pred_onx = sess.run([label_name],
{input_name: X_test.astype(numpy.float32)})[0]
print(pred_onx)
###############################################
# With DMatrix
# ++++++++++++
#
# Huge datasets cannot be handled with the scikit-learn API.
# DMatrix must be used. Let's see how to convert the trained
# model.
dtrain = DMatrix(X_train, label=y_train)
param = {'objective': 'multi:softmax', 'num_class': 3}
bst = train_xgb(param, dtrain, 10)
initial_type = [('float_input', FloatTensorType([None, 4]))]
onx = convert_xgboost(bst, initial_types=initial_type)
sess = rt.InferenceSession(onx.SerializeToString())
input_name = sess.get_inputs()[0].name
label_name = sess.get_outputs()[0].name
pred_onx = sess.run([label_name],
{input_name: X_test.astype(numpy.float32)})[0]
print(pred_onx)
##################################
# Display the ONNX graph
# ++++++++++++++++++++++
#
xgb_alltrain = DMatrix(
X_train,
label=y_train,
feature_names=predictors,
feature_types=feature_types)
xgb_alltest = DMatrix(
X_test,
feature_names=predictors,
feature_types=feature_types)
print('\t[XGBoost ] training...')
xgb_model = train_xgb(
xgb_params,
xgb_train,
num_boost_round=15000,
evals=[(xgb_val, 'valid_1')],
verbose_eval=200,
early_stopping_rounds=100,
)
y_pred = xgb_model.predict(X_train[val_idx, :])
score = roc_auc_score(y_train[val_idx], y_pred)
#print('\t[XGBoost ] best iteration: \033[92m%i\033[0m' % xgb_model.best_iteration)
print('\t[XGBoost ] oof ROC-AUC is: \033[92m%.4f\033[0m' % score)
y_train_xgb.append(xgb_model.predict(xgb_alltrain))
y_pred_xgb.append(xgb_model.predict(xgb_alltest))
xgb_scores.append(score)
xgb_feature_importances.append(
xgb_model.get_fscore())