真實(shí)情況	                                  預(yù)測(cè)結(jié)果
            正例	                 反例
          正例	         TP（真正例）	           FN（假反例）
          反例	         FP（假正例）	           TN（真反例）

import numpy
import matplotlib
from matplotlib import pyplot as plt
# true = [真實(shí)組1，真實(shí)組2...真實(shí)組N]，predict = [預(yù)測(cè)組1，預(yù)測(cè)組2...預(yù)測(cè)組N]
def evaluation(true, predict):
    num = len(true)  # 確定有幾組
    (TP, FP, FN, TN) = ([0] * num for i in range(4))  # 賦初值
    for m in range(0, len(true)):
        if (len(true[m]) != len(predict[m])):  # 樣本數(shù)都不等，顯然是有錯(cuò)誤的
            print("真實(shí)結(jié)果與預(yù)測(cè)結(jié)果樣本數(shù)不一致。")
        else:
            for i in range(0, len(true[m])):  # 對(duì)每一組數(shù)據(jù)分別計(jì)數(shù)
                if (predict[m][i] == 1) and ((true[m][i] == 1)):
                    TP[m] += 1.0
                elif (predict[m][i] == 1) and ((true[m][i] == 0)):
                    FP[m] += 1.0
                elif (predict[m][i] == 0) and ((true[m][i] == 1)):
                    FN[m] += 1.0
                elif (predict[m][i] == 0) and ((true[m][i] == 0)):
                    TN[m] += 1.0
    (P, R) = ([0] * num for i in range(2))
    for m in range(0, num):
        if (TP[m] + FP[m] == 0):
            P[m] = 0  # 預(yù)防一些分母為0的情況
        else:
            P[m] = TP[m] / (TP[m] + FP[m])
        if (TP[m] + FN[m] == 0):
            R[m] = 0  # 預(yù)防一些分母為0的情況
        else:
            R[m] = TP[m] / (TP[m] + FN[m])
    plt.title("P-R")
    plt.xlabel("P")
    plt.ylabel("R")
    #plt.plot(P, R)
    #plt.show()
if __name__ == "__main__":
    # 簡(jiǎn)單舉例
    myarray_ture = numpy.random.randint(0, 2, (3, 100))
    myarray_predict = numpy.random.randint(0, 2, (3, 100))
    evaluation(myarray_ture,myarray_predict)

from sklearn import svm, datasets
from sklearn.model_selection import train_test_split
import numpy as np
iris = datasets.load_iris()
# 鳶尾花數(shù)據(jù)導(dǎo)入
x = iris.data
#每一列代表了萼片或花瓣的長(zhǎng)寬，一共4列，每一列代表某個(gè)被測(cè)量的鳶尾植物，iris.shape=(150,4)
y = iris.target
#target是一個(gè)數(shù)組，存儲(chǔ)了data中每條記錄屬于哪一類鳶尾植物，所以數(shù)組的長(zhǎng)度是150,所有不同值只有三個(gè)
random_state = np.random.RandomState(0)
#給定狀態(tài)為0的隨機(jī)數(shù)組
n_samples, n_features = x.shape
x = np.c_[x, random_state.randn(n_samples, 200 * n_features)]
#添加合并生成特征測(cè)試數(shù)據(jù)集
x_train, x_test, y_train, y_test = train_test_split(x[y < 2], y[y < 2],
                                                    test_size=0.25,
                                                    random_state=0)
#根據(jù)此模型訓(xùn)練簡(jiǎn)單數(shù)據(jù)分類器
classifier = svm.LinearSVC(random_state=0)#線性分類支持向量機(jī)
classifier.fit(x_train, y_train)
y_score = classifier.decision_function(x_test)
from sklearn.metrics import precision_recall_curve
import matplotlib.pyplot as plt
precision, recall, _ =precision_recall_curve(y_test, y_score)
plt.fill_between(recall, precision,color='b')
plt.xlabel('Recall')
plt.ylabel('Precision')
plt.ylim([0.0, 1.0])
plt.xlim([0.0, 1.0])
plt.plot(recall, precision)
plt.title("Precision-Recall")
plt.show()