print('*** Fitting with hyperparam:', hyperparam, flush=True)
if num_durations > 0:
surv_model.fit(X_train_std, y_train_discrete,
batch_size, n_epochs, verbose=True)
else:
surv_model.fit(X_train_std, (y_train[:, 0], y_train[:, 1]),
batch_size, n_epochs, verbose=True)
surv_model.save_net(model_filename)
else:
print('*** Loading ***', flush=True)
surv_model.load_net(model_filename)
if num_durations > 0:
surv_df = surv_model.interpolate(10).predict_surv_df(X_test_std)
else:
surv_df = surv_model.predict_surv_df(X_test_std)
surv = surv_df.to_numpy().T
print()
print('[Test data statistics]')
sorted_y_test_times = np.sort(y_test[:, 0])
print('Quartiles:')
print('- Min observed time:', np.min(y_test[:, 0]))
print('- Q1 observed time:',
sorted_y_test_times[int(0.25 * len(sorted_y_test_times))])
print('- Median observed time:', np.median(y_test[:, 0]))
print('- Q3 observed time:',
sorted_y_test_times[int(0.75 * len(sorted_y_test_times))])
print('- Max observed time:', np.max(y_test[:, 0]))
print('Mean observed time:', np.mean(y_test[:, 0]))
print('Fraction censored:', 1. - np.mean(y_test[:, 1]))
print('Time elapsed: %f second(s)' % elapsed)
np.savetxt(time_elapsed_filename,
np.array(elapsed).reshape(1, -1))
surv_model.save_net(model_filename)
else:
print('*** Loading ***', flush=True)
surv_model.load_net(model_filename)
elapsed = float(np.loadtxt(time_elapsed_filename))
print('Time elapsed (from previous fitting): '
+ '%f second(s)' % elapsed)
if num_durations > 0:
surv_df = \
surv_model.interpolate(10).predict_surv_df(fold_X_val_std)
else:
surv_df = surv_model.predict_surv_df(fold_X_val_std)
ev = EvalSurv(surv_df, fold_y_val[:, 0], fold_y_val[:, 1],
censor_surv='km')
sorted_fold_y_val = np.sort(np.unique(fold_y_val[:, 0]))
time_grid = np.linspace(sorted_fold_y_val[0],
sorted_fold_y_val[-1], 100)
surv = surv_df.to_numpy().T
cindex_scores.append(ev.concordance_td('antolini'))
integrated_brier_scores.append(
ev.integrated_brier_score(time_grid))
print(' c-index (td):', cindex_scores[-1])
print(' Integrated Brier score:',
integrated_brier_scores[-1])