print("\t\t\t OPTION 2: using CART to forecast REGRESSION.")
print("\t\t\t OPTION x: exit.")
c = input("\n\t\t\t Please input your option:")
if c == '1':
print("\t\t STEP1: Load dataset...")
xx, yy = ds.load(True, r'.\dataset.dat')
xx = ds.discrete_vec(xx, K=DS.DATAUtil.g_SuperParam['K'])
y_c = yy[:, 1]
xx_t, y_t = ds.load(True, r'.\testds.dat')
x_t = ds.discrete_vec(xx_t, K=DS.DATAUtil.g_SuperParam['K'])
y_t_c = y_t[:, 1]
print("\t\t STEP2: Build CART Tree...")
model = DecisionTree.CART(c_r=True)
model.train(xx, y_c)
print("\t\t STEP3: Display CART Tree...")
PlotUtil.init_Plot("CART Decision Tree - (C)")
model.display_Tree()
PlotUtil.show_Plot()
print("\t\t STEP4: Predict by CART Tree...")
y_p = model.predict(x_t)
y_p = y_p.flatten()
if y_p.shape[0] > 0:
eval_y = np.zeros_like(y_p, dtype=int)
eval_y = np.where((y_p == y_t_c), 1, 0)
y_p = ds.y_int2str(y_p)
y_p = np.reshape(y_p, (-1, 1))
print("The predict result is:")
