如何在Python中获得RandomForest模型的准确性？

Question

我得到了这个脚本，用RandomForest和LinearRegression预测秒数据集的值。这样可行，线性回归的准确率是18％，太糟糕了。

所以我尝试随机森林，但我不知道如何计算该模型的准确性..

import pandas as pd

from sklearn.ensemble import RandomForestRegressor
from sklearn.datasets import make_regression

import numpy as np
import pandas as pd
import scipy
import matplotlib.pyplot as plt
from pylab import rcParams
import urllib
import sklearn
from sklearn.linear_model import RidgeCV, LinearRegression, Lasso

from sklearn.preprocessing import StandardScaler, PolynomialFeatures
from sklearn.model_selection import GridSearchCV

data = pd.read_csv('EncuestaVieja.csv')
X = data[['Edad','Sexo','v1','v2','v3']]
y = data['Alumna']

dataP = pd.read_csv('EncuestaVieja_test.csv')
X_p = dataP[['Edad','Sexo','v1','v2','v3']]
y_p = dataP['Alumna']

dataT = pd.read_csv('EncuestaVieja_test_2.csv')
X_t = dataT[['Edad','Sexo','v1','v2','v3']]
y_t = dataT['Alumna']
regr = linear_model.LinearRegression()

regr.fit(X, y)

lr = RandomForestRegressor(n_estimators=50)
lr.fit(X, y)

X_test = pd.read_csv('EncuestaNueva.csv')[['Edad','Sexo','v1','v2','v3']]

predictions = regr.predict(X_test)


predictions2 = lr.predict(X_test)
print( 'RandomForest Accuracy: ')
print(((predictions2)))
print( '')
from sklearn.linear_model import LinearRegression
regressor = LinearRegression()
regressor.fit(X_p,y_p)
accuracy = regressor.score(X_t,y_t)
print( 'Linear Regression Accuracy: ', accuracy*100,'%')
print(((predictions)))

OUTPUT：

RandomForest Accuracy: 
[ 1.64  2.54  2.6   2.38  1.64  1.32  1.68  2.56  3.    2.28  2.38  2.68
  2.9   2.5   2.78  1.96  1.56  2.6   2.12  2.76  2.74  1.66  1.68  2.12
  2.3   2.36  2.28  2.28  2.82  1.7   1.86  2.36  1.24]

Linear Regression Accuracy:  18.1336149086 %
[ 1.2681851   1.02802219  3.13377072  2.96885127  2.30808853  1.98814349
  2.39233726  2.8638321   1.86640316  2.63073399  2.21166731  2.25201016
  1.95065189  2.65360517  3.08855254  1.01229733  2.18225606  2.41802534
  2.43539473  2.50227407  1.71105799  1.88238089  2.12152321  3.33525397
  2.72820453  2.43241713  2.88757874  2.6242382   2.63087916  1.98379487
  2.25430306  1.96810279  0.8554685 ]

Answer 1

我认为这是用score（）方法处理的

lr.score(x_test, y_test)

这将返回模型的R ^ 2值。看起来在你的情况下你只有一个x_test。请注意，这不是准确性。回归模型不像分类模型那样使用精确度。而是计算不同的度量，例如均方误差或确定系数。这些指标可以显示预测值与已知值的匹配程度，以及回归模型与回归线的拟合程度。