分类预测输出precision，recall，accuracy，auc和tp，tn，fp，fn矩阵

此次我做的实验是二分类问题，输出precision，recall，accuracy，auc

# -*- coding: utf-8 -*-

#from sklearn.neighbors import

import numpy as np

from pandas import read_csv

import pandas as pd

import sys

import importlib

from sklearn.neighbors import KNeighborsClassifier

from sklearn.ensemble import GradientBoostingClassifier

from sklearn import svm

from sklearn import cross_validation

from sklearn.metrics import hamming_loss

from sklearn import metrics

importlib.reload(sys)

from sklearn.linear_model import LogisticRegression

from imblearn.combine import SMOTEENN

from sklearn.tree import DecisionTreeClassifier

from sklearn.ensemble import RandomForestClassifier  #92%

from sklearn import tree

from xgboost.sklearn import XGBClassifier

from sklearn.linear_model import SGDClassifier

from sklearn import neighbors

from sklearn.naive_bayes import BernoulliNB

import matplotlib as mpl

import matplotlib.pyplot as plt

from sklearn.metrics import confusion_matrix

from numpy import mat

def metrics_result(actual, predict):

    print('准确度:{0:.3f}'.format(metrics.accuracy_score(actual, predict)))

    print('精密度:{0:.3f}'.format(metrics.precision_score(actual, predict,average='weighted')))

    print('召回:{0:0.3f}'.format(metrics.recall_score(actual, predict,average='weighted')))

    print('f1-score:{0:.3f}'.format(metrics.f1_score(actual, predict,average='weighted')))

    print('auc:{0:.3f}'.format(metrics.roc_auc_score(test_y, predict)))

输出混淆矩阵

    matr=confusion_matrix(test_y,predict)

    matr=mat(matr)

    conf=np.matrix([[0,0],[0,0]])

    conf[0,0]=matr[1,1]

    conf[1,0]=matr[1,0]

    conf[0,1]=matr[0,1]

    conf[1,1]=matr[0,0]

    print(conf)

全代码：

# -*- coding: utf-8 -*-

#from sklearn.neighbors import

import numpy as np

from pandas import read_csv

import pandas as pd

import sys

import importlib

from sklearn.neighbors import KNeighborsClassifier

from sklearn.ensemble import GradientBoostingClassifier

from sklearn import svm

from sklearn import cross_validation

from sklearn.metrics import hamming_loss

from sklearn import metrics

importlib.reload(sys)

from sklearn.linear_model import LogisticRegression

from imblearn.combine import SMOTEENN

from sklearn.tree import DecisionTreeClassifier

from sklearn.ensemble import RandomForestClassifier  #92%

from sklearn import tree

from xgboost.sklearn import XGBClassifier

from sklearn.linear_model import SGDClassifier

from sklearn import neighbors

from sklearn.naive_bayes import BernoulliNB

import matplotlib as mpl

import matplotlib.pyplot as plt

from sklearn.metrics import confusion_matrix

from numpy import mat

def metrics_result(actual, predict):

    print('准确度:{0:.3f}'.format(metrics.accuracy_score(actual, predict)))

    print('精密度:{0:.3f}'.format(metrics.precision_score(actual, predict,average='weighted')))

    print('召回:{0:0.3f}'.format(metrics.recall_score(actual, predict,average='weighted')))

    print('f1-score:{0:.3f}'.format(metrics.f1_score(actual, predict,average='weighted')))

    print('auc:{0:.3f}'.format(metrics.roc_auc_score(test_y, predict)))

'''分类0-1'''

root1="D:/ProgramData/station3/10.csv"

root2="D:/ProgramData/station3/more+average2.csv"

root3="D:/ProgramData/station3/new_10.csv"

root4="D:/ProgramData/station3/more+remove.csv"

root5="D:/ProgramData/station3/new_10 2.csv"

root6="D:/ProgramData/station3/new10.csv"

root7="D:/ProgramData/station3/no_-999.csv"

root=root4

data1 = read_csv(root) #数据转化为数组

data1=data1.values

print(root)

time=1

accuracy=[]

aucc=[]

pre=[]

recall=[]

for i in range(time):

    train, test= cross_validation.train_test_split(data1, test_size=0.2, random_state=i)

    test_x=test[:,:-1]

    test_y=test[:,-1]

    train_x=train[:,:-1]

    train_y=train[:,-1]

# =============================================================================

#     print(train_x.shape)

#     print(train_y.shape)

#     print(test_x.shape)

#     print(test_y.shape)

#     print(type(train_x))

# =============================================================================

    #X_Train=train_x

    #Y_Train=train_y

    X_Train, Y_Train = SMOTEENN().fit_sample(train_x, train_y)

    #clf = RandomForestClassifier()  #82

    #clf = LogisticRegression()  #82 

    #penalty=’l2’, dual=False, tol=0.0001, C=1.0, fit_intercept=True, intercept_scaling=1, class_weight=None, random_state=None, solver=’liblinear’, max_iter=100, multi_class=’ovr’, verbose=0, warm_start=False, n_jobs=1

    #clf=svm.SVC()

    clf= XGBClassifier()

    #from sklearn.ensemble import RandomForestClassifier  #92%

    #clf = DecisionTreeClassifier()

    #clf = GradientBoostingClassifier()

    #clf=neighbors.KNeighborsClassifier()

    #clf=BernoulliNB()

    print(clf)

    clf.fit(X_Train, Y_Train)

    predict=clf.predict(test_x)

    matr=confusion_matrix(test_y,predict)

    matr=mat(matr)

    conf=np.matrix([[0,0],[0,0]])

    conf[0,0]=matr[1,1]

    conf[1,0]=matr[1,0]

    conf[0,1]=matr[0,1]

    conf[1,1]=matr[0,0]

    print(conf)

    #a=metrics_result(test_y, predict)

    #a=metrics_result(test_y,predict)

    '''accuracy'''

    aa=metrics.accuracy_score(test_y, predict)

    #print(metrics.accuracy_score(test_y, predict))

    accuracy.append(aa)

    '''auc'''

    bb=metrics.roc_auc_score(test_y, predict, average=None)

    aucc.append(bb)

    '''precision'''

    cc=metrics.precision_score(test_y, predict, average=None)

    pre.append(cc[1])

# =============================================================================

#     print('cc')

#     print(type(cc))

#     print(cc[1])

#     print('cc')

# =============================================================================

    '''recall'''

    dd=metrics.recall_score(test_y, predict, average=None)

    #print(metrics.recall_score(test_y, predict,average='weighted'))

    recall.append(dd[1])

    f=open('D:\ProgramData\station3\predict.txt', 'w')

    for i in range(len(predict)):

        f.write(str(predict[i]))

        f.write('\n')

    f.write("写好了")

    f.close()

    f=open('D:\ProgramData\station3\y_.txt', 'w')

    for i in range(len(predict)):

        f.write(str(test_y[i]))

        f.write('\n')

    f.write("写好了")

    f.close()

# =============================================================================

#     f=open('D:/ProgramData/station3/predict.txt', 'w')

#     for i in range(len(predict)):

#        f.write(str(predict[i]))

#        f.write('\n')

#     f.write("写好了")

#     f.close()

#

#     f=open('D:/ProgramData/station3/y.txt', 'w')

#     for i in range(len(test_y)):

#        f.write(str(test_y[i]))

#        f.write('\n')

#     f.write("写好了")

#     f.close()

#

# =============================================================================

# =============================================================================

#     print('调用函数auc：', metrics.roc_auc_score(test_y, predict, average='micro'))

#

#     fpr, tpr, thresholds = metrics.roc_curve(test_y.ravel(),predict.ravel())

#     auc = metrics.auc(fpr, tpr)

#     print('手动计算auc：', auc)

#     #绘图

#     mpl.rcParams['font.sans-serif'] = u'SimHei'

#     mpl.rcParams['axes.unicode_minus'] = False

#     #FPR就是横坐标,TPR就是纵坐标

#     plt.plot(fpr, tpr, c = 'r', lw = 2, alpha = 0.7, label = u'AUC=%.3f' % auc)

#     plt.plot((0, 1), (0, 1), c = '#808080', lw = 1, ls = '--', alpha = 0.7)

#     plt.xlim((-0.01, 1.02))

#     plt.ylim((-0.01, 1.02))

#     plt.xticks(np.arange(0, 1.1, 0.1))

#     plt.yticks(np.arange(0, 1.1, 0.1))

#     plt.xlabel('False Positive Rate', fontsize=13)

#     plt.ylabel('True Positive Rate', fontsize=13)

#     plt.grid(b=True, ls=':')

#     plt.legend(loc='lower right', fancybox=True, framealpha=0.8, fontsize=12)

#     plt.title(u'大类问题一分类后的ROC和AUC', fontsize=17)

#     plt.show()

# =============================================================================

sum_acc=0

sum_auc=0

sum_pre=0

sum_recall=0

for i in range(time):

    sum_acc+=accuracy[i]

    sum_auc+=aucc[i]

    sum_pre+=pre[i]

    sum_recall+=recall[i]

acc1=sum_acc*1.0/time

auc1=sum_auc*1.0/time

pre1=sum_pre*1.0/time

recall1=sum_recall*1.0/time

print("acc",acc1)

print("auc",auc1)

print("pre",pre1)

print("recall",recall1)

# =============================================================================

#

# data1 = read_csv(root2) #数据转化为数组

# data1=data1.values

#

#

# accuracy=[]

# auc=[]

# pre=[]

# recall=[]

# for i in range(30):

#     train, test= cross_validation.train_test_split(data1, test_size=0.2, random_state=i)

#     test_x=test[:,:-1]

#     test_y=test[:,-1]

#     train_x=train[:,:-1]

#     train_y=train[:,-1]

#     X_Train, Y_Train = SMOTEENN().fit_sample(train_x, train_y)

#

#     #clf = RandomForestClassifier()  #82

#     clf = LogisticRegression()  #82

#     #clf=svm.SVC()

#     #clf= XGBClassifier()

#     #from sklearn.ensemble import RandomForestClassifier  #92%

#     #clf = DecisionTreeClassifier()

#     #clf = GradientBoostingClassifier()

#

#     #clf=neighbors.KNeighborsClassifier()  65.25%

#     #clf=BernoulliNB()

#     clf.fit(X_Train, Y_Train)

#     predict=clf.predict(test_x)

#

#     '''accuracy'''

#     aa=metrics.accuracy_score(test_y, predict)

#     accuracy.append(aa)

#

#     '''auc'''

#     aa=metrics.roc_auc_score(test_y, predict)

#     auc.append(aa)

#

#     '''precision'''

#     aa=metrics.precision_score(test_y, predict,average='weighted')

#     pre.append(aa)

#

#     '''recall'''

#     aa=metrics.recall_score(test_y, predict,average='weighted')

#     recall.append(aa)

#

#

# sum_acc=0

# sum_auc=0

# sum_pre=0

# sum_recall=0

# for i in range(30):

#     sum_acc+=accuracy[i]

#     sum_auc+=auc[i]

#     sum_pre+=pre[i]

#     sum_recall+=recall[i]

#

# acc1=sum_acc*1.0/30

# auc1=sum_auc*1.0/30

# pre1=sum_pre*1.0/30

# recall1=sum_recall*1.0/30

# print("more 的 acc:", acc1)

# print("more 的 auc:", auc1)

# print("more 的 precision:", pre1)

# print("more 的 recall:", recall1)

#

# =============================================================================

    #X_train, X_test, y_train, y_test = cross_validation.train_test_split(X_Train,Y_Train, test_size=0.2, random_state=i)

输出结果：

分类预测输出precision，recall，accuracy，auc和tp，tn，fp，fn矩阵的更多相关文章

目标检测的评价标准mAP, Precision, Recall, Accuracy
目录 metrics 评价方法 TP , FP , TN , FN 概念计算流程 Accuracy , Precision ,Recall Average Precision PR曲线 AP计算 A ...
机器学习：评价分类结果（Precision - Recall 的平衡、P - R 曲线）
一.Precision - Recall 的平衡 1)基础理论调整阈值的大小,可以调节精准率和召回率的比重: 阈值:threshold,分类边界值,score > threshold 时分类为 ...
机器学习基础梳理—(accuracy,precision,recall浅谈)
一.TP TN FP FN TP:标签为正例,预测为正例(P),即预测正确(T) TN:标签为负例,预测为负例(N),即预测正确(T) FP:标签为负例,预测为正例(P),即预测错误(F) FN:标签 ...
Precision,Recall,F1的计算
Precision又叫查准率,Recall又叫查全率.这两个指标共同衡量才能评价模型输出结果. TP: 预测为1(Positive),实际也为1(Truth-预测对了) TN: 预测为0(Negati ...
评价指标整理：Precision, Recall, F-score, TPR, FPR, TNR, FNR, AUC, Accuracy
针对二分类的结果,对模型进行评估,通常有以下几种方法: Precision.Recall.F-score(F1-measure)TPR.FPR.TNR.FNR.AUCAccuracy 真实结果 1 ...
分类指标准确率(Precision)和正确率(Accuracy)的区别
http://www.cnblogs.com/fengfenggirl/p/classification_evaluate.html 一.引言分类算法有很多,不同分类算法又用很多不同的变种.不同的分 ...
Precision/Recall、ROC/AUC、AP/MAP等概念区分
1. Precision和Recall Precision,准确率/查准率.Recall,召回率/查全率.这两个指标分别以两个角度衡量分类系统的准确率. 例如,有一个池塘,里面共有1000条鱼,含10 ...
机器学习--如何理解Accuracy, Precision, Recall, F1 score
当我们在谈论一个模型好坏的时候,我们常常会听到准确率(Accuracy)这个词,我们也会听到＂如何才能使模型的Accurcy更高＂．那么是不是准确率最高的模型就一定是最好的模型? 这篇博文会向大家解释 ...
通过Precision/Recall判断分类结果偏差极大时算法的性能
当我们对某些问题进行分类时,真实结果的分布会有明显偏差. 例如对是否患癌症进行分类,testing set 中可能只有0.5%的人患了癌症. 此时如果直接数误分类数的话,那么一个每次都预测人没有癌症的 ...

随机推荐

操作数据库的时候，使用自带的DbProviderFactory类 (涉及抽象工厂和工厂方法)
微软自带的DbProviderFactory https://msdn.microsoft.com/en-us/library/system.data.common.dbproviderfactory ...
bzoj 1799: [Ahoi2009]self 同类分布数位dp
1799: [Ahoi2009]self 同类分布 Time Limit: 50 Sec Memory Limit: 64 MB[Submit][Status][Discuss] Descripti ...
CentOS6.4x86EngCustomize120g__20160307.rar
安装的镜像包: CentOS-6.4-i386-bin-DVD1to2(CentOS-6.4-i386-bin-DVD1.iso / CentOS-6.4-i386-bin-DVD2.iso) 1. ...
Java回顾之多线程同步
在这篇文章里,我们关注线程同步的话题.这是比多线程更复杂,稍不留意,我们就会“掉到坑里”,而且和单线程程序不同,多线程的错误是否每次都出现,也是不固定的,这给调试也带来了很大的挑战. 在这篇文章里,我 ...
angular的表单
一般来讲表单可能遇到的问题:1.如何数据绑定.2.验证表单.3.显示出错信息.4.整个form的验证.5.避免提交没有验证通过的表单.6.防止多系提交. input属性:nameng-modelng- ...
charles抓取https中出现unknow
http正常抓包,https则出现unknown 1.安装证书 Help->SSL Proxying->Install Charles Root Certificate 但是!!!装完并没 ...
Hadoop异常
org.apache.hadoop.mapred.InvalidInputException: Input path does not exist: hdfs://xxx:49000/user/had ...
bzoj3862
题解: 这一道题目和模板有不同的地方就是在于可以修改只有一条边和i相邻于是我们还要记录与这个点相邻的点有没有改变代码: #pragma GCC optimize(2) #include<bi ...
Use appropriate component from RM component palette
--------------------------- 错误 --------------------------- Opened report contains the following non- ...
单链表是否有环的问题解决与讨论（java实现）
单链表是否有环的问题经常在面试中遇到,一般面试中会要求空间为O(1);再者求若有环,则求环产生时的起始位置. 下面采用java实现. //单链表class ListNode{ int val; Lis ...

分类预测输出precision，recall，accuracy，auc和tp，tn，fp，fn矩阵

分类预测输出precision，recall，accuracy，auc和tp，tn，fp，fn矩阵的更多相关文章

随机推荐

热门专题