初识机器学习 | 2.如何评价模型好坏

最新推荐文章于 2022-12-17 14:30:05 发布

小哲嗨数

最新推荐文章于 2022-12-17 14:30:05 发布

阅读量234

点赞数

分类专栏：【机器学习】

本文链接：https://blog.csdn.net/ganzheyu/article/details/103115622

版权

在这里插入图片描述
下例为乳房癌的概率，通过多个角度评估模型。

import pandas as pd
from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression

# 1)读取数据
cancer = load_breast_cancer()

# 2)数据集划分
x_train, x_test, y_train, y_test = train_test_split(cancer.data, cancer.target)

# 3)特征工程
transformer = StandardScaler()
x_train = transformer.fit_transform(x_train)
x_test = transformer.fit_transform(x_test)

# 4)KNN算法
estimator = LogisticRegression()
estimator.fit(x_train, y_train)

# 5)模型评估
score = estimator.score(x_test, y_test)
print(u"准确率： \n", score)
y_predict = estimator.predict(x_test)

准确率： 
 0.986013986013986


/Users/cleland/.pyenv/versions/3.7.1/envs/base/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.
  FutureWarning)

准确度

accuracy = 正确的数据量 / 样本量

def score(y_predict, y_ture):
    return sum(y_predict == y_test) / y_ture.size

print(u"准确率: ", score(y_predict, y_test))
print(u"sklearn 准确率: ", estimator.score(x_test, y_test))

准确率:  0.986013986013986
sklearn 准确率:  0.986013986013986

混淆矩阵

0 - Negative(阴性，消极)
1 - Postive(阳性，积极)

	预测值_0	预测值_1
真实值_0	TN	FP
真实值_1	FN	TP

def TN(y_predict, y_true):
    return sum((y_predict == 0) & (y_true == 0))

def FN(y_predict, y_true):
    return sum((y_predict == 0) & (y_true == 1))

def FP(y_predict, y_true):
    return sum((y_predict == 1) & (y_true == 0)</