How to get forecase of next 12 month by Extreme machine learning

Question

I have a time series data from 2017 to 2022. How can I get the prediction for next 12 months in 2023. Prediction(X_test) can just give the result of last 5 monhs in 2022.

m = 14
per = (1736 - m)/1826
size = int(len(data) * per)
d_train, d_test = data[0:size], data[size:len(data)]
mean_train = np.mean(d_train)
sd_train = np.std(d_train)
d_train = (d_train-mean_train)/sd_train
d_test = (d_test-mean_train)/sd_train
X_train = np.array([d_train[i][0] for i in range(m)])
y_train = np.array(d_train[m][0])
for i in range(1,(d_train.shape[0]-m)):
    l = np.array([d_train[j][0] for j in range(i,i+m)])
    X_train = np.vstack([X_train,l])
    y_train = np.vstack([y_train,d_train[i+m]])

X_test = np.array([d_test[i][0] for i in range(m)])
y_test = np.array(d_test[m][0])
for i in range(1,(d_test.shape[0]-m)):
    l = np.array([d_test[j][0] for j in range(i,i+m)])
    X_test = np.vstack([X_test,l])
    y_test = np.vstack([y_test,d_test[i+m]])
......
def predict(X):
    out = hidden_nodes(X)
    out = np.dot(out, output_weights)
    return out
prediction = predict(X_test)

give forecast of next 12 months (2023)