keras入门－简单数据集的运用

Keras:

Keras是一个高层神经网络API，Keras由纯Python编写而成并基Tensorflow、Theano以及CNTK后端。Keras为支持快速实验而生.

Keras的核心数据结构是“模型”，模型是一种组织网络层的方式。Keras中主要的模型是Sequential模型，Sequential是一系列网络层按顺序构成的栈。

下面我就用两个经典数据集iris 和　mnist　来解释Keras是如何完成逻辑分类的

iris：

# coding=utf-8import matplotlib.pyplot as pltimport seaborn as snsimport numpy as npfrom sklearn.cross_validation import train_test_splitfrom sklearn.linear_model import LogisticRegressionCVfrom keras.models import Sequentialfrom keras.layers.core import Dense, Activationfrom keras.utils import np_utils

这里用到sklearn的两种切分数据集的方法：

from sklearn.model_selection import ShuffleSplitrs = ShuffleSplit(n_splits=1, train_size=0.6, test_size=0.4, random_state=1)＃n_splits表示只需要一个样本结果，现在训练－验证集比例为：　６：４，　random_state为洗牌次数rs.get_n_splits(X)X_trainset = NoneX_testset = Noney_trainset = Noney_testset = Nonefor train_index, test_index in rs.split(X, y):X_trainset, X_testset = X[train_index], X[test_index]y_trainset, y_testset = y[train_index], y[test_index]

２from sklearn.cross_validation import train_test_splittrain_X, test_X, train_y, test_y = train_test_split(X, y, train_size=0.1, random_state=0)＃随机切分

所有的Ｙ都转化为one-hot函数：

def one_hot_encode_object_array(arr):    uniques, ids = np.unique(arr, return_inverse=True)    return np_utils.to_categorical(ids, len(uniques))

iris = sns.load_dataset("iris")# 导入数据集irisX = iris.values[:, :4]y = iris.values[:, 4]#划分数据集train_X, test_X, train_y, test_y = train_test_split(X, y, train_size=0.6, random_state=0)print train_Xtrain_y_ohe = one_hot_encode_object_array(train_y)test_y_ohe = one_hot_encode_object_array(test_y)

模型：

#定义模型model = Sequential()# 输入层大小，隐含层大小，激活函数model.add(Dense(200,input_shape=(4,)))model.add(Activation('relu'))#输出层大小，激活函数,两层网络model.add(Dense(3))model.add(Activation('softmax'))#编译模型model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=["accuracy"])#训练模型model.fit(train_X, train_y_ohe, nb_epoch=100, batch_size=1, verbose=1)

#评估模型成绩# predict_y = model.predict_proba(test_X)# print predict_yloss, accuracy = model.evaluate(test_X,test_y_ohe,verbose=0)print ("Accuracy = {:.2f}".format(accuracy))

mnist:

import numpy as npimport kerasnp.random.seed(1337)#随机种子from keras.datasets import mnistfrom keras.models import Sequentialfrom keras.layers import Densefrom keras.utils import np_utils

batch_size = 128nb_classes = 10＃类别，数字０－９np_epoch = 10＃梯度下降循环训练次数img_size = 28*28＃输入图片大小

(X_train,y_train),(X_test,y_test) = mnist.load_data()＃加载数据集，已经划分好，shuffle-splitX_train = X_train.reshape(y_train.shape[0],img_size).astype("float32")/255X_test =  X_test.reshape(y_test.shape[0],img_size).astype("float32")/255＃生成图片向量，颜色１－２５５print X_train.shapeprint X_test.shape

#老规矩，把Ｙ变为one-hot编码Y_train = np_utils.to_categorical(y_train,nb_classes)Y_test = np_utils.to_categorical(y_test,nb_classes)

model = Sequential([Dense(10,input_shape=(img_size,),activation="softmax"),])＃优化模型，随机梯度下降（rmsprop),loss_funtion(交叉熵）model.compile(optimizer='rmsprop',loss='categorical_crossentropy',metrics=['accuracy'])model.fit(X_train,Y_train,batch_size=batch_size,nb_epoch=1,verbose=1,validation_data=(X_test,Y_test))

评估模型成绩：

score = model.evaluate(X_test,Y_test,verbose=0)print ('accuracy:{}'.format(score[1]))

my github:

https://github.com/TFknight/PythonStudy/tree/master/keras-learn