Python机器学习Sklearn入门之神经网络

Python机器学习Sklearn入门之神经网络

MLP 神经网络算法

MLP 神经网络算法函数位于 neural_network 神经网络模块，函数名是 MLPClassifier，接口是

MLPClassifier(hidden_layer_size = (100,),activation = 'relu',solver = 'adam',alpha = 0.0001,batch_size = 'auto',learning_rate = 'constant',learning_rate_init = 0.001,power_t = 0.5,max_iter = 200,shuffle = True,random_state = None,tol = 0.0001,verbose = False,warm_start = False,momentum = 0.9,nesterovs_momentum = True,early_stopping = False,validation_fraction = 0.1,beta_1 = 0.9,beta_2 = 0.999,epsilon = 1e-08)

建模、预测

#2
print('\n2# 建模')
mx =zai.mx_MLP(x_train.values,y_train.values)

#3
print('\n3# 预测')
y_pred = mx.predict(x_test.values)
df9['y_predsr']=y_pred
df9['y_test'],df9['y_pred']=y_test,y_pred
df9['y_pred']=round(df9['y_predsr']).astype(int) 

mx_MLP() 函数

def mx_MLP(train_x,train_y):
	mx = MLPClassifier()
	mx.fit(train_x,train_y)
	return mx

保存数据结果并显示信息

#4
df9.to_csv('tmp/iris_9.csv',index=False)
print('\n4# df9')
print(df9.tail())

输出

4# df9
     x1   x2   x3   x4  y_predsr  y_test  y_pred
33  6.4  2.8  5.6  2.1         1       1       1
34  5.8  2.8  5.1  2.4         1       1       1
35  5.3  3.7  1.5  0.2         2       2       2
36  5.5  2.3  4.0  1.3         3       3       3
37  5.2  3.4  1.4  0.2         2       2       2

检验测试结果

#5   
dacc=zai.ai_acc_xed(df9,1,False)
print('\n5# mx:mx_sum,kok:{0:.2f}%'.format(dacc))  

输出

5# mx:mx_sum,kok:94.74%

不过，MLP 神经网络算法每次运行的结果是不同的

再运行一次，结果

4# df9
     x1   x2   x3   x4  y_predsr  y_test  y_pred
33  6.4  2.8  5.6  2.1         1       1       1
34  5.8  2.8  5.1  2.4         1       1       1
35  5.3  3.7  1.5  0.2         2       2       2
36  5.5  2.3  4.0  1.3         3       3       3
37  5.2  3.4  1.4  0.2         2       2       2

5# mx:mx_sum,kok:97.37%

MLP_reg 神经网络回归算法

MLP_reg 神经网络算法函数位于 neural_network 神经网络模块，函数名是 MLPRegressor，接口是

MLPRegressor(hidden_layer_size = (100,),activation = 'relu',solver = 'adam',alpha = 0.0001,batch_size = 'auto',learning_rate = 'constant',learning_rate_init = 0.001,power_t = 0.5,max_iter = 200,shuffle = True,random_state = None,tol = 0.0001,verbose = False,warm_start = False,momentum = 0.9,nesterovs_momentum = True,early_stopping = False,validation_fraction = 0.1,beta_1 = 0.9,beta_2 = 0.999,epsilon = 1e-08)

建模、预测

#2
print('\n2# 建模')
mx =zai.mx_MLP_reg(x_train.values,y_train.values)

#3
print('\n3# 预测')
y_pred = mx.predict(x_test.values)
df9['y_predsr']=y_pred
df9['y_test'],df9['y_pred']=y_test,y_pred
df9['y_pred']=round(df9['y_predsr']).astype(int)  

mx_MLP_reg() 函数

def mx_MLP_reg(train_x,train_y):
	mx = MLPRegressor()
	mx.fit(train_x,train_y)
	return mx

保存数据结果并显示信息

#4
df9.to_csv('tmp/iris_9.csv',index=False)
print('\n4# df9')
print(df9.tail())

输出

4# df9
     x1   x2   x3   x4  y_predsr  y_test  y_pred
33  6.4  2.8  5.6  2.1  2.571980       1       3
34  5.8  2.8  5.1  2.4  2.323973       1       2
35  5.3  3.7  1.5  0.2  1.488991       2       1
36  5.5  2.3  4.0  1.3  2.100562       3       2
37  5.2  3.4  1.4  0.2  1.469982       2       1

检验测试结果

#5   
dacc=zai.ai_acc_xed(df9,1,False)
print('\n5# mx:mx_sum,kok:{0:.2f}%'.format(dacc))   

输出

5# mx:mx_sum,kok:44.74%

不过，MLP_reg 神经网络算法每次运行的结果是不同的

再运行一次，结果

4# df9
     x1   x2   x3   x4  y_predsr  y_test  y_pred
33  6.4  2.8  5.6  2.1  2.373282       1       2
34  5.8  2.8  5.1  2.4  2.322128       1       2
35  5.3  3.7  1.5  0.2  1.929890       2       2
36  5.5  2.3  4.0  1.3  1.909459       3       2
37  5.2  3.4  1.4  0.2  1.855990       2       2

5# mx:mx_sum,kok:34.21%

Python机器学习入门源代码和数据集 请点这里