# ----------------house data-------------------
m, o, n, p = generate_data()
print(len(n))
print(len(p))
ratio = 0.7
i, j, ii, jj = [], [], [], []
random.shuffle(o)
lens = int(len(o) * ratio)
X_train = [t[1] for t in o[:lens]]
y_train = [t[0] for t in o[:lens]]
X_test = [t[1] for t in o[lens:]]
y_test = [t[0] for t in o[lens:]]
print("=========My Model=============")
model = MulticlassPerceptron(epoch=50, early_stopping=True)
model.fit(X_train, y_train, m)
model.model_analysis(X_test, y_test)
print("=======skit-learn==========")
model = Perceptron(max_iter=50)
model.fit(X_train, y_train)
predicted = model.predict(X_test)
errors = 0
for i in range(len(predicted)):
if predicted[i] != y_test[i]:
errors += 1
print("accuracy:", str(1 - errors * 1.0 / len(predicted)))
# # ---------------------------------------------
#
#
# # -----------------position data------------------
