numerical values for use in other CELAVI methods.
-
uncertainty_methods.apply_array_uncertainty(quantity, run)¶ Use model run number to access one element in a parameter list.
When quantity is a List, then the value of quantity with the index run is returned. When quantity is a Dict, the value assigned to the key “value” is returned. When quantity is a float, it is returned as-is.
- Parameters
quantity (List, Dict, or float) – A data structure containing a range of parameter values (floats), distribution parameters, or a single float
run (int) – The current CELAVI run number
- Returns
- Return type
A single float representing the value of quantity during run
-
uncertainty_methods.apply_stoch_uncertainty(quantity, seed=1, distn=<scipy.stats._continuous_distns.triang_gen object>)¶ Draw from distribution if parameters exist.
When quantity is a Dict, it must contain distribution parameters as well as a key named “value”. Once a parameter value has been drawn from the distribution, it is stored under “value”. When quantity is a float, it is returned as-is.
distn defaults to a triangular distribution. If another distribution is used, this method will need to be edited to use the relevant parameters.
- Parameters
quantity (Dict or float) – Contains distribution parameters and a key:value pair for storing the random draw.
seed (int or an instance of np.random.default_rng) – Defines the current random state. Must be passed in from Scenario for reproducibility.
distn (Distribution available in scipy.stats) – Defaults to the triangular distribution.