tensorflow之逻辑回归模型实现
前面一篇介绍了用tensorflow实现线性回归模型预测sklearn内置的波士顿房价,现在这一篇就记一下用逻辑回归分类sklearn提供的乳腺癌数据集,该数据集有569个样本,每个样本有30维,为二分类数据集,212个正样本,357个负样本。
首先,加载数据,并划分训练集和测试集:
# 加载乳腺癌数据集,该数据及596个样本,每个样本有30维,共有两类
cancer = skd.load_breast_cancer()
# 将数据集的数据和标签分离
X_data = cancer.data
Y_data = cancer.target
x_train,x_test,y_train,y_test = train_test_split(X_data,Y_data,test_size=0.3,random_state=0)
这里还要注意一下,因为后面要用到的损失函数为交叉熵,而数据集的标签是一维的,所以我们需要将其转换为位数来表示类别,也就是[0,1]=>[[0,1],[1,0]]这种形式的标签:
y_train_new = np.zeros([y_train.shape[0], 2], dtype='int32')
y_test_new = np.zeros([y_test.shape[0], 2], dtype='int32')
for i in range(y_train.shape[0]):
if y_train[i] == 0:
y_train_new[i,0] = 0
y_train_new[i,1] = 1
else:
y_train_new[i,0] = 1
y_train_new[i,1] = 0
for i in range(y_test.shape[0]):
if y_test[i] == 0:
y_test_new[i,0] = 0
y_test_new[i,1] = 1
else:
y_test_new[i,0] = 1
y_test_new[i,1] = 0
接着就是老套路,初始化训练参数,设置模型输入输出,然后是构建模型,二分类逻辑回归模型的数学表达式为:

用tensorflow实现代码如下:
pred = tf.nn.sigmoid(tf.matmul(X, W) + b) # sigmoid
然后是构建交叉熵表达式,其数学公式为:

用tensorflow实现代码如下:
cost = tf.reduce_mean(-tf.reduce_sum(Y*tf.log(pred), reduction_indices=1))
对于训练好的模型的测试性能的评价,这里我采用计算ROC曲线的方式来展现。首先是模型对于输入数据的预测会得到一个对应的1*2的输出,这个输出不一定是很好的[0 1]或者[1 0],而是一个类似于概率的量,也就是类似于这样[0.9527, 0.0473],所以我们需要判断,如果第一位小于第二位,依照前面的设定应该是0,反之为1,当然可能会有第一位等于第二位的情况,那就只能是存在误差了:
y_test_pred = sess.run(pred, feed_dict={X: x_test})
y_scores = np.empty((y_test_pred.shape[0]))
for i in range(y_scores.shape[0]):
if y_test_pred[i,0]<y_test_pred[i,1]:
y_scores[i]=0
else:
y_scores[i]=1
然后roc曲线和其AUC值这里调用的是sklearn提供的函数:
fpr, tpr, thresholds = roc_curve((y_test), y_scores)
AUC_ROC = roc_auc_score(y_test, y_scores)
然后是画出precision recall curve(精确率-召回率曲线),同样也是调用sklearn的函数:
precision, recall, thresholds = precision_recall_curve(y_test, y_scores)
precision = np.fliplr([precision])[0] #so the array is increasing (you won't get negative AUC)
recall = np.fliplr([recall])[0] #so the array is increasing (you won't get negative AUC) AUC_prec_rec = np.trapz(precision,recall)
接着是计算置信度矩阵,得到矩阵后顺带的就可以计算准确率、灵敏性、特异性等几个参数了,同样的,还是调用sklearn的函数即可:
confusion = confusion_matrix(y_test, y_pred)
accuracy = float(confusion[0,0]+confusion[1,1])/float(np.sum(confusion))
specificity = float(confusion[0,0])/float(confusion[0,0]+confusion[0,1])
sensitivity = float(confusion[1,1])/float(confusion[1,1]+confusion[1,0])
precision = float(confusion[1,1])/float(confusion[1,1]+confusion[0,1])
同样的衡量指标还有Jaccard系数和F1分数:
jaccard_index = jaccard_similarity_score(y_test, y_pred, normalize=True)
F1_score = f1_score(y_test, y_pred, labels=None, average='binary', sample_weight=None)
这里涉及到了很多模型性能的评价指标,具体的含义就不细究了,下一次专门写一篇来总结吧。完整的代码如下:
from __future__ import print_function
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
from sklearn import datasets as skd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import scale
from sklearn.metrics import roc_curve
from sklearn.metrics import roc_auc_score
from sklearn.metrics import confusion_matrix
from sklearn.metrics import precision_recall_curve
from sklearn.metrics import jaccard_similarity_score
from sklearn.metrics import f1_score
# 加载乳腺癌数据集,该数据及596个样本,每个样本有30维,共有两类
cancer = skd.load_breast_cancer()
# 将数据集的数据和标签分离
X_data = cancer.data
Y_data = cancer.target
print("X_data.shape = ", X_data.shape)
print("Y_data.shape = ", Y_data.shape)
# 将数据和标签分成训练集和测试集
x_train,x_test,y_train,y_test = train_test_split(X_data,Y_data,test_size=0.3,random_state=0)
x_train = scale(x_train)
x_test = scale(x_test)
y_train_new = np.zeros([y_train.shape[0], 2], dtype='int32')
y_test_new = np.zeros([y_test.shape[0], 2], dtype='int32')
for i in range(y_train.shape[0]):
if y_train[i] == 0:
y_train_new[i,0] = 0
y_train_new[i,1] = 1
else:
y_train_new[i,0] = 1
y_train_new[i,1] = 0
for i in range(y_test.shape[0]):
if y_test[i] == 0:
y_test_new[i,0] = 0
y_test_new[i,1] = 1
else:
y_test_new[i,0] = 1
y_test_new[i,1] = 0
print("x_train.shape = ", x_train.shape)
print("x_test.shape = ", x_test.shape)
print("y_train.shape = ", y_train_new.shape)
print("y_test.shape = ", y_test_new.shape)
# 初始化参数
learning_rate = 0.01
training_epochs = 50000
display_step = 50
# 定义图模型输入
X = tf.placeholder(tf.float32, [None, 30]) # mnist data image of shape 28*28=784
Y = tf.placeholder(tf.float32, [None, 2]) # 0-9 digits recognition => 10 classes#
# 设置模型权重和偏置
W = tf.Variable(tf.random_normal([30, 2]),dtype=tf.float32, name="weight")
b = tf.Variable(tf.random_normal([2]),dtype=tf.float32, name="bias")
# 构建模型
#pred = tf.nn.softmax(tf.matmul(x, W) + b) # Softmax
pred = tf.nn.sigmoid(tf.matmul(X, W) + b) # sigmoid
# 使用交叉熵来最小化训练误差
cost = tf.reduce_mean(-tf.reduce_sum(Y*tf.log(pred), reduction_indices=1))
# 梯度下降法
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)#
# 初始化器
init = tf.global_variables_initializer()
# 开始训练
with tf.Session() as sess:#
# 初始化
sess.run(init)#
# 迭代训练
for epoch in range(training_epochs):
avg_cost = 0.
sess.run(optimizer, feed_dict={X: x_train, Y: y_train_new})
# 显示训练信息
if (epoch+1) % display_step == 0:
c = sess.run(cost, feed_dict={X: x_train, Y:y_train_new})
print("Epoch:", '%04d' % (epoch+1), "cost=", "{:.9f}".format(c))
print("完成训练!")
# 测试模型
correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(Y, 1))
# 计算准确度
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
print("Accuracy:", accuracy.eval({X: x_test, Y: y_test_new}))
# 计算模型预测
y_test_pred = sess.run(pred, feed_dict={X: x_test})
y_pred = np.empty((y_test_pred.shape[0]), dtype='int32')
# 二值化
for i in range(y_pred.shape[0]):
if y_test_pred[i,0]<y_test_pred[i,1]:
y_pred[i]=0
else:
y_pred[i]=1
# 得到ROC曲线
fpr, tpr, thresholds = roc_curve((y_test), y_pred)
AUC_ROC = roc_auc_score(y_test, y_pred)
roc_curve =plt.figure()
plt.plot(fpr,tpr,'-',label='Area Under the Curve (AUC = %0.4f)' % AUC_ROC)
plt.title('ROC curve')
plt.xlabel("FPR (False Positive Rate)")
plt.ylabel("TPR (True Positive Rate)")
plt.legend(loc="lower right")
plt.show()
# 得到precision_recall曲线
precision, recall, thresholds = precision_recall_curve(y_test, y_pred)
precision = np.fliplr([precision])[0] #so the array is increasing (you won't get negative AUC)
recall = np.fliplr([recall])[0] #so the array is increasing (you won't get negative AUC)
AUC_prec_rec = np.trapz(precision,recall)
print("\nArea under Precision-Recall curve: " +str(AUC_prec_rec))
prec_rec_curve = plt.figure()
plt.plot(recall,precision,'-',label='Area Under the Curve (AUC = %0.4f)' % AUC_prec_rec)
plt.title('Precision - Recall curve')
plt.xlabel("Recall")
plt.ylabel("Precision")
plt.legend(loc="lower right")
plt.show()
# 计算置信度矩阵
threshold_confusion = 0.5
print("\nConfusion matrix: Costum threshold (for positive) of " +str(threshold_confusion))
confusion = confusion_matrix(y_test, y_pred)
print(confusion)
accuracy = 0
if float(np.sum(confusion))!=0:
accuracy = float(confusion[0,0]+confusion[1,1])/float(np.sum(confusion))
print("Global Accuracy: " +str(accuracy))
specificity = 0
if float(confusion[0,0]+confusion[0,1])!=0:
specificity = float(confusion[0,0])/float(confusion[0,0]+confusion[0,1])
print("Specificity: " +str(specificity))
sensitivity = 0
if float(confusion[1,1]+confusion[1,0])!=0:
sensitivity = float(confusion[1,1])/float(confusion[1,1]+confusion[1,0])
print("Sensitivity: " +str(sensitivity))
precision = 0
if float(confusion[1,1]+confusion[0,1])!=0:
precision = float(confusion[1,1])/float(confusion[1,1]+confusion[0,1])
print("Precision: " +str(precision))
#Jaccard similarity index
jaccard_index = jaccard_similarity_score(y_test, y_pred, normalize=True)
print("\nJaccard similarity score: " +str(jaccard_index))
#F1 score
F1_score = f1_score(y_test, y_pred, labels=None, average='binary', sample_weight=None)
print("\nF1 score (F-measure): " +str(F1_score))
print("y_test[0:20]=", y_test[0:20])
print("y_pred[0:20]=", y_pred[0:20])
最总的结果显示如下:



古人学问无遗力,少壮工夫老始成。
纸上得来终觉浅,绝知此事要躬行。
-- 陆游 《冬夜读书示子聿》
tensorflow之逻辑回归模型实现的更多相关文章
- 利用Tensorflow实现逻辑回归模型
官方mnist代码: #下载Mnist数据集 import tensorflow.examples.tutorials.mnist.input_data mnist = input_data.read ...
- 深度学习实践系列(1)- 从零搭建notMNIST逻辑回归模型
MNIST 被喻为深度学习中的Hello World示例,由Yann LeCun等大神组织收集的一个手写数字的数据集,有60000个训练集和10000个验证集,是个非常适合初学者入门的训练集.这个网站 ...
- 逻辑回归模型(Logistic Regression, LR)基础
逻辑回归模型(Logistic Regression, LR)基础 逻辑回归(Logistic Regression, LR)模型其实仅在线性回归的基础上,套用了一个逻辑函数,但也就由于这个逻辑函 ...
- Python之逻辑回归模型来预测
建立一个逻辑回归模型来预测一个学生是否被录取. import numpy as np import pandas as pd import matplotlib.pyplot as plt impor ...
- 逻辑回归模型(Logistic Regression)及Python实现
逻辑回归模型(Logistic Regression)及Python实现 http://www.cnblogs.com/sumai 1.模型 在分类问题中,比如判断邮件是否为垃圾邮件,判断肿瘤是否为阳 ...
- 逻辑回归模型(Logistic Regression, LR)--分类
逻辑回归(Logistic Regression, LR)模型其实仅在线性回归的基础上,套用了一个逻辑函数,但也就由于这个逻辑函数,使得逻辑回归模型成为了机器学习领域一颗耀眼的明星,更是计算广告学的核 ...
- 吴恩达机器学习笔记22-正则化逻辑回归模型(Regularized Logistic Regression)
针对逻辑回归问题,我们在之前的课程已经学习过两种优化算法:我们首先学习了使用梯度下降法来优化代价函数
- tensorflow 实现逻辑回归——原以为TensorFlow不擅长做线性回归或者逻辑回归,原来是这么简单哇!
实现的是预测 低 出生 体重 的 概率.尼克·麦克卢尔(Nick McClure). TensorFlow机器学习实战指南 (智能系统与技术丛书) (Kindle 位置 1060-1061). Kin ...
- tensorflow学习笔记五----------逻辑回归
在逻辑回归中使用mnist数据集.导入相应的包以及数据集. import numpy as np import tensorflow as tf import matplotlib.pyplot as ...
随机推荐
- 为 git 设定 socks5 代理
为 git 设定 socks5 代理 查看当前设定 git config --global -l 为 git 设定全局代理 git config --global http.proxy socks5h ...
- SQL 查询每组的第一条记录
CREATE TABLE [dbo].[test1]( [program_id] [int] NULL, [person_id] [int] NULL ) ON [PRIMARY] /*查询每组分组中 ...
- #P1909 买铅笔 的题解
题目描述 P老师需要去商店买n支铅笔作为小朋友们参加NOIP的礼物.她发现商店一共有 33种包装的铅笔,不同包装内的铅笔数量有可能不同,价格也有可能不同.为了公平起 见,P老师决定只买同一种包装的铅笔 ...
- cross_val_score 交叉验证与 K折交叉验证,嗯都是抄来的,自己作个参考
因为sklearn cross_val_score 交叉验证,这个函数没有洗牌功能,添加K 折交叉验证,可以用来选择模型,也可以用来选择特征 sklearn.model_selection.cross ...
- Linux 下面搭建KMS服务器
1. 下载安装Linux 版的KMS服务软件 # wget https://github.com/Wind4/vlmcsd/releases/download/svn1111/binaries.tar ...
- android开发如何在页面之间传参
第一个页面跳转 传递值 Button bn1=(Button)findViewById(R.id.btn_Login); //跳转bn1.setOnClickListener(new OnClickL ...
- Linux引导过程!(重点)
1.开机自检:服务器开机后,将根据主板BIOS (基本输入系统),中的设置对CPU .内存.显卡.键盘等设备进行初步检测,并初始化部分硬件. 2.MBR 引导:当从本机硬盘中启动系统时,首先根据硬盘中 ...
- Linux CentOS7 VMware 安装PHP5 、安装PHP7
一.安装PHP5 PHP官网www.php.net 当前主流版本为5.6/7.1 cd /usr/local/src/ wget http://cn2.php.net/distributions/ph ...
- 「NOIP2017」逛公园
传送门 Luogu 解题思路 考虑 \(\text{DP}\). 设 \(f[u][k]\) 表示从 \(u\) 到 \(n\) 走过不超过 \(Mindis(u, n) + k\) 距离的方案数. ...
- Controller生命周期
1. 实例化 alloc/init, initWithNibName 2.awakeFromNib 从nib创建Controller对象 3.get/set outlets 4. viewDidLoa ...