Surprise at the end

1 2.6 Neural Network Regression Algorithm

12.6.1 Classes, Parameters, Properties, and Methods

kind

class sklearn.neural_network.MLPRegressor(hidden_layer_sizes=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, n_iter_no_change=10, max_fun=15000)

parameter

parameter

explain

hidden_layer_sizes

tuple, length = n_layers - 2, default=(100,) The ith element represents the number of neurons in the ith hidden layer.

activation

{'identity', 'logistic', 'tanh', 'relu'}, default='relu' Activation function for the hidden layer.

'identity' , no-op activation, used to implement a linear bottleneck, returns f(x) = x

'logistic' , the logistic sigmoid function, returns f(x) = 1 / (1 + exp(-x)) .

'tanh' , the hyperbolic tan function, returns f(x) = tanh(x) .

'relu' is the corrected linear unit function, returning f(x) = max(0 , x)

solver

{'lbfgs', 'sgd', 'adam'}, default='adam' weight optimization solver.

'lbfgs' is an optimizer in the family of quasi-Newton methods.

'sgd' refers to Stochastic Gradient Descent.

'adam' refers to the gradient-based stochastic optimizer proposed by Jinma, Dideric, and Jimmyba Note : In terms of training time and validation scores, the default solver 'adam' works well on relatively large datasets ( with thousands or more training samples ) works reasonably well. However, for small datasets, 'lbfgs' can converge faster and perform better.

alpha

float, default=0.0001 . L2 penalty ( regular term ) parameter.

Attributes

Attributes

category

introduce

loss_

float

Current loss calculated with the loss function.

best_loss_

float

The minimum loss achieved by the solver during the entire fitting process.

loss_curve_

list of shape (n_iter_,)

The ith element in the list represents the loss for the ith iteration.

t_

int

The number of training samples seen by the solver during fitting.

coefs_

list of shape (n_layers - 1,)

The ith element in the list represents the weight matrix corresponding to the ith layer.

intercepts_

list of shape (n_layers - 1,)

The ith element in the list represents the bias vector corresponding to layer i+1 .

n_iter_

int

The number of iterations the solver has run.

n_layers_

int

layers.

n_outputs_

int

number of outputs.

out_activation_

str

The name of the output activation function.

loss_curve_

list of shape (n_iters,)

The loss value is evaluated at the end of each training step.

t_

int

Mathematically equal to n iters*X.shape[0] , representing the time step, used by the optimizer's learning rate scheduler.

method

fit(X, y)

Fit the model to the data matrix X and target y .

get_params([deep])

Get the parameters for this estimator.

predict(X)

A multi-layer perceptron model is used for prediction.

score(X, y[, sample_weight])

Returns the predicted coefficient of determination R2 .

set_params(**params)

Set the parameters for this estimator.

12.6.2 Neural Network Regression Algorithms

def MLPRegressor_make_regression():       warnings.filterwarnings("ignore")       myutil = util()       X,y = datasets.make_regression(n_samples=100,n_features=1,n_informative=2,noise=50,random_state=8)       X_train,X_test,y_train,y_test = train_test_split(X, y, random_state=8,test_size=0.3)       clf = MLPRegressor(max_iter=20000).fit(X,y)       title = "MLPRegressor make_regression数据集(有噪音)"       myutil.draw_line(X[:,0],y,clf,title)

       picture

def My_MLPRegressor(solver,hidden_layer_sizes,activation,level,alpha,mydata,title):    warnings.filterwarnings("ignore")    myutil = util()    X,y = mydata.data,mydata.target    X_train,X_test,y_train,y_test = train_test_split(X, y, random_state=8,test_size=0.3)   clf = MLPRegressor(solver=solver,hidden_layer_sizes=hidden_layer_sizes,activation=activation,alpha=alpha,max_iter=10000).fit(X_train,y_train)   mytitle = "MLPRegressor("+title+"):solver:"+solver+",node:"+str(hidden_layer_sizes)+",activation:"+activation+",level="+str(level)+",alpha="+str(alpha)   myutil.print_scores(clf,X_train,y_train,X_test,y_test,mytitle)def MLPRegressor_base():    mydatas = [datasets.load_diabetes(), datasets.load_boston()]    titles = ["糖尿病数据","波士顿房价数据"]    for (mydata,title) in zip(mydatas, titles):        ten = [10]        hundred = [100]        two_ten = [10,10]        Parameters = [['lbfgs',hundred,'relu',1,0.0001], ['lbfgs',ten,'relu',1,0.0001], ['lbfgs',two_ten,'relu',2,0.0001],['lbfgs',two_ten,'tanh',2,0.0001],['lbfgs',two_ten,'tanh',2,1]]       for Parameter in Parameters:              My_MLPRegressor(Parameter[0],Parameter[1],Parameter[2],Parameter[3],Parameter[4],mydata,title)

output

MLPRegressor(糖尿病数据):solver:lbfgs,node:[100],activation:relu,level=1,alpha=0.0001:68.83%MLPRegressor(糖尿病数据):solver:lbfgs,node:[100],activation:relu,level=1,alpha=0.0001:28.78%MLPRegressor(糖尿病数据):solver:lbfgs,node:[10],activation:relu,level=1,alpha=0.0001:53.50%MLPRegressor(糖尿病数据):solver:lbfgs,node:[10],activation:relu,level=1,alpha=0.0001:45.41%MLPRegressor(糖尿病数据):solver:lbfgs,node:[10, 10],activation:relu,level=2,alpha=0.0001:68.39%MLPRegressor(糖尿病数据):solver:lbfgs,node:[10, 10],activation:relu,level=2,alpha=0.0001:31.62%MLPRegressor(糖尿病数据):solver:lbfgs,node:[10, 10],activation:tanh,level=2,alpha=0.0001:64.18%MLPRegressor(糖尿病数据):solver:lbfgs,node:[10, 10],activation:tanh,level=2,alpha=0.0001:31.46%MLPRegressor(糖尿病数据):solver:lbfgs,node:[10, 10],activation:tanh,level=2,alpha=1:-0.00%MLPRegressor(糖尿病数据):solver:lbfgs,node:[10, 10],activation:tanh,level=2,alpha=1:-0.01%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[100],activation:relu,level=1,alpha=0.0001:90.04%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[100],activation:relu,level=1,alpha=0.0001:63.90%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[10],activation:relu,level=1,alpha=0.0001:85.23%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[10],activation:relu,level=1,alpha=0.0001:68.49%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[10, 10],activation:relu,level=2,alpha=0.0001:90.12%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[10, 10],activation:relu,level=2,alpha=0.0001:63.48%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[10, 10],activation:tanh,level=2,alpha=0.0001:18.19%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[10, 10],activation:tanh,level=2,alpha=0.0001:18.25%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[10, 10],activation:tanh,level=2,alpha=1:85.37%MLPRegressor(波士顿房价数据):solver:lbfgs,node:[10, 10],activation:tanh,level=2,alpha=1:

73.75%

data

solver

node

activation

level

alpha

training score

test score

diabetes

lbfgs

[100]

relu

1

0.0001

68.83%

28.78%

lbfgs

[10]

relu

1

0.0001

53.50%

45.41%

lbfgs

[10,10]

relu

2

0.0001

68.39%

31.62%

lbfgs

[10,10]

tanh

2

0.0001

64.18%

31.46%

lbfgs

[10,10]

tanh

2

1

-0.00%

-0.00%

Boston house prices

lbfgs

[100]

relu

1

0.0001

90.04%

63.90%

lbfgs

[10]

relu

1

0.0001

85.23%

68.49%

lbfgs

[10,10]

relu

2

0.0001

90.12%

63.48%

lbfgs

[10,10]

tanh

2

0.0001

18.19%

18.25%

lbfgs

[10,10]

tanh

2

1

85.37%

73.75%

Surprise at the end of the article

Penetration testing environment and code

Experimental code:

Link: https://pan.baidu.com/s/14XsCng6laiSiT_anuwr5dw?pwd=78dy

Extraction code: 78dy

surroundings

Install tomcat, Apache and MySQL on Windows

Install tomcat, Apache and MySQL on Linux

operate

1. Copy the sec in tomcat to the tomcat directory, such as %TOMCAT-HOME%\webapps\

2. Copy the sec in Apache to the Apache directory, such as \htdocs\

3. Under the sec directory in tomcat

include.jsp

<%StringWindows_IP="127.0.0.1";StringLinux_IP="192.168.0.150";StringJSP_PORT="8080";StringPHP_PORT="8100";%>

  • String Windows_IP: IP address of Windows

  • String Linux_IP: IP address of Linux

  • String JSP_PORT: The port number of the JSP

  • String PHP_PORT: The port number of PHP

3. Include.php in the sec directory in Apache

$windows_ip="http://127.0.0.1";$linux_ip="http://192.168.0.150";$jsp_port="8080";$php_port="8100";?>

  • $windows_ip: IP address of Windows

  • $linux_ip=: IP address of Linux

  • $jsp_port=: The port number of the JSP

  • $php_port: The port number of PHP

Open browse and enter http://192.168.0.106:8080/sec/

192.168.0.106 is the local IP address

Database configuration

Create a sec database under MySQL, root/123456. Import 4 csv files under DB into sec database

Penetration testing operating system virtual machine file vmx file

1) Windows 2000 Professional

Link: https://pan.baidu.com/s/13OSz_7H1mIpMKJMq92nEqg?pwd=upsm

Extraction code: upsm

2) Windows Server 2003 Standard x64 Edition

Link: https://pan.baidu.com/s/1Ro-BoTmp-1kq0W_lB9Oiww?pwd=ngsb

Extraction code: ngsb

Power-on password: 123456

3) Windows 7 x64

Link: https://pan.baidu.com/s/1-vLtP58-GXmkau0OLNoGcg?pwd=zp3o

Extraction code: zp3o

4) Debian 6 (Kali Linux)

Link: https://pan.baidu.com/s/1Uw6SXS8z_IxdkNpLr9y0zQ?pwd=s2i5

Extraction code: s2i5

Power-on password: jerry/123456

Installed Apache, Tomcat, MySQL, vsftpd and supporting Web security testing practice teaching plan.

start Tomcat

#/usr/local/apache-tomcat-8.5.81/bin/startup.sh

start MySQL

#service mysql start

start Apache

#/etc/init.d/apache2 start

Open a browser and enter 127.0.0.1:8080/sec/

5) Metasploitable2-Linux (with vsftpd 2.3.4)

Link: https://pan.baidu.com/s/1a71zOXGi_9aLrXyEnvkHwQ?pwd=17g6

Extraction code: 17g6

Power -on password : see the page prompt

After decompression, it is directly vmx file, which can be used directly

picture