标签: random-forest

尝试使用如下所示的y数据拟合Random Forest Regressor模型时:

[  0.00000000e+00   1.36094276e+02   4.46608221e+03   8.72660888e+03
   1.31375786e+04   1.73580193e+04   2.29420671e+04   3.12216341e+04
   4.11395711e+04   5.07972062e+04   6.14904935e+04   7.34275322e+04
   7.87333933e+04   8.46302456e+04   9.71074959e+04   1.07146672e+05
   1.17187952e+05   1.26953374e+05   1.37736003e+05   1.47239359e+05
   1.53943242e+05   1.78806710e+05   1.92657725e+05   2.08912711e+05
   2.22855152e+05   2.34532982e+05   2.41391255e+05   2.48699216e+05
   2.62421197e+05   2.79544300e+05   2.95550971e+05   3.13524275e+05
   3.23365158e+05   3.24069067e+05   3.24472999e+05   3.24804951e+05

和X数据看起来像这样:

[ 735233.27082176  735234.27082176  735235.27082176  735236.27082176
  735237.27082176  735238.27082176  735239.27082176  735240.27082176
  735241.27082176  735242.27082176  735243.27082176  735244.27082176
  735245.27082176  735246.27082176  735247.27082176  735248.27082176

使用以下代码:

regressor = RandomForestRegressor(n_estimators=150, min_samples_split=1)
rgr = regressor.fit(X,y) 

我收到此错误:

ValueError: Number of labels=600 does not match number of samples=1

我假设我的一组值的格式错误,但是从文档中我不太清楚.

我在python中使用RandomForestRegressor,我想创建一个图表来说明功能重要性的排名.这是我使用的代码:

from sklearn.ensemble import RandomForestRegressor

MT= pd.read_csv("MT_reduced.csv") 
df = MT.reset_index(drop = False)

columns2 = df.columns.tolist()

# Filter the columns to remove ones we don't want.
columns2 = [c for c in columns2 if c not in["Violent_crime_rate","Change_Property_crime_rate","State","Year"]]

# Store the variable we'll be predicting on.
target = "Property_crime_rate"

# Let’s randomly split our data with 80% as the train set and 20% as the test set:

# Generate the training set.  Set random_state to be able to replicate results.
train2 = …

如何使用randomForest具有观察权重的R 包？我知道这个包里没有这样的选择.我有两个问题:

1. 使用randomForest包有没有解决这个问题的方法？此刻我正在从权重数据中抽取样本,因此我至少可以模拟它:

   m = dim(data)[1]
   sample(data, m, replace=TRUE, prob=weights)
   

   Run Code Online (Sandbox Code Playgroud)

   它有其他(更好的)解决方案吗？

2. 是否有任何替代randomForest方案.我找到了party包(cforest),但它在内存管理方面很糟糕(或者我不能像使用randomForest包那样使用它).我有大约200k观测值和30-40个变量.

编辑:

很抱歉没有澄清细节.我正在使用randomForest包来回归问题(不是分类).这是一个时间序列,每个观察都有它的重量.稍后,此权重用于确定测试观察的模型性能.y变量是连续的.

我在没有找到解决方案的情况下进行了广泛的研究.我已经清理了我的数据集如下:

library("raster")
impute.mean <- function(x) replace(x, is.na(x) | is.nan(x) | is.infinite(x) , 
mean(x, na.rm = TRUE))
losses <- apply(losses, 2, impute.mean)
colSums(is.na(losses))
isinf <- function(x) (NA <- is.infinite(x))
infout <- apply(losses, 2, is.infinite)
colSums(infout)
isnan <- function(x) (NA <- is.nan(x))
nanout <- apply(losses, 2, is.nan)
colSums(nanout)

问题出现了运行预测算法:

options(warn=2)
p  <-   predict(default.rf, losses, type="prob", inf.rm = TRUE, na.rm=TRUE, nan.rm=TRUE)

所有的研究都表明它应该是数据中的NA或Inf或NaN,但我没有发现任何数据.我正在制作数据和randomForest摘要可用于[删除] Traceback的调查并没有显示太多(对我来说):

4: .C("classForest", mdim = as.integer(mdim), ntest = as.integer(ntest), 
       nclass = as.integer(object$forest$nclass), maxcat = as.integer(maxcat), 
       nrnodes = as.integer(nrnodes), jbt = as.integer(ntree), …

我的大小38 MB的训练集(12个属性与420000行).我运行下面的R代码片段,训练使用模型randomForest.这对我来说需要几个小时.

rf.model <- randomForest(
              Weekly_Sales~.,
              data=newdata,
              keep.forest=TRUE,
              importance=TRUE,
              ntree=200,
              do.trace=TRUE,
              na.action=na.roughfix
            )

我认为,由于na.roughfix,它是需要长时间来执行.有这么多的NA's训练集中.

可能有人让我知道我怎么能提高性能？

我的系统配置是:

Intel(R) Core i7 CPU @ 2.90 GHz
RAM - 8 GB
HDD - 500 GB
64 bit OS

我正在尝试提取我使用PySpark训练的随机森林对象的要素重要性.但是,我没有看到在文档中的任何地方执行此操作的示例,也不是RandomForestModel的方法.

如何从RandomForestModelPySpark中的回归器或分类器中提取要素重要性？

以下是文档中提供的示例代码,以帮助我们开始; 但是,没有提到其中的特征重要性.

from pyspark.mllib.tree import RandomForest
from pyspark.mllib.util import MLUtils

# Load and parse the data file into an RDD of LabeledPoint.
data = MLUtils.loadLibSVMFile(sc, 'data/mllib/sample_libsvm_data.txt')
# Split the data into training and test sets (30% held out for testing)
(trainingData, testData) = data.randomSplit([0.7, 0.3])

# Train a RandomForest model.
#  Empty categoricalFeaturesInfo indicates all features are continuous.
#  Note: Use larger numTrees in practice.
#  Setting featureSubsetStrategy="auto" lets the algorithm choose.
model = RandomForest.trainClassifier(trainingData, numClasses=2, …

我只是想做一个简单的RandomForestRegressor示例.但在测试准确性时,我得到了这个错误

/Users/noppanit/anaconda/lib/python2.7/site-packages/sklearn/metrics/classification.pyc

Run Code Online (Sandbox Code Playgroud)

in accuracy_score(y_true,y_pred,normalize,sample_weight)177 178#计算每种可能表示的准确性 - > 179 y_type,y_true,y_pred = _check_targets(y_true,y_pred)180如果y_type.startswith('multilabel'):181 differing_labels = count_nonzero(y_true - y_pred,axis = 1)

/Users/noppanit/anaconda/lib/python2.7/site-packages/sklearn/metrics/classification.pyc

Run Code Online (Sandbox Code Playgroud)

in _check_targets(y_true,y_pred)90 if(y_type不在["binary","multiclass","multilabel-indicator",91"multilabel-sequences"]):---> 92引发ValueError("{0}是不支持".format(y_type)"93 94如果["binary","multiclass"]中的y_type:

ValueError: continuous is not supported

Run Code Online (Sandbox Code Playgroud)

这是数据的样本.我无法显示真实数据.

target, func_1, func_2, func_2, ... func_200
float, float, float, float, ... float

这是我的代码.

import pandas as pd
import numpy as np
from sklearn.preprocessing import Imputer
from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor, ExtraTreesRegressor, GradientBoostingRegressor
from sklearn.cross_validation import train_test_split
from sklearn.metrics import accuracy_score
from sklearn import tree

train = pd.read_csv('data.txt', sep='\t')

labels = …

我第一次使用随机森林进行处理,我遇到了一些我无法弄清楚的问题.当我对所有数据集(大约3000行)进行分析时,我没有收到任何错误消息.但是当我对我的数据集的子集(大约300行)执行相同的分析时,我收到一个错误:

dataset <- read.csv("datasetNA.csv", sep=";", header=T)
names (dataset)
dataset2 <- dataset[complete.cases(dataset$response),]
library(randomForest)
dataset2 <- na.roughfix(dataset2)
data.rforest <- randomForest(dataset2$response ~ dataset2$predictorA + dataset2$predictorB+ dataset2$predictorC + dataset2$predictorD + dataset2$predictorE + dataset2$predictorF + dataset2$predictorG + dataset2$predictorH + dataset2$predictorI, data=dataset2, ntree=100, keep.forest=FALSE, importance=TRUE)

# subset of my original dataset:
groupA<-dataset2[dataset2$order=="groupA",]
data.rforest <- randomForest(groupA$response ~ groupA$predictorA + groupA$predictorB+ groupA$predictorC + groupA$predictorD + groupA$predictorE + groupA$predictorF + groupA$predictorG + groupA$predictorH + groupA$predictorI, data=groupA, ntree=100, keep.forest=FALSE, importance=TRUE)

Error in randomForest.default(m, y, ...) : Can't have empty classes …

R 3.0.0的一个新特性是引入了长向量.但是,.C()和.Fortran()不接受长矢量输入.在R-bloggers上我发现:

这是一种预防措施,因为现有代码不太可能被编写来处理长向量(并且R包装器通常假设长度(x)是整数)

我使用R-package randomForest,这个包显然需要.Fortran(),因为它崩溃了,留下了错误信息

randomForest.default出错:.Fortran不支持long向量(参数20)

如何克服这个问题？我在Windows 7 64位计算机上使用randomForest 4.6-7(在R 3.0.2下构建).

当运行Spark的RandomForest算法时,即使使用相同的种子,我似乎在不同的运行中在树中获得不同的分割.任何人都可以解释我是否做错了(可能),或者实施是错误的(我认为不太可能)？这是我的运行方案:

//read data into rdd
//convert string rdd to LabeledPoint rdd
// train_LP_RDD is RDD of LabeledPoint
// call random forest
val seed = 123417
val numTrees = 10
val numClasses = 2
val categoricalFeaturesInfo: Map[Int, Int] = Map() 
val featureSubsetStrategy = "auto"
val impurity = "gini"
val maxDepth = 8
val maxBins = 10
val rfmodel = RandomForest.trainClassifier(train_LP_RDD, numClasses, categoricalFeaturesInfo,
                        numTrees, featureSubsetStrategy, impurity, maxDepth, maxBins,seed)
println(rfmodel.toDebugString)

在两个不同的运行中,此代码段的输出是不同的.例如,两个结果的差异显示如下:

sdiff -bBWs run1.debug run2.debug

If (feature 2 <= 15.96)             |         If (feature …