Getting error while modeling with SHO model

[ajaykhoj]

Hi,

I am using this Celerite code for very first time. In my study, I am modeling a data set with SHO model. As documented, I am following all step like, set the prior values for SHO model, when I run the code and simulate data, outcome containing error given as : In first place code is given - """"""""". code """"""""""""""" import celerite from celerite import terms Q3 = 1.0 / np.sqrt(2.0) w3 = 0.008 S3 = np.var(y) / (w3 * Q3) bounds_non_periodic = dict(log_S0=(-10, 10), log_Q=(-10, 10), log_omega0=(-10, 10)) kernel3 = terms.SHOTerm(log_S0=np.log(S3), log_Q=np.log(Q3), log_omega0=np.log(w3)) #kernel3.freeze_parameter("log_Q") kernel_combined = kernel3

gp = celerite.GP(kernel_combined) gp.compute(MJD, dy) # You always need to call compute once. print("Initial log likelihood: {0}".format(gp.log_likelihood(y)))

from scipy.optimize import minimize def neg_log_like(params, y, gp): gp.set_parameter_vector(params) return -gp.log_likelihood(y)

initial_params = gp.get_parameter_vector() bounds = gp.get_parameter_bounds() r = minimize(neg_log_like, initial_params, method="L-BFGS-B", bounds=bounds, args=(y, gp)) gp.set_parameter_vector(r.x) print(r)

x = np.linspace(np.min(MJD), np.max(MJD), 500) pred_mean, pred_var = gp.predict(y, x, return_var=True) pred_std = np.sqrt(pred_var) pred_mean """""""""""""""""""""""""""""""""""""""""""""""""""""

Error : - when I generate the mean and var using gp.predict command, getting outcome as : array([3.07314316e-08, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,

why is that, why am i getting "nan"?

I have gone through many article in which used oscillator models are written as : SHO * n, where n <= 4, how to build a kernel for this ?