我正在尝试在R中的插入符包中创建预测模型,并从terminal / cmd调用新数据的预测。这是可重复的例子:
# Sonar_training.R
## learning and saving model
library(caret)
library(mlbench)
data(Sonar)
set.seed(107)
inTrain <- createDataPartition(y = Sonar$Class, p = .75,list = FALSE)
training <- Sonar[ inTrain,]
testing <- Sonar[-inTrain,]
saveRDS(testing,"test.rds")
ctrl <- trainControl(method = "repeatedcv",
repeats = 3)
plsFit <- train(Class ~ .,data = training,method = "pls",
tuneLength = 15,
trControl = ctrl,
preProc = c("center", "scale"))
plsClasses <- predict(plsFit, newdata = testing)
saveRDS(plsFit,"fit.rds")
这是由Rscript.exe调用的脚本:
# script.R
##reading model and predict test data
t <- Sys.time()
pls <- readRDS("fit.rds")
testing <- readRDS("test.rds")
head(predict(pls, newdata = testing))
print(Sys.time() - t)
我在终端中使用以下语句运行:
pawel@pawel-MS-1753:~$ Rscript script.R
Loading required package: pls
Attaching package: ‘pls’
The following object is masked from ‘package:stats’:
loadings
[1] M M R M R R
Levels: M R
Time difference of 2.209697 secs
有没有办法更快/更有效率?例如,是否有可能在每次执行时都不加载包?在这种情况下, readRDS 是否正确读取模型?
答案 0 :(得分:1)
您可以尝试使用“profvis”包来分析您的代码:
#library(profvis)
profvis({
for (i in 1:100){
#your code here
}
})
我尝试过,99%的执行时间是培训时间,1%是保存/加载RDS数据,其余成本约为0(加载包,加载数据......):
因此,如果您不想自己优化训练功能,那么您似乎很少有办法缩短执行时间。
答案 1 :(得分:0)
我已经看到这种情况发生在PLS分类模型中,而且我不确定这个问题。但是,请尝试使用AType get(int n) {
Links<AType> current = list;
while (n > 0 && current instanceof Cons) {
current = current.getNext();
n--;
}
return current.getElem();
}
。你会得到大致相同的答案,它应该很快完成。