|
ROOT
Reference Guide |
|
Go to the documentation of this file.
11 #ifndef ROOT_TMVA_Envelope
12 #define ROOT_TMVA_Envelope
112 return fDataLoader->GetDataSetInfo().GetDataSetManager();
TDirectory * RootBaseDir()
Utility method to get base dir directory from current file.
Abstract base class for all high level ml algorithms, you can book ml methods like BDT,...
DataSetManager * GetDataLoaderDataSetManager()
Utility method to get TMVA::DataSetManager pointer from the DataLoader.
void SetModelPersistence(Bool_t status=kTRUE)
Method enable model persistence, then algorithms model is saved in xml or serialized files.
std::vector< OptionMap > fMethods
void SetVerbose(Bool_t status)
Method enable print extra information in the algorithms.
TFile * GetFile()
Method to get the pointer to TFile object.
TString fTransformations
flag for extra information
DataSetInfo & GetDataLoaderDataSetInfo()
Utility method to get TMVA::DataSetInfo reference from the DataLoader.
Class that contains all the data information.
std::vector< OptionMap > & GetMethods()
Method get the Booked methods in a option map object.
DataInputHandler & GetDataLoaderDataInput()
Utility method to get TMVA::DataInputHandler reference from the DataLoader.
Bool_t fVerbose
flag to save the trained model
Bool_t HasMethod(TString methodname, TString methodtitle)
function to check methods booked
Envelope(const TString &name, DataLoader *dataloader=nullptr, TFile *file=nullptr, const TString options="")
timer to measute the time.
Bool_t fSilentFile
List of transformations to test.
virtual void BookMethod(TString methodname, TString methodtitle, TString options="")
Method to book the machine learning method to perform the algorithm.
UInt_t fJobs
procpool object
A ROOT file is a suite of consecutive data records (TKey instances) with a well defined format.
DataLoader * GetDataLoader()
Method to get the pointer to TMVA::DataLoader object.
virtual void ParseOptions()
Method to parse the internal option string.
Bool_t IsModelPersistence()
Method to see if the algorithm model is saved in xml or serialized files.
TStopwatch fTimer
number of jobs to run some high level algorithm in parallel
Class that contains all the data information.
std::shared_ptr< TFile > fFile
data
void SetDataLoader(DataLoader *dalaloader)
Method to set the pointer to TMVA::DataLoader object.
Bool_t IsSilentFile()
Method to see if a file is available to save results.
void WriteDataInformation(TMVA::DataSetInfo &fDataSetInfo, TMVA::Types::EAnalysisType fAnalysisType)
method to save Train/Test information into the output file.
This class provides a simple interface to execute the same task multiple times in parallel,...
virtual void Evaluate()=0
Virtual method to be implemented with your algorithm.
#define ClassDef(name, id)
TProcPool fWorkers
if true dont produce file output
Describe directory structure in memory.
void SetFile(TFile *file)
Method to set the pointer to TFile object, with a writable file.
Bool_t fModelPersistence
file to save the results
~Envelope()
Default destructor.
create variable transformations
Bool_t IsVerbose()
Method to see if the algorithm should print extra information.
std::shared_ptr< DataLoader > fDataLoader
Booked method information.