两类贝叶斯点机 - 内部系统误差 [英] Two Class Bayes Point Machine - Internal system Error
问题描述
使用两级贝叶斯点机 - 一切都很好 - 然后它开始摔倒了。数据总是相同的布局和格式,只会改变大小。
我做错了什么?
错误日志:
记录从UTC开始02/26/2019 14:09:14:
$
运行工作:" ; / DLL" Microsoft.Analytics.Modules.TrainGenericModel.Dll,版本= 6.0.0.0,文化=中性公钥= 69c3241e6f0468ca; Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel;润QUOT; / Output0" .. \..\Trained model \Trained
model.ilearner" / learner".. \ .. \未经训练的模型\未经训练的model.ilearner" / trainingData" .. \..\Dataset \Dataset.dataset" / labelColumnIndexOrName"%7B%22isFilter%22%3Atrue%2C%22rules%22%3A%5B%7B%22ruleType%22%3A%22ColumnNames%22%2C%22columns%22%3A%5B%22Result%22%5D %2C%22exclude%22%3Afalse%7D%5D%7D" 
/ ContextFile" .. \..\_context\ContextFile.txt""
[Start] Program :: Main
[Start ]鸟;    DataLabModuleDescriptionParser :: ParseModuleDescriptionString
[停止] DataLabModuleDescriptionParser :: ParseModuleDescriptionString。持续时间= 00:00:00.0028180
$
[开始] DllModuleMethod :: DllModuleMethod
[停止] DllModuleMethod :: DllModuleMethod。持续时间= 00:00:00.0000357
[开始] DllModuleMethod ::执行
[开始] DataLabModuleBinder :: BindModuleMethod
[详细] moduleMethodDescription Microsoft.Analytics.Modules.TrainGenericModel.Dll,版本= 6.0.0.0,文化=中性公钥= 69c3241e6f0468ca; Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel;运行
[冗长]鸟;            assemblyFullName Microsoft.Analytics.Modules.TrainGenericModel.Dll,版本= 6.0.0.0,文化=中性公钥= 69c3241e6f0468ca
[开始]鸟;      &NBSP ;     DataLabModuleBinder :: LoadModuleAssembly
[详细] 加载moduleAssembly Microsoft.Analytics.Modules.TrainGenericModel.Dll,版本= 6.0.0.0,文化=中性公钥= 69c3241e6f0468ca
[停止]鸟;             DataLabModuleBinder :: LoadModuleAssembly。持续时间= 00:00:00.0088478
[详细] moduleTypeName Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel
[详细] moduleMethodName运行
[信息] 模块友好名称:列车模型
[信息] 模块发布状态:发布
[停止] DataLabModuleBinder :: BindModuleMethod。持续时间= 00:00:00.0103677
[开始] ParameterArgumentBinder :: InitializeParameterValues
[详细] parameterInfos count = 3
[详细] parameterInfos [0] name = learner,type = Microsoft.Analytics.MachineLearning.ILearner
[开始]   CustomSerializationHandler :: HandleArgumentString
[停止] CustomSerializationHandler :: HandleArgumentString。持续时间= 00:00:00.0369341
[详细] parameterInfos [1] name = trainingData,type = Microsoft.Numerics.Data.Local.DataTable
[开始]    DataTableDatasetHandler :: HandleArgumentString
[停止] DataTableDatasetHandler :: HandleArgumentString。持续时间= 00:00:00.4719581
[详细] parameterInfos [2] name = labelColumnIndexOrName,type = Microsoft.Analytics.Modules.Common.Dll.ColumnSelection
[停止] ParameterArgumentBinder :: InitializeParameterValues。持续时间= 00:00:00.6572974
[详细] 开始调用方法运行...
[停止] DllModuleMethod ::执行。持续时间= 00:00:10.5020130
[严重] 错误:抱歉,您似乎遇到了内部系统错误。请通过浏览器中的完整URL和您遇到故障的时间联系amlforum@microsoft.com。我们可以使用您的
帮助找到此错误并进一步调查。谢谢。
[严重] {" InputParameters":{"数据表":[{"行和QUOT;:58886,"列":11," estimatedSize":34750464," ColumnTypes":{" System.String":6, " System.Int32":2," System.Double":3}," IsComplete":真,"统计与QUOT;:{" 0":[2745,0]," 1 QUOT;:[18.001562340794077 ,13.0,1.0,144.0,16.869850712427727,144.0,0.0]," 2英寸:[45.1818089189281,22.0,0.0,1381.0,70.112990184473176,589.0,0.0]," 3英寸:221,0]," 4英寸:[10.481693441565058,10.0,1.0,40.0,4.1408662904714539,36.0,0.0]," 5英寸:[5,0]," 6英寸:[11.960634106579253,9.9,4.0,34.3,6.2098705498810736,245.0,0.0]," ; 7英寸;:[11,0]," 8英寸;:[1412,0]," 9":[14.274897768569796,8.0,0.0,500.0,20.32413600988907,274.0,0.0]," 10":[2, 0]}}],"学习者":[{" LearnerKind":"学习者"," LearnerSettings":{"名称":" BackwardCompatibleBinaryBayesPointMachineClassifier","我应变":假,"设置":[[" AllowUnknownLevels","真"],[" RandomNumberSeed"," 2342"],[" TrainingIterationCount"," 100"],[" AddBias","真"]],"权重":[]}}],"通用":{" labelColumnIndexOrName":" {\" isFilter\":真,\" ; rules\":[{\" ruleType\":\" ColumnNames\",\" columns\":[\" Result\"] ,\" exclude\":假}]}"}}," OutputParameters":[]," ModuleType":" Microsoft.Analytics.Modules.TrainGenericModel.Dll","的ModuleVersion" :"
版本= 6.0.0.0"," AdditionalModuleInfo":" Microsoft.Analytics.Modules.TrainGenericModel.Dll,版本= 6.0.0.0,文化=中性公钥= 69c3241e6f0468ca; Microsoft.Analytics.Modules.TrainGenericModel .Dll.TrainGenericModel;运行"," Errors":" Microsoft.Analytics.Exceptions.ErrorMapping + ModuleException:
抱歉,您似乎遇到了内部系统错误。请通过浏览器中的完整URL和您遇到故障的时间联系amlforum@microsoft.com。我们可以在您的帮助下找到此错误并进一步调查。谢谢。
---> System.Reflection.TargetInvocationException:调用目标抛出了异常。 ---> MicrosoftResearch.Infer.Factors.ImproperMessageException:推理期间分布不正确(高斯(m / v = -5.798,1 / v = -1.61e-15))。
无法对此型号进行推断。\\\\ n 在MicrosoftResearch.Infer.Factors.IsPositiveOp.XAverageConditional_Helper(Bernoulli isPositive,Gaussian x,Boolean forceProper)\ r \ n 在MicrosoftResearch.Infer.Models.User.CompoundSparseBinaryBpmTraining_EP.Changed_FeatureCount_WeightConstraints_ZeroFeatureValueInstanceCounts_InstanceCount_InstanceFeatureC14(的Int32
numberOfIterations)\r\\\
  在MicrosoftResearch.Infer.Models.User.CompoundSparseBinaryBpmTraining_EP.Execute(Int32 numberOfIterations,Boolean initialise)\\ nn 在MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.InferenceAlgorithms`3.RunTrainingAlgorithm(Int32
iterationCount)\ r \ n 在MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.InferenceAlgorithms`3.TrainInternal(双[] [] featureValues,的Int32 [] [] featureIndexes,的TLabel []标签,的Int32 iterationCount,的Int32 batchNumber)\r\\\
  
。在MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.CompoundBinaryFactorizedInferenceAlgorithms.TrainInternal(双[] [] featureValues,的Int32 [] [] featureIndexes,布尔[]标签,的Int32 iterationCount,的Int32 batchNumber)\r\\\
 &NBSP ;在MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.InferenceAlgorithms`3.Train
(双[] [] featureValues,的Int32 [] [] featureIndexes,的TLabel []标签,的Int32 iterationCount,的Int32 batchNumber)\r\\\
&NBSP ; 在MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.NativeDataFormatBayesPointMachineClassifier`7.TrainOnSingleBatch(TInstanceSource
instanceSource,TLabelSource LABELSOURCE)\r\\\
  在MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.StandardDataFormatBayesPointMachineClassifier`9.TrainIncremental(TInstanceSource instanceSource,TLabelSource LABELSOURCE)\r\\\
  
at Microsoft.Analytics.Modules.BayesPointMachineClassifiers.Dll.BackwardCompatibleBinaryBayesPointMachineClassifier.Train(DataTable instanceSource)\ r \ n 在System.Dynamic.UpdateDelegates.UpdateAndExecuteVoid2 [T0,T1](CallSite站点,T0 arg0,T1 arg1)\\\\ n 在Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel.RunImpl(ILearner学习者,数据表trainingData,ColumnSelection labelColumnIndexOrName)以m
:\\AzureMLVS15-004\\_work\\117\ \\\\\Product \\Source \\Modules \\TrainGenericModel.Dll \\TrainGenericModel.cs:line
53 \\\\ n ---内部异常堆栈跟踪结束--- \\ n at System.RuntimeMethodHandle.InvokeMethod(Object target,Object [] arguments,Signature sig,Boolean constructor)\ r \ n at System.Reflection.RuntimeMethodInfo.UnsafeInvokeInternal(Object
obj,Object [] parameters,Object [] arguments)\ r \ n 在System.Reflection.RuntimeMethodInfo.Invoke(Object obj,BindingFlags invokeAttr,Binder binder,Object []参数,CultureInfo文化)\\ n 在Microsoft.MetaAnalytics.DllModuleHost.DllModuleMethod.Execute(Dictionary`2
moduleDescription)\ r \ n 在Microsoft.MetaAnalytics.DllModuleHost.Program.Main(String [] args)\ r \ n ---内部异常堆栈跟踪结束---",","警告":[],"持续时间":"00:00:10.4980309"}
模块完成后运行时间为00:00:10.8442992,退出代码为-2
模块因负退出代码-2-
为
记录结束于UTC 02/26/2019 14:09:26。
< blockquote>
嗨Mark,
对不起您的体验。您是否使用浏览器中的完整URL和遇到故障的时间联系了 amlforum@microsoft.com?如果没有,请通过azcommunity@microsoft.com与我们联系 ?
请向我们提供您的订阅ID和此主题的URL。
问候,
宇通
Been using Two Class Bayes Point Machine - everything fine - then it started to fall over. The data is always the same layout and format and only changes in size.
Am I doing anything wrong?
Error Log:
Record Starts at UTC 02/26/2019 14:09:14:
Run the job:"/dll "Microsoft.Analytics.Modules.TrainGenericModel.Dll, Version=6.0.0.0, Culture=neutral, PublicKeyToken=69c3241e6f0468ca;Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel;Run" /Output0 "..\..\Trained model\Trained
model.ilearner" /learner "..\..\Untrained model\Untrained model.ilearner" /trainingData "..\..\Dataset\Dataset.dataset" /labelColumnIndexOrName "%7B%22isFilter%22%3Atrue%2C%22rules%22%3A%5B%7B%22ruleType%22%3A%22ColumnNames%22%2C%22columns%22%3A%5B%22Result%22%5D%2C%22exclude%22%3Afalse%7D%5D%7D"
/ContextFile "..\..\_context\ContextFile.txt""
[Start] Program::Main
[Start] DataLabModuleDescriptionParser::ParseModuleDescriptionString
[Stop] DataLabModuleDescriptionParser::ParseModuleDescriptionString. Duration = 00:00:00.0028180
[Start] DllModuleMethod::DllModuleMethod
[Stop] DllModuleMethod::DllModuleMethod. Duration = 00:00:00.0000357
[Start] DllModuleMethod::Execute
[Start] DataLabModuleBinder::BindModuleMethod
[Verbose] moduleMethodDescription Microsoft.Analytics.Modules.TrainGenericModel.Dll, Version=6.0.0.0, Culture=neutral, PublicKeyToken=69c3241e6f0468ca;Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel;Run
[Verbose] assemblyFullName Microsoft.Analytics.Modules.TrainGenericModel.Dll, Version=6.0.0.0, Culture=neutral, PublicKeyToken=69c3241e6f0468ca
[Start] DataLabModuleBinder::LoadModuleAssembly
[Verbose] Loaded moduleAssembly Microsoft.Analytics.Modules.TrainGenericModel.Dll, Version=6.0.0.0, Culture=neutral, PublicKeyToken=69c3241e6f0468ca
[Stop] DataLabModuleBinder::LoadModuleAssembly. Duration = 00:00:00.0088478
[Verbose] moduleTypeName Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel
[Verbose] moduleMethodName Run
[Information] Module FriendlyName : Train Model
[Information] Module Release Status : Release
[Stop] DataLabModuleBinder::BindModuleMethod. Duration = 00:00:00.0103677
[Start] ParameterArgumentBinder::InitializeParameterValues
[Verbose] parameterInfos count = 3
[Verbose] parameterInfos[0] name = learner , type = Microsoft.Analytics.MachineLearning.ILearner
[Start] CustomSerializationHandler::HandleArgumentString
[Stop] CustomSerializationHandler::HandleArgumentString. Duration = 00:00:00.0369341
[Verbose] parameterInfos[1] name = trainingData , type = Microsoft.Numerics.Data.Local.DataTable
[Start] DataTableDatasetHandler::HandleArgumentString
[Stop] DataTableDatasetHandler::HandleArgumentString. Duration = 00:00:00.4719581
[Verbose] parameterInfos[2] name = labelColumnIndexOrName , type = Microsoft.Analytics.Modules.Common.Dll.ColumnSelection
[Stop] ParameterArgumentBinder::InitializeParameterValues. Duration = 00:00:00.6572974
[Verbose] Begin invoking method Run ...
[Stop] DllModuleMethod::Execute. Duration = 00:00:10.5020130
[Critical] Error: Sorry, it seems that you have encountered an internal system error. Please contact amlforum@microsoft.com with the full URL in the browser and the time you experienced the failure. We can locate this error with your
help and investigate further. Thank you.
[Critical] {"InputParameters":{"DataTable":[{"Rows":58886,"Columns":11,"estimatedSize":34750464,"ColumnTypes":{"System.String":6,"System.Int32":2,"System.Double":3},"IsComplete":true,"Statistics":{"0":[2745,0],"1":[18.001562340794077,13.0,1.0,144.0,16.869850712427727,144.0,0.0],"2":[45.1818089189281,22.0,0.0,1381.0,70.112990184473176,589.0,0.0],"3":[221,0],"4":[10.481693441565058,10.0,1.0,40.0,4.1408662904714539,36.0,0.0],"5":[5,0],"6":[11.960634106579253,9.9,4.0,34.3,6.2098705498810736,245.0,0.0],"7":[11,0],"8":[1412,0],"9":[14.274897768569796,8.0,0.0,500.0,20.32413600988907,274.0,0.0],"10":[2,0]}}],"Learner":[{"LearnerKind":"learner","LearnerSettings":{"name":"BackwardCompatibleBinaryBayesPointMachineClassifier","isTrained":false,"settings":[["AllowUnknownLevels","True"],["RandomNumberSeed","2342"],["TrainingIterationCount","100"],["AddBias","True"]],"weights":[]}}],"Generic":{"labelColumnIndexOrName":"{\"isFilter\":true,\"rules\":[{\"ruleType\":\"ColumnNames\",\"columns\":[\"Result\"],\"exclude\":false}]}"}},"OutputParameters":[],"ModuleType":"Microsoft.Analytics.Modules.TrainGenericModel.Dll","ModuleVersion":"
Version=6.0.0.0","AdditionalModuleInfo":"Microsoft.Analytics.Modules.TrainGenericModel.Dll, Version=6.0.0.0, Culture=neutral, PublicKeyToken=69c3241e6f0468ca;Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel;Run","Errors":"Microsoft.Analytics.Exceptions.ErrorMapping+ModuleException:
Sorry, it seems that you have encountered an internal system error. Please contact amlforum@microsoft.com with the full URL in the browser and the time you experienced the failure. We can locate this error with your help and investigate further. Thank you.
---> System.Reflection.TargetInvocationException: Exception has been thrown by the target of an invocation. ---> MicrosoftResearch.Infer.Factors.ImproperMessageException: Improper distribution during inference (Gaussian(m/v=-5.798, 1/v=-1.61e-15)).
Cannot perform inference on this model.\r\n at MicrosoftResearch.Infer.Factors.IsPositiveOp.XAverageConditional_Helper(Bernoulli isPositive, Gaussian x, Boolean forceProper)\r\n at MicrosoftResearch.Infer.Models.User.CompoundSparseBinaryBpmTraining_EP.Changed_FeatureCount_WeightConstraints_ZeroFeatureValueInstanceCounts_InstanceCount_InstanceFeatureC14(Int32
numberOfIterations)\r\n at MicrosoftResearch.Infer.Models.User.CompoundSparseBinaryBpmTraining_EP.Execute(Int32 numberOfIterations, Boolean initialise)\r\n at MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.InferenceAlgorithms`3.RunTrainingAlgorithm(Int32
iterationCount)\r\n at MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.InferenceAlgorithms`3.TrainInternal(Double[][] featureValues, Int32[][] featureIndexes, TLabel[] labels, Int32 iterationCount, Int32 batchNumber)\r\n
at MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.CompoundBinaryFactorizedInferenceAlgorithms.TrainInternal(Double[][] featureValues, Int32[][] featureIndexes, Boolean[] labels, Int32 iterationCount, Int32 batchNumber)\r\n
at MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.InferenceAlgorithms`3.Train(Double[][] featureValues, Int32[][] featureIndexes, TLabel[] labels, Int32 iterationCount, Int32 batchNumber)\r\n at MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.NativeDataFormatBayesPointMachineClassifier`7.TrainOnSingleBatch(TInstanceSource
instanceSource, TLabelSource labelSource)\r\n at MicrosoftResearch.Infer.Learners.BayesPointMachineClassifierInternal.StandardDataFormatBayesPointMachineClassifier`9.TrainIncremental(TInstanceSource instanceSource, TLabelSource labelSource)\r\n
at Microsoft.Analytics.Modules.BayesPointMachineClassifiers.Dll.BackwardCompatibleBinaryBayesPointMachineClassifier.Train(DataTable instanceSource)\r\n at System.Dynamic.UpdateDelegates.UpdateAndExecuteVoid2[T0,T1](CallSite site, T0 arg0, T1 arg1)\r\n
at Microsoft.Analytics.Modules.TrainGenericModel.Dll.TrainGenericModel.RunImpl(ILearner learner, DataTable trainingData, ColumnSelection labelColumnIndexOrName) in m:\\AzureMLVS15-004\\_work\\117\\s\\Product\\Source\\Modules\\TrainGenericModel.Dll\\TrainGenericModel.cs:line
53\r\n --- End of inner exception stack trace ---\r\n at System.RuntimeMethodHandle.InvokeMethod(Object target, Object[] arguments, Signature sig, Boolean constructor)\r\n at System.Reflection.RuntimeMethodInfo.UnsafeInvokeInternal(Object
obj, Object[] parameters, Object[] arguments)\r\n at System.Reflection.RuntimeMethodInfo.Invoke(Object obj, BindingFlags invokeAttr, Binder binder, Object[] parameters, CultureInfo culture)\r\n at Microsoft.MetaAnalytics.DllModuleHost.DllModuleMethod.Execute(Dictionary`2
moduleDescription)\r\n at Microsoft.MetaAnalytics.DllModuleHost.Program.Main(String[] args)\r\n --- End of inner exception stack trace ---","Warnings":[],"Duration":"00:00:10.4980309"}
Module finished after a runtime of 00:00:10.8442992 with exit code -2
Module failed due to negative exit code of -2
Record Ends at UTC 02/26/2019 14:09:26.
Hi Mark,
Sorry for your experience. Have you contacted amlforum@microsoft.com with the full URL in the browser and the time you experienced the failure? If not, please reach out to us at azcommunity@microsoft.com? Please provide us with your subscription ID and the URL for this thread.
Regards,
Yutong
这篇关于两类贝叶斯点机 - 内部系统误差的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!