评估一个预测模型性能通常都有那些指标

对于不同类型的模型，会有不同的评估指标，那么我们从最直接的回归和分类这两个类型，对于结果连续的回归问题，

一般使用的大致为:MSE(均方差),MAE（绝对平均差）,RMSE(根均方差)这三种评估方法，这三种方式公式此处补贴出来。

对于离散的分类问题，我们一般看ROC曲线，以及AUC曲线，一般好的模型，ROC曲线，在一开始就直接上升到1,然后一直保持1，也就是使得AUC=1.0或者尽可能的让其

接近这个值，这是我们奋斗的目标.

摘个实际的例子：--出自《预测分析核心算法》这本书.

#-*-coding:utf-8-*-
__author__ ='gxjun'
import pandas as pd
import matplotlib.pyplot as plt
from pandas import DataFrame
from random import uniform
import math 
import numpy as np
import random 
import pylab as pl
from sklearn import datasets,linear_model
from sklearn.metrics import roc_curve ,auc


##计算RP值
def confusionMatrix(predicted ,actual , threshold):
    if len(predicted) != len(actual):
        return -1;
    tp=0.0;
    fp=0.0;
    tn=0.0;
    fn=0.0;
    for i in range(len(actual)):
        if actual[i] >0.5:
            if predicted[i] > threshold:
                tp+=1.0;
            else:
                fn+=1.0;
        else:
            if predicted[i]<threshold:
                tn+=1.0;
            else:
                fp+=1.0;
    rtn=[tp,tn,fp,fn];
    return rtn;
target_url =("https://archive.ics.uci.edu/ml/machine-learning-databases/undocumented/connectionist-bench/sonar/sonar.all-data")
data = pd.read_csv(target_url,header=None,prefix='V');
print('-'*80)
print(data.head())
print('-'*80)
print(data.tail())
print('-'*80)
print(data.describe())
print('-'*80)
label = [];
dataRows = [];

for i in range(208):
    if data.iat[i,-1]=='M':
        label.append(1.0);
    else:
        label.append(0);
print label        
dataRows=data.iloc[:,0:-1];
x_train = np.array(dataRows);
y_train = np.array(label);
print "x_train shape: {} , y_train shape: {}".format(x_train.shape,y_train.shape);
print "x_test shape: {} , y_test shape: {}".format(x_test.shape,y_test.shape);
x_test = np.array(dataRows[0:int(208/3)]);
y_test = np.array(label[0:int(208/3)]);
#train model
rockModel = linear_model.LinearRegression();
rockModel.fit(x_train,y_train);
prob = rockModel.predict(x_train);
print('-'*80);
confusionMatrain = confusionMatrix(prob,y_train,threshold=0.5);

#print confusionMatrain
fpr ,tpr,threshold = roc_curve(y_train,prob);
roc_auc = auc(fpr,tpr);

plt.clf();
plt.plot(fpr,tpr,label='ROC curve(area =%0.2f)'%roc_auc);
pl.plot([0,1],[0,1],'k-');
pl.xlim([0.0,1.0]);
pl.ylim([0.0,1.0]);
pl.xlabel("FP rate}");
pl.ylabel("TP rate}");
pl.title("ROC");
pl.legend(loc="lower right");
pl.show()

结果为：