Algorithm 3. Gradient boosting model.

Result: Prediction of delamination size

Input: Sensor features with ground truth ${\left(m_p,n_p\right)}_{p=1}^x$

1 clf = GradientBoostingRegressor (mp, np)

2 clf.fit (loss = 'squared_error', learn_r = 0.1, n_est = 100, subsample = 1.0, c = 'friedman_mse', m_s_s = 2, m_s_l = 1, max_depth = 3, alpha = 0.9, valid_frac = 0.1)

Output: Gradient Boosting Prediction model