UZH

Physik-Institut

CMS Wiki Pages

User Tools

Site Tools

Trace:
btag:tmva

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
Next revisionBoth sides next revision
btag:tmva [2014/12/17 14:40] – vlambertbtag:tmva [2014/12/17 14:49] – vlambert
Line 18: Line 18:
 **5)** Create 2D Pt/Eta Histograms for the weighted ntuples with **createEtaPtWeightHists.py** (make sure "weight_norm*weight_category" are set for the weight in Draw() for the histograms). There will be 12 histograms created, 9 for the individual flavour/category files and 3 combined histograms, one for each flavour. **5)** Create 2D Pt/Eta Histograms for the weighted ntuples with **createEtaPtWeightHists.py** (make sure "weight_norm*weight_category" are set for the weight in Draw() for the histograms). There will be 12 histograms created, 9 for the individual flavour/category files and 3 combined histograms, one for each flavour.
  
-**6)** Make the final weighted ntuples making sure that the new Pt/Eta histogram files are pointed to in **addWeightBranch.py**. There should be six new branches created: +**6)** Make the final weighted ntuples making sure that the new Pt/Eta histogram files are pointed to in **addWeightBranch.py**. There should be six new branches created:\\ 
--weight_etaPt   : the Pt/Eta weight, specific for a flavour/category file (for category dedicated training) +-**weight_etaPt**   : the Pt/Eta weight, specific for a flavour/category file (for category dedicated training)\\ 
--weight_etaPtInc: the Pt/Eta weight, inclusive for the flavour +-**weight_etaPtInc**: the Pt/Eta weight, inclusive for the flavour\\ 
--weight_category: the category weight from the evaluation sample +-**weight_category**: the category weight from the evaluation sample\\ 
--weight_norm    : the normalization weight from the training sample +-**weight_norm**    : the normalization weight from the training sample\\ 
--weight_flavour : the ratio of the flavour prevalences in the evaluation process +-**weight_flavour** : the ratio of the flavour prevalences in the evaluation process\\ 
--weight         : (weight_etaPtInc) x (weight_norm x weight_category) x (weight_flavour)  +-**weight**         : (weight_etaPtInc) x (weight_norm x weight_category) x (weight_flavour) //-- this can be used for combined trainings//\\
  
 +The training samples are now ready for the training process with **tmva_training.py**. Make sure to create a directory called "//weights//" to save the output class and xml files from the training. 
  
 === Evaluation Samples === === Evaluation Samples ===
 +**1)** Make the trees really flat without vectors and set variables that are not defined for a given vertex category to a default value. For this, run your ntuples through **createNewTree.py** which will produce sets of new flat ntuples split in event range such as //CombinedSVV2NoVertex_DUSG_0_249999.root// with the shared tree name //"tree"//. 
  
 +**2)** The evaluation trees can be skimmed as well to make the evaluation process faster. The script **skimTT.py** will reference the event ranges in the file names for the flat trees and copy new skimmed trees that contain 10% of the events from each of the flavour/category files. The output will be one combined root file for each flavour/category such as //CombinedSVV2NoVertex_DUSG.root//.
btag/tmva.txt · Last modified: 2014/12/17 14:53 by vlambert

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
CC Attribution-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki