smash.BayesianOptimize

class smash.BayesianOptimize(data=None)

Represents bayesian optimize optional results.

Attributes:

time_steppandas.DatetimeIndex

A list of length n containing the returned time steps.

rr_statesFortranDerivedTypeArray

A list of length n of RR_StatesDT for each time_step.

q_domainnumpy.ndarray

An array of shape (nrow, ncol, n) representing simulated discharges on the domain for each time_step.

internal_fluxesdict[str, numpy.ndarray]

A dictionary where keys are the names of the internal fluxes and the values are array of shape (nrow, ncol, n) representing an internal flux on the domain for each time_step.

control_vectornumpy.ndarray

An array of shape (k,) representing the control vector solution of the optimization (it can be transformed).

costfloat

Cost value.

n_iterint

Number of iterations performed.

projgnumpy.ndarray

Projected gradient value (infinity norm of the Jacobian matrix).

log_lkhfloat

Log likelihood component value.

log_priorfloat

Log prior component value.

log_hfloat

Log h component value.

serr_munumpy.ndarray

An array of shape (ng, ntime_step) representing the mean of structural errors for each gauge and each time_step.

serr_sigmanumpy.ndarray

An array of shape (ng, ntime_step) representing the standard deviation of structural errors for each gauge and each time_step.

See also

smash.bayesian_optimize

Model bayesian assimilation using numerical optimization algorithms.

Notes

The object’s available attributes depend on what is requested by the user during a call to bayesian_optimize in return_options.