SVM支持向量机Tensorflow实现
来源:互联网 发布:制作书本的软件 编辑:程序博客网 时间:2024/06/05 05:00
一、tensorflow实现SVM
# -- coding: utf-8 --import matplotlib.pyplot as pltimport numpy as npimport tensorflow as tffrom sklearn import datasets# 获取数据iris = datasets.load_iris()x_vals = np.array([[x[0], x[3]] for x in iris.data])y_vals = np.array([1 if y == 0 else -1 for y in iris.target])# 分离训练和测试集train_indices = np.random.choice(len(x_vals),int(len(x_vals)*0.8),replace=False)test_indices = np.array(list(set(range(len(x_vals))) - set(train_indices)))x_vals_train = x_vals[train_indices]x_vals_test = x_vals[test_indices]y_vals_train = y_vals[train_indices]y_vals_test = y_vals[test_indices]batch_size = 100# 初始化feedinx_data = tf.placeholder(shape=[None, 2], dtype=tf.float32)y_target = tf.placeholder(shape=[None, 1], dtype=tf.float32)# 创建权值参数A = tf.Variable(tf.random_normal(shape=[2, 1]))b = tf.Variable(tf.random_normal(shape=[1, 1]))A2 = tf.Variable(tf.random_normal(shape=[2, 1]))b2 = tf.Variable(tf.random_normal(shape=[1, 1]))# 定义线性模型: y = Ax + bmodel_output = tf.subtract(tf.matmul(x_data, A), b)model_output2 = tf.subtract(tf.matmul(x_data, A2), b2)# Declare vector L2 'norm' function squaredl2_norm = tf.reduce_sum(tf.square(A))# Loss = max(0, 1-pred*actual) + alpha * L2_norm(A)^2alpha = tf.constant([0.01])classification_term = tf.reduce_mean(tf.maximum(0., tf.subtract(1., tf.multiply(model_output, y_target))))classification_term2 = tf.reduce_mean(tf.maximum(0., tf.subtract(1., tf.multiply(model_output2, y_target))))loss = tf.add(classification_term, tf.multiply(alpha, l2_norm))loss2 = tf.add(classification_term2,[0])my_opt = tf.train.GradientDescentOptimizer(0.01)train_step = my_opt.minimize(loss)my_opt2 = tf.train.GradientDescentOptimizer(0.01)train_step2 = my_opt2.minimize(loss2)with tf.Session() as sess: init = tf.global_variables_initializer() sess.run(init)# Training loop loss_vec = [] train_accuracy = [] test_accuracy = [] for i in range(20000): rand_index = np.random.choice(len(x_vals_train), size=batch_size) rand_x = x_vals_train[rand_index] rand_y = np.transpose([y_vals_train[rand_index]]) sess.run(train_step, feed_dict={x_data: rand_x, y_target: rand_y}) sess.run(train_step2, feed_dict={x_data: rand_x, y_target: rand_y}) [[a1], [a2]] = sess.run(A) [[b]] = sess.run(b) slope = -a2/a1 y_intercept = b/a1 best_fit = [] [[a12], [a22]] = sess.run(A2) [[b2]] = sess.run(b2) slope2 = -a22/a12 y_intercept2 = b2/a12 best_fit2 = [] x1_vals = [d[1] for d in x_vals] for i in x1_vals: best_fit.append(slope*i+y_intercept) best_fit2.append(slope2*i+y_intercept2)# Separate I. setosa setosa_x = [d[1] for i, d in enumerate(x_vals) if y_vals[i] == 1] setosa_y = [d[0] for i, d in enumerate(x_vals) if y_vals[i] == 1] not_setosa_x = [d[1] for i, d in enumerate(x_vals) if y_vals[i] == -1] not_setosa_y = [d[0] for i, d in enumerate(x_vals) if y_vals[i] == -1] plt.plot(setosa_x, setosa_y, 'o', label='I. setosa') plt.plot(not_setosa_x, not_setosa_y, 'x', label='Non-setosa') plt.plot(x1_vals, best_fit, 'r-', label='Linear Separator + w', linewidth=3) plt.plot(x1_vals, best_fit2, 'r-', label='Linear Separator', color='b', linewidth=3) plt.ylim([0, 10]) plt.legend(loc='lower right') plt.title('Sepal Length vs Pedal Width') plt.xlabel('Pedal Width') plt.ylabel('Sepal Length') plt.show()代码中创建了两个线性模型,再计算损失函数loss时候,一个加了||w||平方,一个没加,所以绘图的时候会有两条线,红色线条实现了支持向量到现行模型距离最大化,可以更好的预测未知模型。
二、tensorflow实现SVM,并保存使用模型
训练代码:
# -- coding: utf-8 --import matplotlib.pyplot as pltimport numpy as npimport tensorflow as tffrom sklearn import datasets# 获取数据iris = datasets.load_iris()x_vals = np.array([[x[0], x[3]] for x in iris.data])y_vals = np.array([1 if y == 0 else -1 for y in iris.target])# 分离训练和测试集train_indices = np.random.choice(len(x_vals),int(len(x_vals)*0.8),replace=False)test_indices = np.array(list(set(range(len(x_vals))) - set(train_indices)))x_vals_train = x_vals[train_indices]x_vals_test = x_vals[test_indices]y_vals_train = y_vals[train_indices]y_vals_test = y_vals[test_indices]batch_size = 100# 初始化feedinx_data = tf.placeholder(shape=[None, 2], dtype=tf.float32)y_target = tf.placeholder(shape=[None, 1], dtype=tf.float32)# 创建权值参数A = tf.Variable(tf.random_normal(shape=[2, 1]))b = tf.Variable(tf.random_normal(shape=[1, 1]))# 定义线性模型: y = Ax + bmodel_output = tf.subtract(tf.matmul(x_data, A), b)# Declare vector L2 'norm' function squaredl2_norm = tf.reduce_sum(tf.square(A))# Loss = max(0, 1-pred*actual) + alpha * L2_norm(A)^2alpha = tf.constant([0.01])classification_term = tf.reduce_mean(tf.maximum(0., tf.subtract(1., tf.multiply(model_output, y_target))))loss = tf.add(classification_term, tf.multiply(alpha, l2_norm))my_opt = tf.train.GradientDescentOptimizer(0.01)train_step = my_opt.minimize(loss)#持久化saver = tf.train.Saver()with tf.Session() as sess: init = tf.global_variables_initializer() sess.run(init)# Training loop for i in range(20000): rand_index = np.random.choice(len(x_vals_train), size=batch_size) rand_x = x_vals_train[rand_index] rand_y = np.transpose([y_vals_train[rand_index]]) sess.run(train_step, feed_dict={x_data: rand_x, y_target: rand_y}) saver.save(sess, "./model/model.ckpt")
使用判断代码:
# -- coding: utf-8 --import matplotlib.pyplot as pltimport numpy as npimport tensorflow as tffrom sklearn import datasets# 获取数据iris = datasets.load_iris()x_vals = np.array([[x[0], x[3]] for x in iris.data])y_vals = np.array([1 if y == 0 else -1 for y in iris.target])# 分离训练和测试集test_indices = np.random.choice(len(x_vals),int(len(x_vals)*0.8),replace=False)x_vals_test = x_vals[test_indices]y_vals_test = y_vals[test_indices]# 初始化feedinx_data = tf.placeholder(shape=[None, 2], dtype=tf.float32)y_target = tf.placeholder(shape=[None, 1], dtype=tf.float32)# 创建权值参数A = tf.Variable(tf.random_normal(shape=[2, 1]))b = tf.Variable(tf.random_normal(shape=[1, 1]))# 定义线性模型: y = Ax + bmodel_output = tf.subtract(tf.matmul(x_data, A), b)#判断准确度result = tf.maximum(0., tf.multiply(model_output, y_target))saver = tf.train.Saver()with tf.Session() as sess: saver.restore(sess, "./model/model.ckpt") y_test = np.reshape(y_vals_test, (120,1)) array = sess.run(result, feed_dict={x_data: x_vals_test, y_target: y_test}) num = np.array(array) zero_num = np.sum(num==[0]) print(num) print(zero_num)
阅读全文
1 0
- SVM支持向量机Tensorflow实现
- MATLAB支持向量机SVM代码实现
- SVM 支持向量机 opencv实现
- python实现支持向量机SVM算法
- 支持向量机(SVM)的实现
- SVM支持向量机代码实现流程
- 支持向量机SVM的MATLAB实现
- 支持向量机SVM
- SVM支持向量机
- svm支持向量机
- SVM支持向量机
- [SVM]支持向量机
- SVM 支持向量机
- 支持向量机SVM
- SVM 支持向量机
- svm支持向量机
- 支持向量机SVM
- svm支持向量机
- 购票系统
- Redis持久化--RDB+AOF
- 机器学习实战——决策树
- docker安装redis
- Android中ListView下拉刷新功能
- SVM支持向量机Tensorflow实现
- JS中函数名后面的括号加与不加的区别
- Android原生登录同步到webview的网页
- theano与keras安装问题
- Python3,在Django中使用easyui
- MyC++基础知识补漏
- MIT 6.824 lab2 启动流程以及raft算法实现
- WCF 一步一步 发布 WCF服务 到 IIS (图)
- OAuth 2.0 ——授权码模式