# oop interface for optimising and using results
# objective is not put through log here, if it is, might want to bound away from zero
bo = BayesOptSearch(est,
X,
param_bounds,
objective_log=False,
objective_log_bound=100,
num_initial_evaluations=5)
# optimisation starts here, use results and apply to model
best_params = bo.optimize(num_iter=5)
est.set_parameters_from_dict(best_params)
est.fit(X)
visualise_fit_2d(est, X)
plt.suptitle("Original fit %s\nOptimised over: %s" %
(str(est.get_parameters()), str(param_bounds)))
if len(param_bounds) == 1:
plt.figure()
plot_bayesopt_model_1d(bo)
plt.title("Objective")
# now change data, with different length scale
X = np.random.randn(200, D) * .1
# reset optimiser, which but initialise from old model, sample 3 random point to update
best_params = bo.re_initialise(new_data=X, num_initial_evaluations=3)
# this optimisation now runs on the "new" objective
best_params = bo.optimize(num_iter=3)
for i, sigma in enumerate(log_sigmas):
est = KernelExpLiteGaussian(np.exp(sigma), lmbda, D, N)
# this is an array num_repetitions x num_folds, each containing a objective
xval_result = est.xvalidate_objective(X, num_folds=5, num_repetitions=2)
O[i] = np.mean(xval_result)
O_lower[i] = np.percentile(xval_result, 10)
O_upper[i] = np.percentile(xval_result, 90)
# best parameter
best_log_sigma = log_sigmas[np.argmin(O)]
# visualisation
plt.figure()
plt.plot([best_log_sigma, best_log_sigma], [np.min(O), np.max(O)], 'r')
plt.plot(log_sigmas, O, 'b-')
plt.plot(log_sigmas, O_lower, 'b--')
plt.plot(log_sigmas, O_upper, 'b--')
plt.xlim([np.min(log_sigmas) - 1, np.max(log_sigmas) + 1])
plt.xlabel("log sigma")
plt.ylabel("objective")
plt.title("lmbda=%.4f" % lmbda)
plt.legend(["Best sigma", "Performance"])
plt.legend(["Best sigma", "Performance", "80% percentile"])
plt.tight_layout()
est.sigma = np.exp(best_log_sigma)
est.fit(X)
visualise_fit_2d(est, X)
plt.show()