MLP Regressor

import numpy as np
import pandas as pd
# fix random seed for reproducibility
np.random.seed(7)
# load pima indians dataset
concrete = pd.read_csv('./concrete.csv', encoding='utf-8')

from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler() # 此處非標準化！而是[min, max]
scaler.fit(concrete)
concrete_norm = scaler.transform(concrete) # 真正做轉換
pd.DataFrame(concrete_norm).describe()


X = concrete_norm[:,0:8]
y = concrete_norm[:,8]

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25,random_state=14)

from sklearn.neural_network import MLPRegressor

"""
# 一個神經元
model_mlp = MLPRegressor(random_state=0, activation='relu', hidden_layer_sizes=1)
model_mlp.fit(X_train, y_train)
mlp_score=model_mlp.score(X_train,y_train)

print('score:',mlp_score)
result = model_mlp.predict(X_test)

import matplotlib.pyplot as plt
plt.scatter(y_test, result)
plt.plot([0,1],[0,1])

correlation1 = pd.DataFrame({'origin':y_test,'predict':result})
correlation1.corr()

""" 

# 五個神經元
model_mlp = MLPRegressor(random_state=0, activation='relu', hidden_layer_sizes=10)
model_mlp.fit(X_train, y_train)
train_score=model_mlp.score(X_train,y_train)
print('training score:',train_score)

# 方法一: 估計 X_test 的performance
test_score=model_mlp.score(X_test,y_test)
print('test score:',test_score)


# 方法二: 用 r2_score 估計 performance
from sklearn.metrics import r2_score
result = model_mlp.predict(X_test)
test2_score=r2_score(y_test,result)
print('test2 score:',test2_score)

plt.scatter(y_test, result)
plt.plot([0,1],[0,1])

correlation1 = pd.DataFrame({'origin':y_test,'predict':result})
correlation1.corr()