Difference: JetFlavourTagging (1 vs. 30)

Revision 302017-03-22 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"
Changed:
<
<

Jet Flavour Tagging Howto

>
>

Jet Flavour Tagging Howto with LCFIVertex

 
Changed:
<
<
This is a detailed record on how the Marlin framework and included LCFI packages are used for jet flavour tagging. b-jet flavour tagging is part of our analysis of the feasibility of the ZZ fusion channel with CLIC ILD at 1.4 TeV.
>
>
This is a detailed record on how the Marlin framework and included LCFIVertex packages are used for jet flavour tagging. b-jet flavour tagging is part of our analysis of the feasibility of the ZZ fusion channel with CLIC ILD at 1.4 TeV.

Note: this wiki refers to the LCFIVertex package, that can be found now on GitHub here.

 

Jet Finder and Truth Tagging

Changed:
<
<
We use the LCFI flavour tagging package. This package consists of a topological vertex finder ZVTOP, which reconstructs secondary interactions, and a multivariate classifier which combines several jet-related variables to tag bottom, charm, and light quark jets (see diagram).
>
>
The LCFI package consists of a topological vertex finder ZVTOP, which reconstructs secondary interactions, and a multivariate classifier which combines several jet-related variables to tag bottom, charm, and light quark jets (see diagram).
  Our steering file will contain the jet finder, flavour tagging and LCFI processors, and we will write new slcio files containing the added collections:

Revision 292013-09-13 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 166 to 166
 

Using 4-Jet background

Changed:
<
<
This is done following the same procedure, but using qqll 1.4TeV, CLIC_ILD DST-s (2645) and the script subqqll.sh and its dependencies, zvresQQllPBS.sh, and train_qnets.xml.
>
>
This is done following the same procedure, but using qqll 1.4TeV, CLIC_ILD DST-s (2645) and the script subqqll.sh and its dependencies, zvresQQllPBS.sh, and train_qnets.xml. The relevant files are:

  • qqll_scripts.tgz: Scripts for signal + qqll background, KT algorithm
  • qnets.tgz: Neural nets for signal + qqll background, KT jet finder algorithm

The outputs are stored in the usual directory: /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/*qnets*. Purity vs. efficiency plot using qnets (1 vertex NN is gibberish):
PEcomparisonQQll.png

 

Useful Links

Line: 192 to 199
 
META FILEATTACHMENT attachment="bkg_scripts.tgz" attr="" comment="Scripts for running processors 1-7 on eeqq background files" date="1375803997" name="bkg_scripts.tgz" path="bkg_scripts.tgz" size="8717" stream="bkg_scripts.tgz" tmpFilename="/usr/tmp/CGItemp45761" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="PEcomparisonBNN.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (signal + background, see text)" date="1375804728" name="PEcomparisonBNN.png" path="PEcomparisonBNN.png" size="14806" stream="PEcomparisonBNN.png" tmpFilename="/usr/tmp/CGItemp45679" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="KTscripts.tgz" attr="" comment="Scripts and templates modified to use FastJet/KT algorithm" date="1376298561" name="KTscripts.tgz" path="KTscripts.tgz" size="13823" stream="KTscripts.tgz" tmpFilename="/usr/tmp/CGItemp50170" user="DanProtopopescu" version="2"
Added:
>
>
META FILEATTACHMENT attachment="qqll_scripts.tgz" attr="" comment="Scripts for signal + qqll background, KT algorithm" date="1379084407" name="qqll_scripts.tgz" path="qqll_scripts.tgz" size="14749" stream="qqll_scripts.tgz" tmpFilename="/usr/tmp/CGItemp45933" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="qnets.tgz" attr="" comment="Neural nets for signal + qqll background, KT jet finder algorithm" date="1379084450" name="qnets.tgz" path="qnets.tgz" size="9675" stream="qnets.tgz" tmpFilename="/usr/tmp/CGItemp46040" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="PEcomparisonQQll.png" attr="" comment="Purity vs. efficiency plot using qnets (1 vertex NN is gibberish)" date="1379084640" name="PEcomparisonQQll.png" path="PEcomparisonQQll.png" size="12863" stream="PEcomparisonQQll.png" tmpFilename="/usr/tmp/CGItemp46082" user="DanProtopopescu" version="1"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 282013-09-11 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 166 to 166
 

Using 4-Jet background

Changed:
<
<
This is done following the same procedure, but using qqll 1.4TeV, CLIC_ILD, DST-s (2091) and the script subqqll.sh and its dependencies, zvresQQllPBS.sh, and train_qnets.xml.
>
>
This is done following the same procedure, but using qqll 1.4TeV, CLIC_ILD DST-s (2645) and the script subqqll.sh and its dependencies, zvresQQllPBS.sh, and train_qnets.xml.
 

Useful Links

Revision 272013-09-11 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 166 to 166
 

Using 4-Jet background

Changed:
<
<
This is done following the same procedure, but using qqll 1.4TeV, CLIC_ILD, DST-s (2091) and the script subl.sh and its dependencies.
>
>
This is done following the same procedure, but using qqll 1.4TeV, CLIC_ILD, DST-s (2091) and the script subqqll.sh and its dependencies, zvresQQllPBS.sh, and train_qnets.xml.
 

Useful Links

Revision 262013-09-10 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 166 to 164
  The relevant scripts and XML templates are here. The outputs are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ and have the prefix kt added the file name.
Added:
>
>

Using 4-Jet background

This is done following the same procedure, but using qqll 1.4TeV, CLIC_ILD, DST-s (2091) and the script subl.sh and its dependencies.

 

Useful Links

Revision 252013-08-12 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 172 to 172
 
Deleted:
<
<
 
META FILEATTACHMENT attachment="BRTotalUncertBands_lm.png" attr="" comment="Higgs branching ratios (from A. Denner et al., EPJ C71, p.1753)" date="1372689661" name="BRTotalUncertBands_lm.png" path="BRTotalUncertBands.png" size="114345" stream="BRTotalUncertBands.png" tmpFilename="/usr/tmp/CGItemp36847" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="Timing-ScreenShot.png" attr="" comment="Screen shot: time used by Marlin processors" date="1372694627" name="Timing-ScreenShot.png" path="Timing-ScreenShot.png" size="72589" stream="Timing-ScreenShot.png" tmpFilename="/usr/tmp/CGItemp32197" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="LCFI_Flow_Diagram.pdf" attr="" comment="LCFI processors - flow diagram" date="1372695037" name="LCFI_Flow_Diagram.pdf" path="LCFI Flow Diagram.pdf" size="89775" stream="LCFI Flow Diagram.pdf" tmpFilename="/usr/tmp/CGItemp32291" user="DanProtopopescu" version="1"
Line: 193 to 190
 
META FILEATTACHMENT attachment="bnets.tgz" attr="" comment="Neural nets (Satoru Jet Finder, trained with signal + eeqq background)" date="1375805087" name="bnets.tgz" path="bnets.tgz" size="18627" stream="bnets.tgz" tmpFilename="/usr/tmp/CGItemp48387" user="DanProtopopescu" version="3"
META FILEATTACHMENT attachment="bkg_scripts.tgz" attr="" comment="Scripts for running processors 1-7 on eeqq background files" date="1375803997" name="bkg_scripts.tgz" path="bkg_scripts.tgz" size="8717" stream="bkg_scripts.tgz" tmpFilename="/usr/tmp/CGItemp45761" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="PEcomparisonBNN.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (signal + background, see text)" date="1375804728" name="PEcomparisonBNN.png" path="PEcomparisonBNN.png" size="14806" stream="PEcomparisonBNN.png" tmpFilename="/usr/tmp/CGItemp45679" user="DanProtopopescu" version="1"
Changed:
<
<
META FILEATTACHMENT attachment="KTscripts.tgz" attr="" comment="Scripts and templates modified to use FastJet/KT algorithm" date="1375891378" name="KTscripts.tgz" path="KTscripts.tgz" size="8986" stream="KTscripts.tgz" tmpFilename="/usr/tmp/CGItemp45647" user="DanProtopopescu" version="1"
>
>
META FILEATTACHMENT attachment="KTscripts.tgz" attr="" comment="Scripts and templates modified to use FastJet/KT algorithm" date="1376298561" name="KTscripts.tgz" path="KTscripts.tgz" size="13823" stream="KTscripts.tgz" tmpFilename="/usr/tmp/CGItemp50170" user="DanProtopopescu" version="2"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 242013-08-07 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 149 to 149
  PEcomparisonBNN.png
Added:
>
>

Using an alternate Jet Finder

The whole sequence can be redone with and alternate jet finder. We used '!FastJet' and we replaced the Durham algorithm with KT. For this, we changed Durham_4Jets to Kt_4Jets and used

<processor name="MyFastJetProcessor4" type="FastJetProcessor">
       <parameter name="algorithm" type="StringVec"> kt_algorithm 1.0 </parameter>
       <parameter name="clusteringMode" type="StringVec"> ExclusiveNJets 4 </parameter>
       <parameter name="jetOut" type="string" lcioOutType="ReconstructedParticle"> Kt_4Jets </parameter>
       <parameter name="recParticleIn" type="string" lcioInType="ReconstructedParticle"> LooseSelectedPandoraPFANewPFOs </parameter>
       <parameter name="recombinationScheme" type="string">E_scheme </parameter>
       <parameter name="recParticleOut" type="string" lcioOutType="ReconstructedParticle"> LooseSelectedPandoraPFOsInKt_4Jets </parameter>
</processor>
The relevant scripts and XML templates are here. The outputs are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ and have the prefix kt added the file name.
 

Useful Links

Added:
>
>
 
META FILEATTACHMENT attachment="BRTotalUncertBands_lm.png" attr="" comment="Higgs branching ratios (from A. Denner et al., EPJ C71, p.1753)" date="1372689661" name="BRTotalUncertBands_lm.png" path="BRTotalUncertBands.png" size="114345" stream="BRTotalUncertBands.png" tmpFilename="/usr/tmp/CGItemp36847" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="Timing-ScreenShot.png" attr="" comment="Screen shot: time used by Marlin processors" date="1372694627" name="Timing-ScreenShot.png" path="Timing-ScreenShot.png" size="72589" stream="Timing-ScreenShot.png" tmpFilename="/usr/tmp/CGItemp32197" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="LCFI_Flow_Diagram.pdf" attr="" comment="LCFI processors - flow diagram" date="1372695037" name="LCFI_Flow_Diagram.pdf" path="LCFI Flow Diagram.pdf" size="89775" stream="LCFI Flow Diagram.pdf" tmpFilename="/usr/tmp/CGItemp32291" user="DanProtopopescu" version="1"
Line: 173 to 193
 
META FILEATTACHMENT attachment="bnets.tgz" attr="" comment="Neural nets (Satoru Jet Finder, trained with signal + eeqq background)" date="1375805087" name="bnets.tgz" path="bnets.tgz" size="18627" stream="bnets.tgz" tmpFilename="/usr/tmp/CGItemp48387" user="DanProtopopescu" version="3"
META FILEATTACHMENT attachment="bkg_scripts.tgz" attr="" comment="Scripts for running processors 1-7 on eeqq background files" date="1375803997" name="bkg_scripts.tgz" path="bkg_scripts.tgz" size="8717" stream="bkg_scripts.tgz" tmpFilename="/usr/tmp/CGItemp45761" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="PEcomparisonBNN.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (signal + background, see text)" date="1375804728" name="PEcomparisonBNN.png" path="PEcomparisonBNN.png" size="14806" stream="PEcomparisonBNN.png" tmpFilename="/usr/tmp/CGItemp45679" user="DanProtopopescu" version="1"
Added:
>
>
META FILEATTACHMENT attachment="KTscripts.tgz" attr="" comment="Scripts and templates modified to use FastJet/KT algorithm" date="1375891378" name="KTscripts.tgz" path="KTscripts.tgz" size="8986" stream="KTscripts.tgz" tmpFilename="/usr/tmp/CGItemp45647" user="DanProtopopescu" version="1"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 232013-08-06 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 138 to 138
 
/afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/qq/ILD/DST/00002091/000/
Changed:
<
<
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. For this, we used slightly modified version of the scripts, which can be found here. The resulting outputs, containing approximately 25k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
>
>
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. For this, we used slightly modified versions of the scripts, which can be found here. The resulting outputs, containing approximately 25k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
  PEcomparisonBkg.png

Training the neural nets with eeqq background added

Changed:
<
<
We can train another set of neural nets but adding this time eeqq background to the input. The resulting neural nets (25k signal + 25k background) are here: bnets.tgz.
>
>
We have trained another set of neural nets but adding this time eeqq background to the input. The resulting neural nets (25k signal + 25k background) are here: bnets.tgz.
 
Added:
>
>
PEcomparisonBNN.png
 

Useful Links

Line: 168 to 170
 
META FILEATTACHMENT attachment="PEcomparisonBkg.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (25k eeqq background)" date="1375279055" name="PEcomparisonBkg.png" path="PEcomparisonBkg.png" size="17521" stream="PEcomparisonBkg.png" tmpFilename="/usr/tmp/CGItemp50307" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="xtractPEv.txt" attr="" comment="Script used to extract the numbers from PurityEfficiencyOutput.txt" date="1375112805" name="xtractPEv.txt" path="xtractPEv.txt" size="1642" stream="xtractPEv.txt" tmpFilename="/usr/tmp/CGItemp45803" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="gnets.tgz" attr="" comment="Neural nets (Satoru Jet Finder, trained with signal only)" date="1375797750" name="gnets.tgz" path="gnets.tgz" size="11972" stream="gnets.tgz" tmpFilename="/usr/tmp/CGItemp45656" user="DanProtopopescu" version="1"
Changed:
<
<
META FILEATTACHMENT attachment="bnets.tgz" attr="" comment="Neural nets (Satoru Jet Finder, trained with signal + eeqq background)" date="1375800358" name="bnets.tgz" path="bnets.tgz" size="12491" stream="bnets.tgz" tmpFilename="/usr/tmp/CGItemp45654" user="DanProtopopescu" version="1"
>
>
META FILEATTACHMENT attachment="bnets.tgz" attr="" comment="Neural nets (Satoru Jet Finder, trained with signal + eeqq background)" date="1375805087" name="bnets.tgz" path="bnets.tgz" size="18627" stream="bnets.tgz" tmpFilename="/usr/tmp/CGItemp48387" user="DanProtopopescu" version="3"
META FILEATTACHMENT attachment="bkg_scripts.tgz" attr="" comment="Scripts for running processors 1-7 on eeqq background files" date="1375803997" name="bkg_scripts.tgz" path="bkg_scripts.tgz" size="8717" stream="bkg_scripts.tgz" tmpFilename="/usr/tmp/CGItemp45761" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="PEcomparisonBNN.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (signal + background, see text)" date="1375804728" name="PEcomparisonBNN.png" path="PEcomparisonBNN.png" size="14806" stream="PEcomparisonBNN.png" tmpFilename="/usr/tmp/CGItemp45679" user="DanProtopopescu" version="1"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 222013-08-06 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 77 to 77
 
<-- Jet theta angle cut (only the Glasgow NN trainer has this option) -->
24.
Changed:
<
<
The neural nets are saved as XML files in gnets/ and will be used for flavour tagging (next step). No slcio output is written at this time.
>
>
The neural nets are saved as XML files in gnets/ and will be used for flavour tagging (next step). No slcio output is written at this time. These neural networks can be downloaded from here: gnets.tgz
 

Flavour Tagging

Line: 129 to 129
 For b-tagging, we've compared the plot produced by MakePurityVsEfficiencyRootPlotGla.C and graphs drawn with the data tabulated in PurityEfficiencyOutput.txt, for 1, 2, 3 (corresponding to distinct neural networks) or any number of vertices (which we don't know yet how to interpret):
PEcomparison.png
Changed:
<
<
Here's the script used to extract the numbers from PurityEfficiencyOutput.txt.
>
>
Here's the script used to extract the numbers from PurityEfficiencyOutput.txt
 

Adding background

Line: 142 to 142
  PEcomparisonBkg.png
Added:
>
>

Training the neural nets with eeqq background added

We can train another set of neural nets but adding this time eeqq background to the input. The resulting neural nets (25k signal + 25k background) are here: bnets.tgz.

 

Useful Links

Line: 161 to 167
 
META FILEATTACHMENT attachment="PEcomparison.png" attr="" comment="B-tagging Purity vs. Efficiency comparison" date="1374664155" name="PEcomparison.png" path="PEcomparison.png" size="12262" stream="PEcomparison.png" tmpFilename="/usr/tmp/CGItemp40207" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="PEcomparisonBkg.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (25k eeqq background)" date="1375279055" name="PEcomparisonBkg.png" path="PEcomparisonBkg.png" size="17521" stream="PEcomparisonBkg.png" tmpFilename="/usr/tmp/CGItemp50307" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="xtractPEv.txt" attr="" comment="Script used to extract the numbers from PurityEfficiencyOutput.txt" date="1375112805" name="xtractPEv.txt" path="xtractPEv.txt" size="1642" stream="xtractPEv.txt" tmpFilename="/usr/tmp/CGItemp45803" user="DanProtopopescu" version="1"
Added:
>
>
META FILEATTACHMENT attachment="gnets.tgz" attr="" comment="Neural nets (Satoru Jet Finder, trained with signal only)" date="1375797750" name="gnets.tgz" path="gnets.tgz" size="11972" stream="gnets.tgz" tmpFilename="/usr/tmp/CGItemp45656" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="bnets.tgz" attr="" comment="Neural nets (Satoru Jet Finder, trained with signal + eeqq background)" date="1375800358" name="bnets.tgz" path="bnets.tgz" size="12491" stream="bnets.tgz" tmpFilename="/usr/tmp/CGItemp45654" user="DanProtopopescu" version="1"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 212013-07-31 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 138 to 138
 
/afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/qq/ILD/DST/00002091/000/
Changed:
<
<
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. For this, we used slightly modified version of the scripts, which can be found here. The resulting outputs, containing approximately 18k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
>
>
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. For this, we used slightly modified version of the scripts, which can be found here. The resulting outputs, containing approximately 25k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
  PEcomparisonBkg.png
Line: 159 to 159
 
META FILEATTACHMENT attachment="LCFI_Hillert.pdf" attr="" comment="LCFI Vertex package talk by S. Hillert" date="1374661268" name="LCFI_Hillert.pdf" path="LCFI_Hillert.pdf" size="806747" stream="LCFI_Hillert.pdf" tmpFilename="/usr/tmp/CGItemp36980" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="LCFI_ecfa_krakow.pdf" attr="" comment="Flavour tagging at the linear collider - talk by M. Wing" date="1374661320" name="LCFI_ecfa_krakow.pdf" path="LCFI_ecfa_krakow.pdf" size="303536" stream="LCFI_ecfa_krakow.pdf" tmpFilename="/usr/tmp/CGItemp36921" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="PEcomparison.png" attr="" comment="B-tagging Purity vs. Efficiency comparison" date="1374664155" name="PEcomparison.png" path="PEcomparison.png" size="12262" stream="PEcomparison.png" tmpFilename="/usr/tmp/CGItemp40207" user="DanProtopopescu" version="2"
Changed:
<
<
META FILEATTACHMENT attachment="PEcomparisonBkg.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (18k eeqq background)" date="1375112388" name="PEcomparisonBkg.png" path="PEcomparisonBkg.png" size="17496" stream="PEcomparisonBkg.png" tmpFilename="/usr/tmp/CGItemp49949" user="DanProtopopescu" version="1"
>
>
META FILEATTACHMENT attachment="PEcomparisonBkg.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (25k eeqq background)" date="1375279055" name="PEcomparisonBkg.png" path="PEcomparisonBkg.png" size="17521" stream="PEcomparisonBkg.png" tmpFilename="/usr/tmp/CGItemp50307" user="DanProtopopescu" version="2"
 
META FILEATTACHMENT attachment="xtractPEv.txt" attr="" comment="Script used to extract the numbers from PurityEfficiencyOutput.txt" date="1375112805" name="xtractPEv.txt" path="xtractPEv.txt" size="1642" stream="xtractPEv.txt" tmpFilename="/usr/tmp/CGItemp45803" user="DanProtopopescu" version="1"
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 202013-07-29 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 129 to 129
 For b-tagging, we've compared the plot produced by MakePurityVsEfficiencyRootPlotGla.C and graphs drawn with the data tabulated in PurityEfficiencyOutput.txt, for 1, 2, 3 (corresponding to distinct neural networks) or any number of vertices (which we don't know yet how to interpret):
PEcomparison.png

Added:
>
>
Here's the script used to extract the numbers from PurityEfficiencyOutput.txt.
 

Adding background

Line: 136 to 138
 
/afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/qq/ILD/DST/00002091/000/
Changed:
<
<
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. For this, we used slightly modified version of the scripts, which can be found here. The resulting outputs, containing approximately 25k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
>
>
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. For this, we used slightly modified version of the scripts, which can be found here. The resulting outputs, containing approximately 18k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
 
Added:
>
>
PEcomparisonBkg.png
 

Useful Links

Line: 156 to 159
 
META FILEATTACHMENT attachment="LCFI_Hillert.pdf" attr="" comment="LCFI Vertex package talk by S. Hillert" date="1374661268" name="LCFI_Hillert.pdf" path="LCFI_Hillert.pdf" size="806747" stream="LCFI_Hillert.pdf" tmpFilename="/usr/tmp/CGItemp36980" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="LCFI_ecfa_krakow.pdf" attr="" comment="Flavour tagging at the linear collider - talk by M. Wing" date="1374661320" name="LCFI_ecfa_krakow.pdf" path="LCFI_ecfa_krakow.pdf" size="303536" stream="LCFI_ecfa_krakow.pdf" tmpFilename="/usr/tmp/CGItemp36921" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="PEcomparison.png" attr="" comment="B-tagging Purity vs. Efficiency comparison" date="1374664155" name="PEcomparison.png" path="PEcomparison.png" size="12262" stream="PEcomparison.png" tmpFilename="/usr/tmp/CGItemp40207" user="DanProtopopescu" version="2"
Added:
>
>
META FILEATTACHMENT attachment="PEcomparisonBkg.png" attr="" comment="B-tagging Purity vs. Efficiency comparison (18k eeqq background)" date="1375112388" name="PEcomparisonBkg.png" path="PEcomparisonBkg.png" size="17496" stream="PEcomparisonBkg.png" tmpFilename="/usr/tmp/CGItemp49949" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="xtractPEv.txt" attr="" comment="Script used to extract the numbers from PurityEfficiencyOutput.txt" date="1375112805" name="xtractPEv.txt" path="xtractPEv.txt" size="1642" stream="xtractPEv.txt" tmpFilename="/usr/tmp/CGItemp45803" user="DanProtopopescu" version="1"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 192013-07-26 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 136 to 136
 
/afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/qq/ILD/DST/00002091/000/
Changed:
<
<
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. The resulting outputs, containing approximately 25k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
>
>
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. For this, we used slightly modified version of the scripts, which can be found here. The resulting outputs, containing approximately 25k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
 

Useful Links

Revision 182013-07-26 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 50 to 50
  The processors listed above could be run in sequence, or split in several steps, invoking a LCIOOutput processor to write intermediate slcio outputs at every step. Here's a script for that, where the intermediate xml files are slight modifications of the files provided in LCFIVertex/steering examples. We found that the most time-consuming processor is ZVTOP_ZVRES with more than 10 s/event.
Changed:
<
<
We found it easier to run processors 1 to 6 on batches of 10 input files, and save the outputs as zvresX_out.slcio. Then we ran processors 7 and 8 on these files to produce a new set ftiX_out.slcio. We used all these to train the neural net, but then again was easier to run flavour tagging (see below) on the individual ftiX_out.slcio files. Again we used all files for the purity vs. efficiency plots (see below).
>
>
We found it easier to run processors 1 to 6 on batches of 10 input files, and save the outputs as /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/runX_out.slcio. Then we ran processors 7 and 8 on these files to produce a new set ftiX_out.slcio. We used half of these files to train the neural net, but then again was easier to run flavour tagging (see below) on the individual ftiX_out.slcio files. We used the other half of the files files for the purity vs. efficiency plots (see below). These slcio output files were stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
  Troubleshooting: The b3_D0CutValue parameter of the IPRPCutProcessor was set to 5O instead of 50, and was causing a crash. For the ZVRESRPCut processor, h1_MCPIDEnable had to be set to false. See also this post.

Neural Network Training

Changed:
<
<
In the previous step, we have extracted the discriminating parameters and truth-tagged the jets. The slcio files created contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use a customised version of the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
>
>
In the previous step, we have extracted the discriminating parameters and truth-tagged the jets. The slcio files created are stored in:
/afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/
These files contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use a customised version of the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
 
  <processor name="MyNeuralNetTrainer" type="GlasgowNeuralNetTrainer"/>
Line: 126 to 130
 PEcomparison.png
Added:
>
>

Adding background

Up to now all that input slcio files contained was signal, so the 'background' in the plots above is not really background. To add background, we've downloaded 1.4!TeV eeqq files to:

/afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/qq/ILD/DST/00002091/000/
We had to run processors 1-7 on these files to identify jets and calculate the discriminating parameters. The resulting outputs, containing approximately 25k events, are stored in /afs/phas.gla.ac.uk/data/ilc/datasets01/1.4tev/ZVRES_out/ as well.
 

Useful Links

Revision 172013-07-24 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 122 to 122
  The cut values and corresponding purity and efficiencies are tabulated in the file PurityEfficiencyOutput.txt. We used another Root macro to plot these values and determine the optimum b-cut.
Changed:
<
<
For b-tagging, we've compared the plot produced by MakePurityVsEfficiencyRootPlotGla.C and graphs drawn with the data tabulated in PurityEfficiencyOutput.txt, for 1, 2, 3 or any number of vertices:
>
>
For b-tagging, we've compared the plot produced by MakePurityVsEfficiencyRootPlotGla.C and graphs drawn with the data tabulated in PurityEfficiencyOutput.txt, for 1, 2, 3 (corresponding to distinct neural networks) or any number of vertices (which we don't know yet how to interpret):
 PEcomparison.png
Added:
>
>
 

Useful Links

Revision 162013-07-24 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 115 to 115
  Once the AIDA Plots processors are run via Marlin, a text file PurityEfficiencyOutput.txt and a RAIDA root file are produced. We customised the MakePurityVsEfficiencyRootPlot.C macro and ran it to use the RAIDA file as input to produce the purity vs. efficiency plots:
Changed:
<
<
root -l bbPurityVsEfficiencyRootPlotGla.C
>
>
root -l MakePurityVsEfficiencyRootPlotGla.C
  Here's a plot of our flavour tagging purity vs. efficiencies (using cca. 25k events):
Screen_Shot_2013-07-16_at_13.40.40.png
Changed:
<
<
The cut values and corresponding purity and efficiencies are tabulated in the file PurityEfficiencyOutput.txt. From these values, one can calculate and optimum b-cut.
>
>
The cut values and corresponding purity and efficiencies are tabulated in the file PurityEfficiencyOutput.txt. We used another Root macro to plot these values and determine the optimum b-cut.

For b-tagging, we've compared the plot produced by MakePurityVsEfficiencyRootPlotGla.C and graphs drawn with the data tabulated in PurityEfficiencyOutput.txt, for 1, 2, 3 or any number of vertices:
PEcomparison.png

Useful Links

 
META FILEATTACHMENT attachment="BRTotalUncertBands_lm.png" attr="" comment="Higgs branching ratios (from A. Denner et al., EPJ C71, p.1753)" date="1372689661" name="BRTotalUncertBands_lm.png" path="BRTotalUncertBands.png" size="114345" stream="BRTotalUncertBands.png" tmpFilename="/usr/tmp/CGItemp36847" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="Timing-ScreenShot.png" attr="" comment="Screen shot: time used by Marlin processors" date="1372694627" name="Timing-ScreenShot.png" path="Timing-ScreenShot.png" size="72589" stream="Timing-ScreenShot.png" tmpFilename="/usr/tmp/CGItemp32197" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="LCFI_Flow_Diagram.pdf" attr="" comment="LCFI processors - flow diagram" date="1372695037" name="LCFI_Flow_Diagram.pdf" path="LCFI Flow Diagram.pdf" size="89775" stream="LCFI Flow Diagram.pdf" tmpFilename="/usr/tmp/CGItemp32291" user="DanProtopopescu" version="1"
Changed:
<
<
META FILEATTACHMENT attachment="Vertexing_Howto.pdf" attr="" comment="Vertexing HowTo (Ben Jeffery)" date="1372697529" name="Vertexing_Howto.pdf" path="Vertexing Howto.pdf" size="761330" stream="Vertexing Howto.pdf" tmpFilename="/usr/tmp/CGItemp32366" user="DanProtopopescu" version="1"
>
>
META FILEATTACHMENT attachment="Vertexing_Howto.pdf" attr="" comment="Vertexing HowTo talk by B. Jeffery" date="1372697529" name="Vertexing_Howto.pdf" path="Vertexing Howto.pdf" size="761330" stream="Vertexing Howto.pdf" tmpFilename="/usr/tmp/CGItemp32366" user="DanProtopopescu" version="1"
 
META FILEATTACHMENT attachment="jet_truth_tag-steer.xml" attr="" comment="Steering file 1: jet finder and truth flavour tagging" date="1372699245" name="jet_truth_tag-steer.xml" path="jet_truth_tag-steer.xml" size="84107" stream="jet_truth_tag-steer.xml" tmpFilename="/usr/tmp/CGItemp32410" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="runLCFI.txt" attr="" comment="Script to run LCFI sequence of processors" date="1372954507" name="runLCFI.txt" path="runLCFI.txt" size="1170" stream="runLCFI.txt" tmpFilename="/usr/tmp/CGItemp32298" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="Screen_Shot_2013-07-16_at_13.40.40.png" attr="" comment="Flavour tagging purity vs. efficiency plots (using cca. 22k events)" date="1373979756" name="Screen_Shot_2013-07-16_at_13.40.40.png" path="Screen Shot 2013-07-16 at 13.40.40.png" size="22046" stream="Screen Shot 2013-07-16 at 13.40.40.png" tmpFilename="/usr/tmp/CGItemp35477" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="zvresRun.sh" attr="" comment="Script to run processors 1-7, see text" date="1373983508" name="zvresRun.sh" path="zvresRun.sh" size="1734" stream="zvresRun.sh" tmpFilename="/usr/tmp/CGItemp35271" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="MakePurityVsEfficiencyRootPlotGla.C" attr="" comment="Customised macro to draw purity vs. efficiency plots" date="1373983605" name="MakePurityVsEfficiencyRootPlotGla.C" path="MakePurityVsEfficiencyRootPlotGla.C" size="2527" stream="MakePurityVsEfficiencyRootPlotGla.C" tmpFilename="/usr/tmp/CGItemp35411" user="DanProtopopescu" version="1"
Added:
>
>
META FILEATTACHMENT attachment="LCFI_Hillert.pdf" attr="" comment="LCFI Vertex package talk by S. Hillert" date="1374661268" name="LCFI_Hillert.pdf" path="LCFI_Hillert.pdf" size="806747" stream="LCFI_Hillert.pdf" tmpFilename="/usr/tmp/CGItemp36980" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="LCFI_ecfa_krakow.pdf" attr="" comment="Flavour tagging at the linear collider - talk by M. Wing" date="1374661320" name="LCFI_ecfa_krakow.pdf" path="LCFI_ecfa_krakow.pdf" size="303536" stream="LCFI_ecfa_krakow.pdf" tmpFilename="/usr/tmp/CGItemp36921" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="PEcomparison.png" attr="" comment="B-tagging Purity vs. Efficiency comparison" date="1374664155" name="PEcomparison.png" path="PEcomparison.png" size="12262" stream="PEcomparison.png" tmpFilename="/usr/tmp/CGItemp40207" user="DanProtopopescu" version="2"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 152013-07-22 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 113 to 113
  to the LCFIVertex/CMakeLists.txt file, sourced the root environment, then ran cmake and make install.
Changed:
<
<
Once the AIDA Plots processors are run via Marlin, a RAIDA root file is produced. We customised the MakePurityVsEfficiencyRootPlot.C macro and ran it to use this RAIDA file as input to produce the purity vs. efficiency plots:
>
>
Once the AIDA Plots processors are run via Marlin, a text file PurityEfficiencyOutput.txt and a RAIDA root file are produced. We customised the MakePurityVsEfficiencyRootPlot.C macro and ran it to use the RAIDA file as input to produce the purity vs. efficiency plots:
 
root -l bbPurityVsEfficiencyRootPlotGla.C
Changed:
<
<
Here's a plot of our flavour tagging purity vs. efficiencies (using cca. 22k events):
>
>
Here's a plot of our flavour tagging purity vs. efficiencies (using cca. 25k events):
 Screen_Shot_2013-07-16_at_13.40.40.png
Added:
>
>
The cut values and corresponding purity and efficiencies are tabulated in the file PurityEfficiencyOutput.txt. From these values, one can calculate and optimum b-cut.
 
META FILEATTACHMENT attachment="BRTotalUncertBands_lm.png" attr="" comment="Higgs branching ratios (from A. Denner et al., EPJ C71, p.1753)" date="1372689661" name="BRTotalUncertBands_lm.png" path="BRTotalUncertBands.png" size="114345" stream="BRTotalUncertBands.png" tmpFilename="/usr/tmp/CGItemp36847" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="Timing-ScreenShot.png" attr="" comment="Screen shot: time used by Marlin processors" date="1372694627" name="Timing-ScreenShot.png" path="Timing-ScreenShot.png" size="72589" stream="Timing-ScreenShot.png" tmpFilename="/usr/tmp/CGItemp32197" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="LCFI_Flow_Diagram.pdf" attr="" comment="LCFI processors - flow diagram" date="1372695037" name="LCFI_Flow_Diagram.pdf" path="LCFI Flow Diagram.pdf" size="89775" stream="LCFI Flow Diagram.pdf" tmpFilename="/usr/tmp/CGItemp32291" user="DanProtopopescu" version="1"

Revision 142013-07-22 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 56 to 56
 

Neural Network Training

Changed:
<
<
The slcio files created at the previous step contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use a customised version of the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
>
>
In the previous step, we have extracted the discriminating parameters and truth-tagged the jets. The slcio files created contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use a customised version of the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
 
  <processor name="MyNeuralNetTrainer" type="GlasgowNeuralNetTrainer"/>

Revision 132013-07-16 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 127 to 127
 
META FILEATTACHMENT attachment="jet_truth_tag-steer.xml" attr="" comment="Steering file 1: jet finder and truth flavour tagging" date="1372699245" name="jet_truth_tag-steer.xml" path="jet_truth_tag-steer.xml" size="84107" stream="jet_truth_tag-steer.xml" tmpFilename="/usr/tmp/CGItemp32410" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="runLCFI.txt" attr="" comment="Script to run LCFI sequence of processors" date="1372954507" name="runLCFI.txt" path="runLCFI.txt" size="1170" stream="runLCFI.txt" tmpFilename="/usr/tmp/CGItemp32298" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="Screen_Shot_2013-07-16_at_13.40.40.png" attr="" comment="Flavour tagging purity vs. efficiency plots (using cca. 22k events)" date="1373979756" name="Screen_Shot_2013-07-16_at_13.40.40.png" path="Screen Shot 2013-07-16 at 13.40.40.png" size="22046" stream="Screen Shot 2013-07-16 at 13.40.40.png" tmpFilename="/usr/tmp/CGItemp35477" user="DanProtopopescu" version="1"
Added:
>
>
META FILEATTACHMENT attachment="zvresRun.sh" attr="" comment="Script to run processors 1-7, see text" date="1373983508" name="zvresRun.sh" path="zvresRun.sh" size="1734" stream="zvresRun.sh" tmpFilename="/usr/tmp/CGItemp35271" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="MakePurityVsEfficiencyRootPlotGla.C" attr="" comment="Customised macro to draw purity vs. efficiency plots" date="1373983605" name="MakePurityVsEfficiencyRootPlotGla.C" path="MakePurityVsEfficiencyRootPlotGla.C" size="2527" stream="MakePurityVsEfficiencyRootPlotGla.C" tmpFilename="/usr/tmp/CGItemp35411" user="DanProtopopescu" version="1"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 122013-07-16 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 23 to 23
 
Changed:
<
<
The JetFinder processor reconstructs 2 and 4 jets events from the input collection (LooseSelectedPandoraPFANewPFOs was used). For the reconstructed 4 jets, MyTrueAngularJetFlavourProcessor determines MC Jet Flavour by angular matching of heavy quarks to jets, and also determines hadronic and partonic charge of the jet.
>
>
The JetFinder processor reconstructs 2 and 4 jets events from the input collection (LooseSelectedPandoraPFANewPFOs was used). For the reconstructed 4 jets, MyTrueAngularJetFlavourProcessor determines MC Jet Flavour by angular matching of heavy quarks to jets, and also determines hadronic and partonic charge of the jet. The LCIOOutput processor creates new slcio files containing the new collections added by the above processors.
  The LCFI processors have the following functions:
Line: 50 to 50
  The processors listed above could be run in sequence, or split in several steps, invoking a LCIOOutput processor to write intermediate slcio outputs at every step. Here's a script for that, where the intermediate xml files are slight modifications of the files provided in LCFIVertex/steering examples. We found that the most time-consuming processor is ZVTOP_ZVRES with more than 10 s/event.
Changed:
<
<
The LCIOOutput processor creates new slcio files containing the new collections added by the above processors.
>
>
We found it easier to run processors 1 to 6 on batches of 10 input files, and save the outputs as zvresX_out.slcio. Then we ran processors 7 and 8 on these files to produce a new set ftiX_out.slcio. We used all these to train the neural net, but then again was easier to run flavour tagging (see below) on the individual ftiX_out.slcio files. Again we used all files for the purity vs. efficiency plots (see below).
  Troubleshooting: The b3_D0CutValue parameter of the IPRPCutProcessor was set to 5O instead of 50, and was causing a crash. For the ZVRESRPCut processor, h1_MCPIDEnable had to be set to false. See also this post.
Line: 54 to 54
  Troubleshooting: The b3_D0CutValue parameter of the IPRPCutProcessor was set to 5O instead of 50, and was causing a crash. For the ZVRESRPCut processor, h1_MCPIDEnable had to be set to false. See also this post.
Deleted:
<
<
 

Neural Network Training

Changed:
<
<
The slcio files created at the previous step contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
>
>
The slcio files created at the previous step contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use a customised version of the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
 
Changed:
<
<
>
>
 
Changed:
<
<
The neural nets are saved as XML files in nnets/ and will be used for flavour tagging (next step). No slcio output is written at this time.
>
>
This processor is a slightly modified version of the NeuralNetTrainer included in the LCFI package, where the polar angle cut was introduced as a steering parameter:
// Theta cut parameters: _jetThetaAngleCut < theta < (180 - _jetThetaAngleCut)
        registerProcessorParameter( "JetThetaAngleCut" ,
                                    "Cut on the jets theta angle"  ,
                                    _jetThetaAngleCut,
                                    (float)15.);
such that we can pass a cut angle different of the 30 default:
<!-- Jet theta angle cut (only the Glasgow NN trainer has this option) -->
                <parameter name="JetThetaAngleCut" type="float"> 24. </parameter>
The neural nets are saved as XML files in gnets/ and will be used for flavour tagging (next step). No slcio output is written at this time.
 

Flavour Tagging

Changed:
<
<
Now we are ready to employ the FlavourTag processor, which will do flavour tagging using the neural nets trained in the previous step. The input slcio file contains the FlavourTagInputs and FTSelectedJets (or Durham_4Jets, not sure if there's a difference at this level) collections.
>
>
Now we are ready to employ the FlavourTag processor, which will do flavour tagging using the neural nets trained in the previous step. The input slcio files contain the FlavourTagInputs and FTSelectedJets (or Durham_4Jets, not sure if there's a difference at this level) collections.
 
  <processor name="MyFlavourTag"/>
  <processor name="MyLCIOOutputProcessor"/>  
Changed:
<
<
The output slcio will contain the collection FlavourTag which will be used for our ZZFusion analysis.
>
>
The output slcio will contain a new collection FlavourTag (or FlavourTagGla in our customised configuration).
 

Purity and Efficiency Studies

Line: 86 to 95
 
Changed:
<
<
We had to provide MyPlot with the actual name of the TrueJetFlavourCollection:
>
>
We had to provide MyPlot with the actual names of our collections:
 
<parameter name="TrueJetFlavourCollection" type="string">TrueJetFlavour_4Jets </parameter>
Added:
>
>
<--In fti-steer.xml this parameter is called "FlavourTagCollection", without the 's' -->
FlavourTagGla
  Note that LCFI must be compiled with ROOT if one wants .root output from PlotProcessor (instead of .txt). For this, we added
Line: 104 to 115
  Once the AIDA Plots processors are run via Marlin, a RAIDA root file is produced. We customised the MakePurityVsEfficiencyRootPlot.C macro and ran it to use this RAIDA file as input to produce the purity vs. efficiency plots:
Changed:
<
<
root -l MakePurityVsEfficiencyRootPlot.C
>
>
root -l bbPurityVsEfficiencyRootPlotGla.C
 
Changed:
<
<

>
>
Here's a plot of our flavour tagging purity vs. efficiencies (using cca. 22k events):
Screen_Shot_2013-07-16_at_13.40.40.png
 
META FILEATTACHMENT attachment="BRTotalUncertBands_lm.png" attr="" comment="Higgs branching ratios (from A. Denner et al., EPJ C71, p.1753)" date="1372689661" name="BRTotalUncertBands_lm.png" path="BRTotalUncertBands.png" size="114345" stream="BRTotalUncertBands.png" tmpFilename="/usr/tmp/CGItemp36847" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="Timing-ScreenShot.png" attr="" comment="Screen shot: time used by Marlin processors" date="1372694627" name="Timing-ScreenShot.png" path="Timing-ScreenShot.png" size="72589" stream="Timing-ScreenShot.png" tmpFilename="/usr/tmp/CGItemp32197" user="DanProtopopescu" version="1"
Line: 115 to 126
 
META FILEATTACHMENT attachment="Vertexing_Howto.pdf" attr="" comment="Vertexing HowTo (Ben Jeffery)" date="1372697529" name="Vertexing_Howto.pdf" path="Vertexing Howto.pdf" size="761330" stream="Vertexing Howto.pdf" tmpFilename="/usr/tmp/CGItemp32366" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="jet_truth_tag-steer.xml" attr="" comment="Steering file 1: jet finder and truth flavour tagging" date="1372699245" name="jet_truth_tag-steer.xml" path="jet_truth_tag-steer.xml" size="84107" stream="jet_truth_tag-steer.xml" tmpFilename="/usr/tmp/CGItemp32410" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="runLCFI.txt" attr="" comment="Script to run LCFI sequence of processors" date="1372954507" name="runLCFI.txt" path="runLCFI.txt" size="1170" stream="runLCFI.txt" tmpFilename="/usr/tmp/CGItemp32298" user="DanProtopopescu" version="1"
Added:
>
>
META FILEATTACHMENT attachment="Screen_Shot_2013-07-16_at_13.40.40.png" attr="" comment="Flavour tagging purity vs. efficiency plots (using cca. 22k events)" date="1373979756" name="Screen_Shot_2013-07-16_at_13.40.40.png" path="Screen Shot 2013-07-16 at 13.40.40.png" size="22046" stream="Screen Shot 2013-07-16 at 13.40.40.png" tmpFilename="/usr/tmp/CGItemp35477" user="DanProtopopescu" version="1"
 
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 112013-07-10 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Line: 36 to 36
  Table of input and output collections for our setup (one can choose other names, of course):
Changed:
<
<
Processor Type Input Collection name Output Collection name
JetFinder SatoruJetFinder LooseSelectedPandoraPFANewPFOs Durham_4Jets
MyTrueAngularJetFlavour TrueAngularJetFlavour MCParticlesSkimmed, Durham_4Jets TrueJetFlavour_4Jets
IPRPCut RPCut LooseSelectedPandoraPFANewPFOs IPFitSelectedParticles
MyPerEventIPFitter PerEventIPFitter IPFitSelectedParticles IPVertex
ZVRESRPCut RPCut RecoMCTruthLink, Durham_4Jets ZVRESSelectedJets
MyZVTOP_ZVRES ZVTOP_ZVRES IPVertex, ZVRESSelectedJets ZVRESDecayChains, ZVRESDecayChainRPTracks, ZVRESSelectedJets
FTRPCut RPCut RecoMCTruthLink, ZVRESDecayChains FTSelectedJets
MyFlavourTagInputs FlavourTagInputs ZVRESDecayChains, FTSelectedJets FlavourTagInputs
>
>
# Processor Type Input Collection name Output Collection name
1 JetFinder SatoruJetFinder LooseSelectedPandoraPFANewPFOs Durham_4Jets
2 MyTrueAngularJetFlavour TrueAngularJetFlavour MCParticlesSkimmed, Durham_4Jets TrueJetFlavour_4Jets
3 IPRPCut RPCut LooseSelectedPandoraPFANewPFOs IPFitSelectedParticles
4 MyPerEventIPFitter PerEventIPFitter IPFitSelectedParticles IPVertex
5 ZVRESRPCut RPCut RecoMCTruthLink, Durham_4Jets ZVRESSelectedJets
6 MyZVTOP_ZVRES ZVTOP_ZVRES IPVertex, ZVRESSelectedJets ZVRESDecayChains, ZVRESDecayChainRPTracks, ZVRESSelectedJets
7 FTRPCut RPCut RecoMCTruthLink, ZVRESDecayChains FTSelectedJets
8 MyFlavourTagInputs FlavourTagInputs ZVRESDecayChains, FTSelectedJets FlavourTagInputs
  Our input slcio files contain the collections: LooseSelectedPandoraPFANewPFOs, MCParticlesSkimmed, PandoraPFANewClusters, PandoraPFANewPFOs, PandoraPFANewReclusterMonitoring, ProngVertices, RecoMCTruthLink, SelectedLDCTracks, SelectedPandoraPFANewPFOs, TightSelectedPandoraPFANewPFOs and V0Vertices.

Line: 102 to 102
  to the LCFIVertex/CMakeLists.txt file, sourced the root environment, then ran cmake and make install.
Changed:
<
<
Once the AIDA Plots processors are run via Marlin, a RAIDA root file is produced. We customised the MakePurityVsEfficiencyRootPlot.C macro and ran it to use this RAIDA file as input to produce the purity vs. efficiency plots.
>
>
Once the AIDA Plots processors are run via Marlin, a RAIDA root file is produced. We customised the MakePurityVsEfficiencyRootPlot.C macro and ran it to use this RAIDA file as input to produce the purity vs. efficiency plots:
root -l MakePurityVsEfficiencyRootPlot.C
 

Revision 102013-07-09 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Jet Flavour Tagging Howto

Deleted:
<
<
 This is a detailed record on how the Marlin framework and included LCFI packages are used for jet flavour tagging. b-jet flavour tagging is part of our analysis of the feasibility of the ZZ fusion channel with CLIC ILD at 1.4 TeV.

Deleted:
<
<
BRTotalUncertBands.png
 

Jet Finder and Truth Tagging

Line: 84 to 79
 

Purity and Efficiency Studies

Changed:
<
<
To determine the optimal cut for our b-tagging, a purity vs. efficiency study was performed. One can use the MakePurityVsEfficiencyRootPlot.C macro provided by the LCFIVertex package.
>
>
To determine the optimal cut for our b-tagging, a purity vs. efficiency study needs to be done. We use the MakePurityVsEfficiencyRootPlot.C macro provided by the LCFIVertex package.
 First, we have to run:
  <processor name="MyAIDAProcessor"/>
Line: 95 to 90
 
<parameter name="TrueJetFlavourCollection" type="string">TrueJetFlavour_4Jets </parameter>
Changed:
<
<
Note that LCFI must be compiled with ROOT if one wants .root output from PlotProcessor (instead of .txt). For this, add as usual
>
>
Note that LCFI must be compiled with ROOT if one wants .root output from PlotProcessor (instead of .txt). For this, we added
 
FIND_PACKAGE( ROOT REQUIRED )
FOREACH( pkg ROOT ) 
Line: 105 to 100
  ENDIF() ENDFOREACH()
Changed:
<
<
to the LCFIVertex/CMakeLists.txt file, source the root environment, then run cmake and make install.
>
>
to the LCFIVertex/CMakeLists.txt file, sourced the root environment, then ran cmake and make install.
 
Changed:
<
<
Once the Plots processors are run via Marlin, a RAIDA root file will be produced. Customise the MakePurityVsEfficiencyRootPlot.C macro and run it to use the RAIDA as input to produce the purity vs. efficiency plots.
>
>
Once the AIDA Plots processors are run via Marlin, a RAIDA root file is produced. We customised the MakePurityVsEfficiencyRootPlot.C macro and ran it to use this RAIDA file as input to produce the purity vs. efficiency plots.
 

Revision 92013-07-05 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"
Changed:
<
<

ZZ Fusion Analysis

>
>

Jet Flavour Tagging Howto

 
Changed:
<
<
This is a detailed record on how the Marlin framework and adjacent packages are used for our analysis of the feasibility of the ZZ fusion channel with CLIC ILD at 1.4 TeV.
>
>
This is a detailed record on how the Marlin framework and included LCFI packages are used for jet flavour tagging. b-jet flavour tagging is part of our analysis of the feasibility of the ZZ fusion channel with CLIC ILD at 1.4 TeV.
 
Line: 11 to 11
 
Changed:
<
<

Jet Finder and Flavour Tagging

>
>

Jet Finder and Truth Tagging

  We use the LCFI flavour tagging package. This package consists of a topological vertex finder ZVTOP, which reconstructs secondary interactions, and a multivariate classifier which combines several jet-related variables to tag bottom, charm, and light quark jets (see diagram).
Line: 53 to 53
  Our input slcio files contain the collections: LooseSelectedPandoraPFANewPFOs, MCParticlesSkimmed, PandoraPFANewClusters, PandoraPFANewPFOs, PandoraPFANewReclusterMonitoring, ProngVertices, RecoMCTruthLink, SelectedLDCTracks, SelectedPandoraPFANewPFOs, TightSelectedPandoraPFANewPFOs and V0Vertices.

Changed:
<
<
The processors listed above could be run in sequence, or split in several steps, invoking a LCIOOutput processor to write intermediate slcio outputs at every step. Here's a script for that. We found that the most time-consuming processor is ZVTOP_ZVRES with more than 10 s/event.
>
>
The processors listed above could be run in sequence, or split in several steps, invoking a LCIOOutput processor to write intermediate slcio outputs at every step. Here's a script for that, where the intermediate xml files are slight modifications of the files provided in LCFIVertex/steering examples. We found that the most time-consuming processor is ZVTOP_ZVRES with more than 10 s/event.
  The LCIOOutput processor creates new slcio files containing the new collections added by the above processors.
Changed:
<
<
Troubleshooting: The b3_D0CutValue parameter of the IPRPCutProcessor was set to 5O instead of 50, and was causing a crash. For the ZVRESRPCut processor, h1_MCPIDEnable had to be set to false.
>
>
Troubleshooting: The b3_D0CutValue parameter of the IPRPCutProcessor was set to 5O instead of 50, and was causing a crash. For the ZVRESRPCut processor, h1_MCPIDEnable had to be set to false. See also this post.
 

Neural Network Training

Line: 69 to 71
  The neural nets are saved as XML files in nnets/ and will be used for flavour tagging (next step). No slcio output is written at this time.
Deleted:
<
<

Flavour Tagging

 
Changed:
<
<
Now are ready to employ the FlavourTag processor, which will do flavour tagging using the neural nets trained in the previous step. Our steering file contains the following processors:
>
>

Flavour Tagging

 
Added:
>
>
Now we are ready to employ the FlavourTag processor, which will do flavour tagging using the neural nets trained in the previous step. The input slcio file contains the FlavourTagInputs and FTSelectedJets (or Durham_4Jets, not sure if there's a difference at this level) collections.
 
  <processor name="MyFlavourTag"/>
Added:
>
>
 
Added:
>
>
The output slcio will contain the collection FlavourTag which will be used for our ZZFusion analysis.
 

Purity and Efficiency Studies

Changed:
<
<
To determine the optimal cut for our b-tagging, a purity vs. efficiency study was performed. One can use the MakePurityVsEfficiencyRootPlot.C macro provided by the LCFIVertex package. Note that LCFI must be compiled with ROOT if one wants .root output from PlotProcessor (instead of .txt). For this, add as usual
>
>
To determine the optimal cut for our b-tagging, a purity vs. efficiency study was performed. One can use the MakePurityVsEfficiencyRootPlot.C macro provided by the LCFIVertex package. First, we have to run:
  <processor name="MyAIDAProcessor"/>
  <processor name="MyPlot"/>
  <processor name="MyLCFIAIDAPlotProcessor"/>
We had to provide MyPlot with the actual name of the TrueJetFlavourCollection:
<parameter name="TrueJetFlavourCollection" type="string">TrueJetFlavour_4Jets </parameter>
Note that LCFI must be compiled with ROOT if one wants .root output from PlotProcessor (instead of .txt). For this, add as usual
 
FIND_PACKAGE( ROOT REQUIRED )
FOREACH( pkg ROOT ) 
Line: 90 to 105
  ENDIF() ENDFOREACH()
Changed:
<
<
to the LCFIVertex/CMakeLists.txt file, source the root environment, then run cmake and make install. We also had to provide MyPlot with the actual name of the TrueJetFlavourCollection:
<parameter name="TrueJetFlavourCollection" type="string">TrueJetFlavour_4Jets </parameter>
Customise the MakePurityVsEfficiencyRootPlot.C macro and run it to produce the purity vs. efficiency plots.

Acceptance Studies

Electrons from Hard Bremsstrahlung

>
>
to the LCFIVertex/CMakeLists.txt file, source the root environment, then run cmake and make install.
 
Changed:
<
<

Luminosity and Event Weights

>
>
Once the Plots processors are run via Marlin, a RAIDA root file will be produced. Customise the MakePurityVsEfficiencyRootPlot.C macro and run it to use the RAIDA as input to produce the purity vs. efficiency plots.
 

Line: 112 to 117
 
META FILEATTACHMENT attachment="Vertexing_Howto.pdf" attr="" comment="Vertexing HowTo (Ben Jeffery)" date="1372697529" name="Vertexing_Howto.pdf" path="Vertexing Howto.pdf" size="761330" stream="Vertexing Howto.pdf" tmpFilename="/usr/tmp/CGItemp32366" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="jet_truth_tag-steer.xml" attr="" comment="Steering file 1: jet finder and truth flavour tagging" date="1372699245" name="jet_truth_tag-steer.xml" path="jet_truth_tag-steer.xml" size="84107" stream="jet_truth_tag-steer.xml" tmpFilename="/usr/tmp/CGItemp32410" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="runLCFI.txt" attr="" comment="Script to run LCFI sequence of processors" date="1372954507" name="runLCFI.txt" path="runLCFI.txt" size="1170" stream="runLCFI.txt" tmpFilename="/usr/tmp/CGItemp32298" user="DanProtopopescu" version="1"
Added:
>
>
META TOPICMOVED by="DanProtopopescu" date="1373037183" from="LinearCollider.ZZfusionMain" to="LinearCollider.JetFlavourTagging"

Revision 82013-07-05 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

ZZ Fusion Analysis

Line: 69 to 69
  The neural nets are saved as XML files in nnets/ and will be used for flavour tagging (next step). No slcio output is written at this time.
Deleted:
<
<

Purity and Efficiency Studies

To determine the optimal cut for our b-tagging, a purity vs. efficiency study was performed.

 

Flavour Tagging

Now are ready to employ the FlavourTag processor, which will do flavour tagging using the neural nets trained in the previous step. Our steering file contains the following processors:

  <processor name="MyFlavourTag"/>
Deleted:
<
<
 
Changed:
<
<
The ZZfusion processor is used in our analysis.
>
>

Purity and Efficiency Studies

To determine the optimal cut for our b-tagging, a purity vs. efficiency study was performed. One can use the MakePurityVsEfficiencyRootPlot.C macro provided by the LCFIVertex package. Note that LCFI must be compiled with ROOT if one wants .root output from PlotProcessor (instead of .txt). For this, add as usual

FIND_PACKAGE( ROOT REQUIRED )
FOREACH( pkg ROOT ) 
 IF( ${pkg}_FOUND )
  INCLUDE_DIRECTORIES( ${${pkg}_INCLUDE_DIRS} )
  ADD_DEFINITIONS( ${${pkg}_DEFINITIONS} )
 ENDIF()
ENDFOREACH()
to the LCFIVertex/CMakeLists.txt file, source the root environment, then run cmake and make install. We also had to provide MyPlot with the actual name of the TrueJetFlavourCollection:
<parameter name="TrueJetFlavourCollection" type="string">TrueJetFlavour_4Jets </parameter>
Customise the MakePurityVsEfficiencyRootPlot.C macro and run it to produce the purity vs. efficiency plots.
 

Acceptance Studies

Revision 72013-07-04 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

ZZ Fusion Analysis

Line: 28 to 28
 
Changed:
<
<
The JetFinder processor reconstructs 4 jets events from the input collection (LooseSelectedPandoraPFANewPFOs was used). For the reconstructed 4 jets, MyTrueAngularJetFlavourProcessor determines MC Jet Flavour by angular matching of heavy quarks to jets, and also determines hadronic and partonic charge of the jet
>
>
The JetFinder processor reconstructs 2 and 4 jets events from the input collection (LooseSelectedPandoraPFANewPFOs was used). For the reconstructed 4 jets, MyTrueAngularJetFlavourProcessor determines MC Jet Flavour by angular matching of heavy quarks to jets, and also determines hadronic and partonic charge of the jet.
  The LCFI processors have the following functions:
Line: 53 to 53
  Our input slcio files contain the collections: LooseSelectedPandoraPFANewPFOs, MCParticlesSkimmed, PandoraPFANewClusters, PandoraPFANewPFOs, PandoraPFANewReclusterMonitoring, ProngVertices, RecoMCTruthLink, SelectedLDCTracks, SelectedPandoraPFANewPFOs, TightSelectedPandoraPFANewPFOs and V0Vertices.

Changed:
<
<
The LCIOOutput processor creates new slcio files containing the new collections added by the above processors. We found that the most time-consuming processor is ZVTOP_ZVRES with more than 10 s/event.
>
>
The processors listed above could be run in sequence, or split in several steps, invoking a LCIOOutput processor to write intermediate slcio outputs at every step. Here's a script for that. We found that the most time-consuming processor is ZVTOP_ZVRES with more than 10 s/event.

The LCIOOutput processor creates new slcio files containing the new collections added by the above processors.

Troubleshooting: The b3_D0CutValue parameter of the IPRPCutProcessor was set to 5O instead of 50, and was causing a crash. For the ZVRESRPCut processor, h1_MCPIDEnable had to be set to false.

 

Neural Network Training

Line: 94 to 98
 
META FILEATTACHMENT attachment="LCFI_Flow_Diagram.pdf" attr="" comment="LCFI processors - flow diagram" date="1372695037" name="LCFI_Flow_Diagram.pdf" path="LCFI Flow Diagram.pdf" size="89775" stream="LCFI Flow Diagram.pdf" tmpFilename="/usr/tmp/CGItemp32291" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="Vertexing_Howto.pdf" attr="" comment="Vertexing HowTo (Ben Jeffery)" date="1372697529" name="Vertexing_Howto.pdf" path="Vertexing Howto.pdf" size="761330" stream="Vertexing Howto.pdf" tmpFilename="/usr/tmp/CGItemp32366" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="jet_truth_tag-steer.xml" attr="" comment="Steering file 1: jet finder and truth flavour tagging" date="1372699245" name="jet_truth_tag-steer.xml" path="jet_truth_tag-steer.xml" size="84107" stream="jet_truth_tag-steer.xml" tmpFilename="/usr/tmp/CGItemp32410" user="DanProtopopescu" version="1"
Added:
>
>
META FILEATTACHMENT attachment="runLCFI.txt" attr="" comment="Script to run LCFI sequence of processors" date="1372954507" name="runLCFI.txt" path="runLCFI.txt" size="1170" stream="runLCFI.txt" tmpFilename="/usr/tmp/CGItemp32298" user="DanProtopopescu" version="1"

Revision 62013-07-03 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

ZZ Fusion Analysis

Line: 57 to 57
 

Neural Network Training

Changed:
<
<
The slcio files created at the previous step contain the collections FTSelectedJets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
>
>
The slcio files created at the previous step contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
 
  <processor name="MyNeuralNetTrainer" type="NeuralNetTrainer"/>

Revision 52013-07-03 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

ZZ Fusion Analysis

Line: 28 to 28
 
Changed:
<
<
The JetFinder processor reconstructs 4 jets events from the input collection (LooseSelectedPandoraPFANewPFOs was used). For the reconstructed 4 jets, TrueAngularJetFlavourProcessor does 'truth tagging', i.e. determines the MC jet flavour (b-jet or c-jet).
>
>
The JetFinder processor reconstructs 4 jets events from the input collection (LooseSelectedPandoraPFANewPFOs was used). For the reconstructed 4 jets, MyTrueAngularJetFlavourProcessor determines MC Jet Flavour by angular matching of heavy quarks to jets, and also determines hadronic and partonic charge of the jet
  The LCFI processors have the following functions:

  • IPRPCut - selects Reconstructed Particles based on track parameters, number of hits etc.
Changed:
<
<
  • PerEventIPFitter - finds the event Interaction Point
  • ZVRESRP - vertex finder for reconstructed particles
  • ZVTOP_ZVRES - topological vertex finder
  • FTRPCut: - flavour tagging reconstructed particle cuts
  • FlavourTagInputs - from vertices and tracks calculates discriminating variables for the neural net
>
>
  • MyPerEventIPFitter - determines IP position and error from the tracks in an event by simple fitting
  • ZVRESRPCut - applies cuts on the d0 and z0 values of the track
  • MyZVTOP_ZVRES - topological vertex finder
  • FTRPCut - flavour tagging reconstructed particle cuts (on d0, z0 and PT)
  • MyFlavourTagInputs - from vertices and tracks calculates discriminating variables for the neural net

Table of input and output collections for our setup (one can choose other names, of course):

Processor Type
<-- -->
Sorted descending
Input Collection name Output Collection name
MyZVTOP_ZVRES ZVTOP_ZVRES IPVertex, ZVRESSelectedJets ZVRESDecayChains, ZVRESDecayChainRPTracks, ZVRESSelectedJets
MyTrueAngularJetFlavour TrueAngularJetFlavour MCParticlesSkimmed, Durham_4Jets TrueJetFlavour_4Jets
JetFinder SatoruJetFinder LooseSelectedPandoraPFANewPFOs Durham_4Jets
IPRPCut RPCut LooseSelectedPandoraPFANewPFOs IPFitSelectedParticles
ZVRESRPCut RPCut RecoMCTruthLink, Durham_4Jets ZVRESSelectedJets
FTRPCut RPCut RecoMCTruthLink, ZVRESDecayChains FTSelectedJets
MyPerEventIPFitter PerEventIPFitter IPFitSelectedParticles IPVertex
MyFlavourTagInputs FlavourTagInputs ZVRESDecayChains, FTSelectedJets FlavourTagInputs

Our input slcio files contain the collections: LooseSelectedPandoraPFANewPFOs, MCParticlesSkimmed, PandoraPFANewClusters, PandoraPFANewPFOs, PandoraPFANewReclusterMonitoring, ProngVertices, RecoMCTruthLink, SelectedLDCTracks, SelectedPandoraPFANewPFOs, TightSelectedPandoraPFANewPFOs and V0Vertices.

  The LCIOOutput processor creates new slcio files containing the new collections added by the above processors. We found that the most time-consuming processor is ZVTOP_ZVRES with more than 10 s/event.

Neural Network Training

Changed:
<
<
The slcio files created at the previous step contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
>
>
The slcio files created at the previous step contain the collections FTSelectedJets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:
 
  <processor name="MyNeuralNetTrainer" type="NeuralNetTrainer"/>

Revision 42013-07-01 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

ZZ Fusion Analysis

Added:
>
>
This is a detailed record on how the Marlin framework and adjacent packages are used for our analysis of the feasibility of the ZZ fusion channel with CLIC ILD at 1.4 TeV.
 
Added:
>
>
BRTotalUncertBands.png
 
Changed:
<
<

Neural Network training

>
>

Jet Finder and Flavour Tagging

 
Changed:
<
<
We use the flavour tagging package LCFIVertex. This package consists of a topological vertex finder ZVTOP, which reconstructs secondary interactions, and a multivariate classifier which combines several jet-related variables to tag bottom, charm, and light quark jets.
>
>
We use the LCFI flavour tagging package. This package consists of a topological vertex finder ZVTOP, which reconstructs secondary interactions, and a multivariate classifier which combines several jet-related variables to tag bottom, charm, and light quark jets (see diagram).
 
Changed:
<
<
Our steering file will contain the following LCFI component processors and a neural net trainer:
>
>
Our steering file will contain the jet finder, flavour tagging and LCFI processors, and we will write new slcio files containing the added collections:
 
  <group name="JetFinders"/>
  <group name="MyTrueAngularJetFlavourProcessorCollection"/>
Line: 18 to 25
 
Changed:
<
<
>
>
 
Changed:
<
<
These processors have the following functions:
>
>
The JetFinder processor reconstructs 4 jets events from the input collection (LooseSelectedPandoraPFANewPFOs was used). For the reconstructed 4 jets, TrueAngularJetFlavourProcessor does 'truth tagging', i.e. determines the MC jet flavour (b-jet or c-jet).

The LCFI processors have the following functions:

 
  • IPRPCut - selects Reconstructed Particles based on track parameters, number of hits etc.
Changed:
<
<
  • PerEventIPFitter - finds the event Impact Parameter
>
>
  • PerEventIPFitter - finds the event Interaction Point
 
  • ZVRESRP - vertex finder for reconstructed particles
  • ZVTOP_ZVRES - topological vertex finder
  • FTRPCut: - flavour tagging reconstructed particle cuts
  • FlavourTagInputs - from vertices and tracks calculates discriminating variables for the neural net
Changed:
<
<
  • NeuralNetTrainer - trains the networks
>
>
The LCIOOutput processor creates new slcio files containing the new collections added by the above processors. We found that the most time-consuming processor is ZVTOP_ZVRES with more than 10 s/event.

Neural Network Training

The slcio files created at the previous step contain the collections Durham_4Jets, FlavourTagInputs and TrueJetFlavour_4Jets, which we will use now to train our neural nets. We use the NeuralNetTrainer code included in the LCFI package. Separate nets were trained for 1, 2, or 3+ vertices to identify b-jets, c-jets, and c-jets with b background. Our steering file contains only:

  <processor name="MyNeuralNetTrainer" type="NeuralNetTrainer"/>
  The neural nets are saved as XML files in nnets/ and will be used for flavour tagging (next step). No slcio output is written at this time.
Added:
>
>

Purity and Efficiency Studies

To determine the optimal cut for our b-tagging, a purity vs. efficiency study was performed.

 

Flavour Tagging

Changed:
<
<
  • TrueJetFlavour - determines MC flavour of reconstructed jets
>
>
Now are ready to employ the FlavourTag processor, which will do flavour tagging using the neural nets trained in the previous step. Our steering file contains the following processors:
 
Changed:
<
<
A new file containing with these collections added is saved to be used in our analysis.
>
>
  <processor name="MyFlavourTag"/>
  <processor name="ZZfusion"/>
 
Changed:
<
<

Acceptance Studies

>
>
The ZZfusion processor is used in our analysis.
 
Changed:
<
<

Four Jet Events

>
>

Acceptance Studies

 
Deleted:
<
<

Purity and Efficiency Studies

 

Electrons from Hard Bremsstrahlung

Luminosity and Event Weights

Deleted:
<
<
-- DanProtopopescu - 2013-06-28
 \ No newline at end of file
Added:
>
>

META FILEATTACHMENT attachment="BRTotalUncertBands_lm.png" attr="" comment="Higgs branching ratios (from A. Denner et al., EPJ C71, p.1753)" date="1372689661" name="BRTotalUncertBands_lm.png" path="BRTotalUncertBands.png" size="114345" stream="BRTotalUncertBands.png" tmpFilename="/usr/tmp/CGItemp36847" user="DanProtopopescu" version="2"
META FILEATTACHMENT attachment="Timing-ScreenShot.png" attr="" comment="Screen shot: time used by Marlin processors" date="1372694627" name="Timing-ScreenShot.png" path="Timing-ScreenShot.png" size="72589" stream="Timing-ScreenShot.png" tmpFilename="/usr/tmp/CGItemp32197" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="LCFI_Flow_Diagram.pdf" attr="" comment="LCFI processors - flow diagram" date="1372695037" name="LCFI_Flow_Diagram.pdf" path="LCFI Flow Diagram.pdf" size="89775" stream="LCFI Flow Diagram.pdf" tmpFilename="/usr/tmp/CGItemp32291" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="Vertexing_Howto.pdf" attr="" comment="Vertexing HowTo (Ben Jeffery)" date="1372697529" name="Vertexing_Howto.pdf" path="Vertexing Howto.pdf" size="761330" stream="Vertexing Howto.pdf" tmpFilename="/usr/tmp/CGItemp32366" user="DanProtopopescu" version="1"
META FILEATTACHMENT attachment="jet_truth_tag-steer.xml" attr="" comment="Steering file 1: jet finder and truth flavour tagging" date="1372699245" name="jet_truth_tag-steer.xml" path="jet_truth_tag-steer.xml" size="84107" stream="jet_truth_tag-steer.xml" tmpFilename="/usr/tmp/CGItemp32410" user="DanProtopopescu" version="1"

Revision 32013-07-01 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

ZZ Fusion Analysis

Line: 11 to 11
 Our steering file will contain the following LCFI component processors and a neural net trainer:
  <group name="JetFinders"/>
Added:
>
>
 

Revision 22013-07-01 - DanProtopopescu

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

ZZ Fusion Analysis

Line: 10 to 10
  Our steering file will contain the following LCFI component processors and a neural net trainer:
Added:
>
>
 

Revision 12013-06-28 - DanProtopopescu

Line: 1 to 1
Added:
>
>
META TOPICPARENT name="WebHome"

ZZ Fusion Analysis

Neural Network training

We use the flavour tagging package LCFIVertex. This package consists of a topological vertex finder ZVTOP, which reconstructs secondary interactions, and a multivariate classifier which combines several jet-related variables to tag bottom, charm, and light quark jets.

Our steering file will contain the following LCFI component processors and a neural net trainer:

  <processor name="IPRPCutProcessor"/>
  <processor name="MyPerEventIPFitterProcessor"/>
  <processor name="ZVRESRPCutProcessor"/>
  <processor name="MyZVTOP_ZVRES"/>
  <processor name="FTRPCutProcessor"/>
  <processor name="MyFlavourTagInputsProcessor"/>
  <processor name="MyNeuralNetTrainer" type="GlasgowNeuralNetTrainer"/>

These processors have the following functions:

  • IPRPCut - selects Reconstructed Particles based on track parameters, number of hits etc.
  • PerEventIPFitter - finds the event Impact Parameter
  • ZVRESRP - vertex finder for reconstructed particles
  • ZVTOP_ZVRES - topological vertex finder
  • FTRPCut: - flavour tagging reconstructed particle cuts
  • FlavourTagInputs - from vertices and tracks calculates discriminating variables for the neural net
  • NeuralNetTrainer - trains the networks

The neural nets are saved as XML files in nnets/ and will be used for flavour tagging (next step). No slcio output is written at this time.

Flavour Tagging

  • TrueJetFlavour - determines MC flavour of reconstructed jets

A new file containing with these collections added is saved to be used in our analysis.

Acceptance Studies

Four Jet Events

Purity and Efficiency Studies

Electrons from Hard Bremsstrahlung

Luminosity and Event Weights

-- DanProtopopescu - 2013-06-28

 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback