LibraryExecutionError-测试通过Rstudio RRS提要发布的Web服务

Question

我在Rstudio中建立了一个模型，并在Azure ML Studio中使用“ AzureML” R包在Web服务中发布了该模型。在Azure ML Studio上测试Web服务时，遇到错误：

Error: AzureML returns error code:
HTTP status code : 400
AzureML error code : LibraryExecutionError
Module execution encountered an internal library error.
The following error occurred during evaluation of R script: R_tryEval: return error: Error: bad restore file magic number (file may be corrupted) -- no data loaded

您对如何解决此类问题有任何见解？我是否R脚本中缺少一些重要的代码？

我使用的模型是用于预测虹膜数据集上的物种的RandomForest

# Iris dataset
df = iris
set.seed(100);

index = createDataPartition(df$Species, p = 0.7, list = FALSE)
ML.train = df[index,]; 
ML.test = df[-index,];  rm(index)

library(randomForest)
model = randomForest::randomForest(Species ~., data = ML.train)

mypredict = function(newdata) {
      require(randomForest)
      predict(model, newdata, type = "response")
}

# Create workspace
wsObj = AzureML::workspace(id = "my Id", auth = "my token")  # I omitted on purpose my Id and my token values

# Publishing
library(devtools)
library(AzureML)
api = AzureML::publishWebService(ws = wsObj,
                                 fun = mypredict,
                                 name = "IrisWebService",
                                 inputSchema = ML.test %>% select(-Species) )

Answer 1

现在不支持RStudio的AzureML，因为已从2019-07-29的CRAN存储库中删除了该软件包。带有此程序包的Azure ML Studio将不起作用，因为该程序包（AzureML）已删除。

可以在https://cloud.r-project.org/web/packages/azuremlsdk/index.html处从CRAN下载适用于R的Azure机器学习SDK>