6.3神经网络实例1--python机器学习

参考彭亮老师的视频教程：转载请注明出处及彭亮老师原创
视频教程： http://pan.baidu.com/s/1kVNe5EJ

1. 简单非线性关系数据集测试(XOR):

X:                  Y

0 0                 0

0 1                 1

1 0                 1

1 1                 0

Code:

from NeuralNetwork import NeuralNetwork

import numpy as np

nn = NeuralNetwork([2,2,1], 'tanh')     

X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])     

y = np.array([0, 1, 1, 0])     

nn.fit(X, y)     

for i in [[0, 0], [0, 1], [1, 0], [1,1]]:    

    print(i, nn.predict(i))

2. 手写数字识别：

每个图片8x8 

识别数字：0,1,2,3,4,5,6,7,8,9

Code:

import numpy as np 

from sklearn.datasets import load_digits 

from sklearn.metrics import confusion_matrix, classification_report 

from sklearn.preprocessing import LabelBinarizer 

from NeuralNetwork import NeuralNetwork

from sklearn.cross_validation import train_test_split

digits = load_digits()  

X = digits.data  

y = digits.target  

X -= X.min() # normalize the values to bring them into the range 0-1  

X /= X.max()

nn = NeuralNetwork([64,100,10],'logistic')  

X_train, X_test, y_train, y_test = train_test_split(X, y)  

labels_train = LabelBinarizer().fit_transform(y_train)  

labels_test = LabelBinarizer().fit_transform(y_test)

print "start fitting"

nn.fit(X_train,labels_train,epochs=3000)  

predictions = []  

for i in range(X_test.shape[0]):  

    o = nn.predict(X_test[i] )  

    predictions.append(np.argmax(o))  

print confusion_matrix(y_test,predictions)  

print classification_report(y_test,predictions)