tmva-users — Open mailing list for TMVA users

Re: [TMVA-users] Fwd: Event normalization in tmva training

[Helge V. <Hel...@ce...>]

Hi Betty,

well, obviously simply scaling by the cross section is 'meaningless' if you
don't take into account the
number of events generated. Just think .. in another world, maybe the
guy/girl responsible for the MC
generation simply generated 2x more of one particular bkg-source just
because he/she regarded it as more
interesting ... if you simply scale by 'cross section' you'd get a
different event sample distribution w/o changing
any physics..

So you'd probably want to scale them such that the generated integrated
luminosity for all bkg sources is the same. Assuming you don't
have any 'preselection cuts' in the MC generation (which is probably a
wrong assumption), then scaling each bkg
by weighting all bkg sources i, bkg(i) by  int_lum(i) is then proportional
to  #events(i)/crosssection(i), so in that case you'd have to
get a multiplicative event weight of
weight(i)=1/int_lumi(i)=crosssection(i)/#events-generated(i)  ..

that would make the distriution of the sum of the bkg events in your
simulation look like that one expected in the data, which
is probably a good way to go as default for any mva training (not, it's not
a 'must' .. you can also somehow weigh the various
types different, get a slightly different classifier... , you'd just have
to watch out how to calculate backgr. rejection then ..)

Cheers,

Helge


On 11 October 2016 at 14:27, Betty Calpas <bet...@ce...> wrote:

> Dear,
>
> any idea?
>
> Regards
>
>
> -------- Forwarded Message --------
> Subject: Event normalization in tmva training
> Date: Tue, 11 Oct 2016 11:30:36 +0200
> From: Betty Calpas <bet...@ce...> <bet...@ce...>
> To: tmv...@li... <tmv...@li...>
> <tmv...@li...>
>
> Dear experts,
>
> to normalize my bkg btw them, I multiplyed each event by
>
> their cross section. But the bkg samples have different
>
> initial number of event, so I wonder if applying the cross
>
> section as a weight is enough? Should I take into account
>
> the initial number of event?
>
> Regards
>
>
>
> ------------------------------------------------------------
> ------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> _______________________________________________
> TMVA-users mailing list
> TMV...@li...
> https://lists.sourceforge.net/lists/listinfo/tmva-users
>
>

[TMVA-users] Fwd: Event normalization in tmva training

[Betty C. <bet...@ce...>]

Dear,

any idea?

Regards



-------- Forwarded Message --------
Subject: 	Event normalization in tmva training
Date: 	Tue, 11 Oct 2016 11:30:36 +0200
From: 	Betty Calpas <bet...@ce...>
To: 	tmv...@li... <tmv...@li...>



Dear experts,

to normalize my bkg btw them, I multiplyed each event by

their cross section. But the bkg samples have different

initial number of event, so I wonder if applying the cross

section as a weight is enough? Should I take into account

the initial number of event?

Regards

[TMVA-users] Event normalization in tmva training

[Betty C. <bet...@ce...>]

Dear experts,

to normalize my bkg btw them, I multiplyed each event by

their cross section. But the bkg samples have different

initial number of event, so I wonder if applying the cross

section as a weight is enough? Should I take into account

the initial number of event?

Regards

Re: [TMVA-users] Fwd: reader->AddVariable(string, float)

[Peter S. <pet...@gm...>]

Great!

On Fri, Sep 30, 2016 at 2:54 PM Betty Calpas <bet...@ce...> wrote:

> Dear Peter,
>
> yes this works.
>
> Regards
>
>
> On 9/30/16 2:50 PM, Peter Speckmayer wrote:
>
> Hi Betty,
>
> If you use the official ROOT, then you have to copy 'by hand' from the
> incoming double variable to another float variable.
>
>
> something like:
>
> Float_t pt_1;
> Double_t pt_1_d;
>
> reader->AddVariable( "pt_1",  &pt_1 );
> ...
>
> inputtree->SetBranchAddress( "pt_1",  &pt_1_d );
>
> and in the event-copying loop you have to pt_1 = pt_1_d;
>
> cheers,
> Peter
>
>
>
> On Fri, Sep 30, 2016 at 1:39 PM Betty Calpas <bet...@ce...> wrote:
>
>> Hi Peter,
>>
>> thank you it help, but I do not understand why the AddVariable can't make
>> the "conversion".
>>
>>
>> Regards
>>
>> On 9/30/16 1:34 PM, Peter Speckmayer wrote:
>>
>> Hi Betty,
>>
>> Could you try to add a new version of "AddVariable" to the Reader?
>>
>> in Reader.h:
>>       void     AddVariable( const TString& expression, Double_t* );
>>
>>
>> in Reader.cxx:
>> void TMVA::Reader::AddVariable( const TString& expression, Double_t*
>> datalink )
>> {
>>    DataInfo().AddVariable( expression, "", "", 0, 0, 'D', kFALSE
>> ,(void*)datalink ); // <= should this be F or rather T?
>> }
>>
>> let me know if this helps.
>>
>> cheers,
>> Peter
>>
>>
>>
>>
>> On Fri, Sep 30, 2016 at 1:10 PM Betty Calpas <bet...@ce...>
>> wrote:
>>
>>> Dear,
>>>
>>> does anyone has an idea?
>>>
>>> Regards
>>>
>>>
>>> -------- Forwarded Message --------
>>> Subject: reader->AddVariable(string, float)
>>> Date: Fri, 30 Sep 2016 10:43:45 +0200
>>> From: Betty Calpas <bet...@ce...> <bet...@ce...>
>>> To: tmv...@li... <tmv...@li...>
>>> <tmv...@li...>
>>>
>>> Dear experts,
>>>
>>> The reader takes only float and int right? but in my tree I have double and
>>>
>>> int. I wonder how I can cast my double in the float? because if I do [1]
>>>
>>> I have the error message [2].
>>>
>>> Regards
>>>
>>>
>>> [1]
>>>
>>> Float_t pt_1;
>>>
>>> reader->AddVariable( "pt_1",  &pt_1 );
>>> ...
>>>
>>> inputtree->SetBranchAddress( "pt_1",  &pt_1 );
>>>
>>> [2]
>>> matching tree with reader variable
>>> Error in <TTree::SetBranchAddress>: The pointer type given "Float_t" (5)
>>> does not correspond to the type needed "Double_t" (8) by the branch: pt_1
>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> TMVA-users mailing list
>>> TMV...@li...
>>> https://lists.sourceforge.net/lists/listinfo/tmva-users
>>>
>>
>>
>> --
>>
>> Cheers,
>> Betty
>>
>
>
> --
>
> Cheers,
> Betty
>

Re: [TMVA-users] Fwd: reader->AddVariable(string, float)

[Betty C. <bet...@ce...>]

Dear Peter,

yes this works.

Regards

On 9/30/16 2:50 PM, Peter Speckmayer wrote:
> Hi Betty,
>
> If you use the official ROOT, then you have to copy 'by hand' from the 
> incoming double variable to another float variable.
>
>
> something like:
>
> Float_t pt_1;
> Double_t pt_1_d;
>
> reader->AddVariable( "pt_1",  &pt_1 );
> ...
>
> inputtree->SetBranchAddress( "pt_1",  &pt_1_d );
>
> and in the event-copying loop you have to pt_1 = pt_1_d;
>
> cheers,
> Peter
>
>
>
> On Fri, Sep 30, 2016 at 1:39 PM Betty Calpas <bet...@ce... 
> <mailto:bet...@ce...>> wrote:
>
>     Hi Peter,
>
>     thank you it help, but I do not understand why the AddVariable
>     can't make the "conversion".
>
>
>     Regards
>
>     On 9/30/16 1:34 PM, Peter Speckmayer wrote:
>>     Hi Betty,
>>
>>     Could you try to add a new version of "AddVariable" to the Reader?
>>
>>     in Reader.h:
>>           void     AddVariable( const TString& expression, Double_t* );
>>
>>
>>     in Reader.cxx:
>>     void TMVA::Reader::AddVariable( const TString& expression,
>>     Double_t* datalink )
>>     {
>>        DataInfo().AddVariable( expression, "", "", 0, 0, 'D', kFALSE
>>     ,(void*)datalink ); // <= should this be F or rather T?
>>     }
>>
>>     let me know if this helps.
>>
>>     cheers,
>>     Peter
>>
>>
>>
>>
>>     On Fri, Sep 30, 2016 at 1:10 PM Betty Calpas
>>     <bet...@ce... <mailto:bet...@ce...>> wrote:
>>
>>         Dear,
>>
>>         does anyone has an idea?
>>
>>         Regards
>>
>>
>>
>>         -------- Forwarded Message --------
>>         Subject: 	reader->AddVariable(string, float)
>>         Date: 	Fri, 30 Sep 2016 10:43:45 +0200
>>         From: 	Betty Calpas <bet...@ce...>
>>         <mailto:bet...@ce...>
>>         To: 	tmv...@li...
>>         <mailto:tmv...@li...>
>>         <tmv...@li...>
>>         <mailto:tmv...@li...>
>>
>>
>>
>>         Dear experts,
>>
>>         The reader takes only float and int right? but in my tree I have double and
>>
>>         int. I wonder how I can cast my double in the float? because if I do [1]
>>
>>         I have the error message [2].
>>
>>         Regards
>>
>>
>>         [1]
>>
>>         Float_t pt_1;
>>
>>         reader->AddVariable( "pt_1",  &pt_1 );
>>         ...
>>
>>         inputtree->SetBranchAddress( "pt_1",  &pt_1 );
>>
>>         [2]
>>         matching tree with reader variable
>>         Error in <TTree::SetBranchAddress>: The pointer type given "Float_t" (5)
>>         does not correspond to the type needed "Double_t" (8) by the branch: pt_1
>>
>>
>>         ------------------------------------------------------------------------------
>>         _______________________________________________
>>         TMVA-users mailing list
>>         TMV...@li...
>>         <mailto:TMV...@li...>
>>         https://lists.sourceforge.net/lists/listinfo/tmva-users
>>
>
>
>     -- 
>
>     Cheers,
>     Betty
>


-- 

Cheers,
Betty

Re: [TMVA-users] Fwd: reader->AddVariable(string, float)

[Peter S. <pet...@gm...>]

Hi Betty,

If you use the official ROOT, then you have to copy 'by hand' from the
incoming double variable to another float variable.


something like:

Float_t pt_1;
Double_t pt_1_d;

reader->AddVariable( "pt_1",  &pt_1 );
...

inputtree->SetBranchAddress( "pt_1",  &pt_1_d );

and in the event-copying loop you have to pt_1 = pt_1_d;

cheers,
Peter



On Fri, Sep 30, 2016 at 1:39 PM Betty Calpas <bet...@ce...> wrote:

> Hi Peter,
>
> thank you it help, but I do not understand why the AddVariable can't make
> the "conversion".
>
>
> Regards
>
> On 9/30/16 1:34 PM, Peter Speckmayer wrote:
>
> Hi Betty,
>
> Could you try to add a new version of "AddVariable" to the Reader?
>
> in Reader.h:
>       void     AddVariable( const TString& expression, Double_t* );
>
>
> in Reader.cxx:
> void TMVA::Reader::AddVariable( const TString& expression, Double_t*
> datalink )
> {
>    DataInfo().AddVariable( expression, "", "", 0, 0, 'D', kFALSE
> ,(void*)datalink ); // <= should this be F or rather T?
> }
>
> let me know if this helps.
>
> cheers,
> Peter
>
>
>
>
> On Fri, Sep 30, 2016 at 1:10 PM Betty Calpas <bet...@ce...> wrote:
>
>> Dear,
>>
>> does anyone has an idea?
>>
>> Regards
>>
>>
>> -------- Forwarded Message --------
>> Subject: reader->AddVariable(string, float)
>> Date: Fri, 30 Sep 2016 10:43:45 +0200
>> From: Betty Calpas <bet...@ce...> <bet...@ce...>
>> To: tmv...@li... <tmv...@li...>
>> <tmv...@li...>
>>
>> Dear experts,
>>
>> The reader takes only float and int right? but in my tree I have double and
>>
>> int. I wonder how I can cast my double in the float? because if I do [1]
>>
>> I have the error message [2].
>>
>> Regards
>>
>>
>> [1]
>>
>> Float_t pt_1;
>>
>> reader->AddVariable( "pt_1",  &pt_1 );
>> ...
>>
>> inputtree->SetBranchAddress( "pt_1",  &pt_1 );
>>
>> [2]
>> matching tree with reader variable
>> Error in <TTree::SetBranchAddress>: The pointer type given "Float_t" (5)
>> does not correspond to the type needed "Double_t" (8) by the branch: pt_1
>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> TMVA-users mailing list
>> TMV...@li...
>> https://lists.sourceforge.net/lists/listinfo/tmva-users
>>
>
>
> --
>
> Cheers,
> Betty
>

Re: [TMVA-users] Fwd: reader->AddVariable(string, float)

[Betty C. <bet...@ce...>]

Hi Peter,

thank you it help, but I do not understand why the AddVariable can't 
make the "conversion".

Regards

On 9/30/16 1:34 PM, Peter Speckmayer wrote:
> Hi Betty,
>
> Could you try to add a new version of "AddVariable" to the Reader?
>
> in Reader.h:
>       void     AddVariable( const TString& expression, Double_t* );
>
>
> in Reader.cxx:
> void TMVA::Reader::AddVariable( const TString& expression, Double_t* 
> datalink )
> {
>    DataInfo().AddVariable( expression, "", "", 0, 0, 'D', kFALSE 
> ,(void*)datalink ); // <= should this be F or rather T?
> }
>
> let me know if this helps.
>
> cheers,
> Peter
>
>
>
>
> On Fri, Sep 30, 2016 at 1:10 PM Betty Calpas <bet...@ce... 
> <mailto:bet...@ce...>> wrote:
>
>     Dear,
>
>     does anyone has an idea?
>
>     Regards
>
>
>
>     -------- Forwarded Message --------
>     Subject: 	reader->AddVariable(string, float)
>     Date: 	Fri, 30 Sep 2016 10:43:45 +0200
>     From: 	Betty Calpas <bet...@ce...>
>     <mailto:bet...@ce...>
>     To: 	tmv...@li...
>     <mailto:tmv...@li...>
>     <tmv...@li...>
>     <mailto:tmv...@li...>
>
>
>
>     Dear experts,
>
>     The reader takes only float and int right? but in my tree I have double and
>
>     int. I wonder how I can cast my double in the float? because if I do [1]
>
>     I have the error message [2].
>
>     Regards
>
>
>     [1]
>
>     Float_t pt_1;
>
>     reader->AddVariable( "pt_1",  &pt_1 );
>     ...
>
>     inputtree->SetBranchAddress( "pt_1",  &pt_1 );
>
>     [2]
>     matching tree with reader variable
>     Error in <TTree::SetBranchAddress>: The pointer type given "Float_t" (5)
>     does not correspond to the type needed "Double_t" (8) by the branch: pt_1
>
>
>     ------------------------------------------------------------------------------
>     _______________________________________________
>     TMVA-users mailing list
>     TMV...@li...
>     <mailto:TMV...@li...>
>     https://lists.sourceforge.net/lists/listinfo/tmva-users
>


-- 

Cheers,
Betty

Re: [TMVA-users] Fwd: reader->AddVariable(string, float)

[Betty C. <bet...@ce...>]

Hi Peter,

I'm running directly from ROOT, should I can't change the method right?

Regards

On 9/30/16 1:34 PM, Peter Speckmayer wrote:
> Hi Betty,
>
> Could you try to add a new version of "AddVariable" to the Reader?
>
> in Reader.h:
>       void     AddVariable( const TString& expression, Double_t* );
>
>
> in Reader.cxx:
> void TMVA::Reader::AddVariable( const TString& expression, Double_t* 
> datalink )
> {
>    DataInfo().AddVariable( expression, "", "", 0, 0, 'D', kFALSE 
> ,(void*)datalink ); // <= should this be F or rather T?
> }
>
> let me know if this helps.
>
> cheers,
> Peter
>
>
>
>
> On Fri, Sep 30, 2016 at 1:10 PM Betty Calpas <bet...@ce... 
> <mailto:bet...@ce...>> wrote:
>
>     Dear,
>
>     does anyone has an idea?
>
>     Regards
>
>
>
>     -------- Forwarded Message --------
>     Subject: 	reader->AddVariable(string, float)
>     Date: 	Fri, 30 Sep 2016 10:43:45 +0200
>     From: 	Betty Calpas <bet...@ce...>
>     <mailto:bet...@ce...>
>     To: 	tmv...@li...
>     <mailto:tmv...@li...>
>     <tmv...@li...>
>     <mailto:tmv...@li...>
>
>
>
>     Dear experts,
>
>     The reader takes only float and int right? but in my tree I have double and
>
>     int. I wonder how I can cast my double in the float? because if I do [1]
>
>     I have the error message [2].
>
>     Regards
>
>
>     [1]
>
>     Float_t pt_1;
>
>     reader->AddVariable( "pt_1",  &pt_1 );
>     ...
>
>     inputtree->SetBranchAddress( "pt_1",  &pt_1 );
>
>     [2]
>     matching tree with reader variable
>     Error in <TTree::SetBranchAddress>: The pointer type given "Float_t" (5)
>     does not correspond to the type needed "Double_t" (8) by the branch: pt_1
>
>
>     ------------------------------------------------------------------------------
>     _______________________________________________
>     TMVA-users mailing list
>     TMV...@li...
>     <mailto:TMV...@li...>
>     https://lists.sourceforge.net/lists/listinfo/tmva-users
>


-- 

Cheers,
Betty

Re: [TMVA-users] Fwd: reader->AddVariable(string, float)

[Peter S. <pet...@gm...>]

Hi Betty,

Could you try to add a new version of "AddVariable" to the Reader?

in Reader.h:
      void     AddVariable( const TString& expression, Double_t* );


in Reader.cxx:
void TMVA::Reader::AddVariable( const TString& expression, Double_t*
datalink )
{
   DataInfo().AddVariable( expression, "", "", 0, 0, 'D', kFALSE
,(void*)datalink ); // <= should this be F or rather T?
}

let me know if this helps.

cheers,
Peter




On Fri, Sep 30, 2016 at 1:10 PM Betty Calpas <bet...@ce...> wrote:

> Dear,
>
> does anyone has an idea?
>
> Regards
>
>
> -------- Forwarded Message --------
> Subject: reader->AddVariable(string, float)
> Date: Fri, 30 Sep 2016 10:43:45 +0200
> From: Betty Calpas <bet...@ce...> <bet...@ce...>
> To: tmv...@li... <tmv...@li...>
> <tmv...@li...>
>
> Dear experts,
>
> The reader takes only float and int right? but in my tree I have double and
>
> int. I wonder how I can cast my double in the float? because if I do [1]
>
> I have the error message [2].
>
> Regards
>
>
> [1]
>
> Float_t pt_1;
>
> reader->AddVariable( "pt_1",  &pt_1 );
> ...
>
> inputtree->SetBranchAddress( "pt_1",  &pt_1 );
>
> [2]
> matching tree with reader variable
> Error in <TTree::SetBranchAddress>: The pointer type given "Float_t" (5)
> does not correspond to the type needed "Double_t" (8) by the branch: pt_1
>
>
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> TMVA-users mailing list
> TMV...@li...
> https://lists.sourceforge.net/lists/listinfo/tmva-users
>

[TMVA-users] Fwd: reader->AddVariable(string, float)

[Betty C. <bet...@ce...>]

Dear,

does anyone has an idea?

Regards



-------- Forwarded Message --------
Subject: 	reader->AddVariable(string, float)
Date: 	Fri, 30 Sep 2016 10:43:45 +0200
From: 	Betty Calpas <bet...@ce...>
To: 	tmv...@li... <tmv...@li...>



Dear experts,

The reader takes only float and int right? but in my tree I have double and

int. I wonder how I can cast my double in the float? because if I do [1]

I have the error message [2].

Regards


[1]

Float_t pt_1;

reader->AddVariable( "pt_1",  &pt_1 );
...

inputtree->SetBranchAddress( "pt_1",  &pt_1 );

[2]
matching tree with reader variable
Error in <TTree::SetBranchAddress>: The pointer type given "Float_t" (5)
does not correspond to the type needed "Double_t" (8) by the branch: pt_1

[TMVA-users] reader->AddVariable(string, float)

[Betty C. <bet...@ce...>]

Dear experts,

The reader takes only float and int right? but in my tree I have double and

int. I wonder how I can cast my double in the float? because if I do [1]

I have the error message [2].

Regards


[1]

Float_t pt_1;

reader->AddVariable( "pt_1",  &pt_1 );
...

inputtree->SetBranchAddress( "pt_1",  &pt_1 );

[2]
matching tree with reader variable
Error in <TTree::SetBranchAddress>: The pointer type given "Float_t" (5) 
does not correspond to the type needed "Double_t" (8) by the branch: pt_1

Re: [TMVA-users] Fwd: Give a specific path to weight file

[Betty C. <bet...@ce...>]

Dear Peter,

thank you for your answer.

when I used [1] I have the error message [2]. In my Makefile I
used 2 TMVA libraries [3], is one missing?

Regards

[1]
(TMVA::gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory";

[2]
main.cc:125:7: error: 'gConfig' is not a member of 'TMVA'
       (TMVA::gConfig().GetIONames()).fWeightFileDir = dir+"_weights";

On 9/27/16 9:30 AM, Peter Speckmayer wrote:

[3]
-lTMVA -lTMVAGui

> Hi,
>
> Please try:
>
>    // If you wish to modify default settings
>
>    // (please check "src/Config.h" to see all available global options)
>
>    //
>
>    // (TMVA::gConfig().GetVariablePlotting()).fTimesRMS = 8.0;
>
>    // (TMVA::gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory";
>
> from the tutorial TMVAClassification.C
>
> Cheers,
> Peter
>
>
> Betty Calpas <bet...@ce... <mailto:bet...@ce...>> 
> schrieb am Di., 27. Sep. 2016, 08:52:
>
>     Dear experts,
>
>     any idea?
>
>     Regards
>
>
>
>     -------- Forwarded Message --------
>     Subject: 	Give a specific path to weight file
>     Date: 	Mon, 26 Sep 2016 14:57:41 +0200
>     From: 	Betty Calpas <bet...@ce...>
>     <mailto:bet...@ce...>
>     To: 	tmv...@li...
>     <mailto:tmv...@li...>
>     <tmv...@li...>
>     <mailto:tmv...@li...>
>
>
>
>     Dear experts,
>
>     when I run TMVAClassification.C, it creates a weight directory
>
>     in the current directory and put the weight file there. I wonder
>
>     how I could set a specific path to write the weight files?
>
>     Regards
>
>     ------------------------------------------------------------------------------
>     _______________________________________________
>     TMVA-users mailing list
>     TMV...@li...
>     <mailto:TMV...@li...>
>     https://lists.sourceforge.net/lists/listinfo/tmva-users
>


-- 

Cheers,
Betty

Re: [TMVA-users] Fwd: Give a specific path to weight file

[Peter S. <pet...@gm...>]

Hi,

Please try:

   // If you wish to modify default settings

   // (please check "src/Config.h" to see all available global options)

   //

   // (TMVA::gConfig().GetVariablePlotting()).fTimesRMS = 8.0;

   // (TMVA::gConfig().GetIONames()).fWeightFileDir = "myWeightDirectory";

from the tutorial TMVAClassification.C

Cheers,
Peter

Betty Calpas <bet...@ce...> schrieb am Di., 27. Sep. 2016, 08:52:

> Dear experts,
>
> any idea?
>
> Regards
>
>
> -------- Forwarded Message --------
> Subject: Give a specific path to weight file
> Date: Mon, 26 Sep 2016 14:57:41 +0200
> From: Betty Calpas <bet...@ce...> <bet...@ce...>
> To: tmv...@li... <tmv...@li...>
> <tmv...@li...>
>
> Dear experts,
>
> when I run TMVAClassification.C, it creates a weight directory
>
> in the current directory and put the weight file there. I wonder
>
> how I could set a specific path to write the weight files?
>
> Regards
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> TMVA-users mailing list
> TMV...@li...
> https://lists.sourceforge.net/lists/listinfo/tmva-users
>

[TMVA-users] Fwd: Give a specific path to weight file

[Betty C. <bet...@ce...>]

Dear experts,

any idea?

Regards



-------- Forwarded Message --------
Subject: 	Give a specific path to weight file
Date: 	Mon, 26 Sep 2016 14:57:41 +0200
From: 	Betty Calpas <bet...@ce...>
To: 	tmv...@li... <tmv...@li...>



Dear experts,

when I run TMVAClassification.C, it creates a weight directory

in the current directory and put the weight file there. I wonder

how I could set a specific path to write the weight files?

Regards

[TMVA-users] Give a specific path to weight file

[Betty C. <bet...@ce...>]

Dear experts,

when I run TMVAClassification.C, it creates a weight directory

in the current directory and put the weight file there. I wonder

how I could set a specific path to write the weight files?

Regards

Re: [TMVA-users] Fwd: Is it possible to train MVA again a single bkg at a time

[Betty C. <bet...@ce...>]

Dear Peter,

thank you for your answer. Please correct me if I'm wrong.
You mean that I can train only sample_1 vs sample_2, and so
if I want to do at the same time sample_1 vs sample_2 and sample_2 vs 
sample_3
it is not possible, I'll have to do it in 2 steps.

Regards

On 9/26/16 10:03 AM, Peter Speckmayer wrote:
> Hi,
>
> With multiclass (for the methods which support it) the training is: 
> each class against the sum of all other classes. Not exactly what you 
> want, but maybe you can work with that.
>
> cheers,
> Peter
>
>
> On Mon, Sep 26, 2016 at 9:32 AM Betty Calpas <bet...@ce... 
> <mailto:bet...@ce...>> wrote:
>
>     Dear experts,
>
>     does anyone know if this is possible?
>
>     Regards
>
>
>
>     -------- Forwarded Message --------
>     Subject: 	Is it possible to train MVA again a single bkg at a time
>     Date: 	Wed, 21 Sep 2016 09:47:51 +0200
>     From: 	Betty Calpas <bet...@ce...>
>     <mailto:bet...@ce...>
>     To: 	tmv...@li...
>     <mailto:tmv...@li...>
>     <tmv...@li...>
>     <mailto:tmv...@li...>
>
>
>
>     Dear experts,
>
>     from TMVAClassification.C you train the MVA for signal
>
>     against the bkgs (bkg1+bkg2 for example), then we have a 1D output file.
>
>     I wonder if I could train in "one go" the MVA for:
>
>     bkg1 against bkg2
>
>     bkg2 against signal
>
>     such that I have a 2D ouput file then when I apply a BDT cuts_1 it kills the
>
>     bkg1, and when I apply the BDT cuts_2, it kills the bkg2?
>
>     - I guess I could do this by running 2 times the TMVAClassification.C,
>     but I wonder
>
>     if it is possible to do this in 1 time?
>
>     Regards
>
>     ------------------------------------------------------------------------------
>     _______________________________________________
>     TMVA-users mailing list
>     TMV...@li...
>     <mailto:TMV...@li...>
>     https://lists.sourceforge.net/lists/listinfo/tmva-users
>


-- 

Cheers,
Betty

Re: [TMVA-users] Fwd: Is it possible to train MVA again a single bkg at a time

[Peter S. <pet...@gm...>]

Hi,

With multiclass (for the methods which support it) the training is: each
class against the sum of all other classes. Not exactly what you want, but
maybe you can work with that.

cheers,
Peter


On Mon, Sep 26, 2016 at 9:32 AM Betty Calpas <bet...@ce...> wrote:

> Dear experts,
>
> does anyone know if this is possible?
>
> Regards
>
>
> -------- Forwarded Message --------
> Subject: Is it possible to train MVA again a single bkg at a time
> Date: Wed, 21 Sep 2016 09:47:51 +0200
> From: Betty Calpas <bet...@ce...> <bet...@ce...>
> To: tmv...@li... <tmv...@li...>
> <tmv...@li...>
>
> Dear experts,
>
> from TMVAClassification.C you train the MVA for signal
>
> against the bkgs (bkg1+bkg2 for example), then we have a 1D output file.
>
> I wonder if I could train in "one go" the MVA for:
>
> bkg1 against bkg2
>
> bkg2 against signal
>
> such that I have a 2D ouput file then when I apply a BDT cuts_1 it kills the
>
> bkg1, and when I apply the BDT cuts_2, it kills the bkg2?
>
> - I guess I could do this by running 2 times the TMVAClassification.C,
> but I wonder
>
> if it is possible to do this in 1 time?
>
> Regards
>
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> TMVA-users mailing list
> TMV...@li...
> https://lists.sourceforge.net/lists/listinfo/tmva-users
>

[TMVA-users] Fwd: Is it possible to train MVA again a single bkg at a time

[Betty C. <bet...@ce...>]

Dear experts,

does anyone know if this is possible?

Regards



-------- Forwarded Message --------
Subject: 	Is it possible to train MVA again a single bkg at a time
Date: 	Wed, 21 Sep 2016 09:47:51 +0200
From: 	Betty Calpas <bet...@ce...>
To: 	tmv...@li... <tmv...@li...>



Dear experts,

from TMVAClassification.C you train the MVA for signal

against the bkgs (bkg1+bkg2 for example), then we have a 1D output file.

I wonder if I could train in "one go" the MVA for:

bkg1 against bkg2

bkg2 against signal

such that I have a 2D ouput file then when I apply a BDT cuts_1 it kills the

bkg1, and when I apply the BDT cuts_2, it kills the bkg2?

- I guess I could do this by running 2 times the TMVAClassification.C,
but I wonder

if it is possible to do this in 1 time?

Regards

Re: [TMVA-users] run mvaeffs.cxx with my predefine number of event.

[Betty C. <bet...@ce...>]

Dear Helge,

ok, thank you for your answer.

Regards

On 9/22/16 6:47 PM, Helge Voss wrote:
> Hi Betty,
>
> that has not been forseen ... I guess you'd have to 'copy' the macro
> code and write something customized for your needs.
>
> Cheers,
>
> Hekge
>
>
> On 22 September 2016 at 16:51, Betty Calpas <bet...@ce...> wrote:
>> Dear experts,
>>
>> I want to plot the figure of merit (FOM) vs the BDT cuts for number of
>> signal and bkg. TMVA::Gui does that and mvaeffs allows you to change the
>> FOM definition, but I wonder how I can enter the number of signal and
>> bkg, not via the screen that pops up, but in the code itself?
>>
>> Regards
>>
>>
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> TMVA-users mailing list
>> TMV...@li...
>> https://lists.sourceforge.net/lists/listinfo/tmva-users


-- 

Cheers,
Betty

Re: [TMVA-users] run mvaeffs.cxx with my predefine number of event.

[Betty C. <bet...@ce...>]

Dear Helge,

ok, thank you for your answer.

Regards

On 9/22/16 6:47 PM, Helge Voss wrote:
> Hi Betty,
>
> that has not been forseen ... I guess you'd have to 'copy' the macro
> code and write something customized for your needs.
>
> Cheers,
>
> Hekge
>
>
> On 22 September 2016 at 16:51, Betty Calpas <bet...@ce...> wrote:
>> Dear experts,
>>
>> I want to plot the figure of merit (FOM) vs the BDT cuts for number of
>> signal and bkg. TMVA::Gui does that and mvaeffs allows you to change the
>> FOM definition, but I wonder how I can enter the number of signal and
>> bkg, not via the screen that pops up, but in the code itself?
>>
>> Regards
>>
>>
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> TMVA-users mailing list
>> TMV...@li...
>> https://lists.sourceforge.net/lists/listinfo/tmva-users


-- 

Cheers,
Betty

Re: [TMVA-users] run mvaeffs.cxx with my predefine number of event.

[Helge V. <Hel...@ce...>]

Hi Betty,

that has not been forseen ... I guess you'd have to 'copy' the macro
code and write something customized for your needs.

Cheers,

Hekge


On 22 September 2016 at 16:51, Betty Calpas <bet...@ce...> wrote:
> Dear experts,
>
> I want to plot the figure of merit (FOM) vs the BDT cuts for number of
> signal and bkg. TMVA::Gui does that and mvaeffs allows you to change the
> FOM definition, but I wonder how I can enter the number of signal and
> bkg, not via the screen that pops up, but in the code itself?
>
> Regards
>
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> TMVA-users mailing list
> TMV...@li...
> https://lists.sourceforge.net/lists/listinfo/tmva-users

[TMVA-users] run mvaeffs.cxx with my predefine number of event.

[Betty C. <bet...@ce...>]

Dear experts,

I want to plot the figure of merit (FOM) vs the BDT cuts for number of 
signal and bkg. TMVA::Gui does that and mvaeffs allows you to change the 
FOM definition, but I wonder how I can enter the number of signal and 
bkg, not via the screen that pops up, but in the code itself?

Regards

[TMVA-users] Is it possible to train MVA again a single bkg at a time

[Betty C. <bet...@ce...>]

Dear experts,

from TMVAClassification.C you train the MVA for signal

against the bkgs (bkg1+bkg2 for example), then we have a 1D output file.

I wonder if I could train in "one go" the MVA for:

bkg1 against bkg2

bkg2 against signal

such that I have a 2D ouput file then when I apply a BDT cuts_1 it kills the

bkg1, and when I apply the BDT cuts_2, it kills the bkg2?

- I guess I could do this by running 2 times the TMVAClassification.C, 
but I wonder

if it is possible to do this in 1 time?

Regards

Re: [TMVA-users] TMVA factory weight

[Betty C. <bet...@ce...>]

Dear Helge,

ok, thank you for clarifying.

Regards

On 9/20/16 2:02 PM, Helge Voss wrote:
> Hi Betty,
>
> these two weight are in the end treated the same, but
>
>> line 223 for the methods AddSignalTree(trees, weight) I wonder what "weight"
> applies the same weight to all event in the tree (typically used to
> reweight to same 'luminosity' or equal number of training events in
> signal and background
>
> while
>
>
>> method SetBackgroundWeightExpression(weight), what is this new weight?
> allows you to set an event weight individually for each event.
> "weight" here is an expression/variable in your training tree.
>
> These are tyipcally used if you have have a Monte Carlo generator that
> produces 'weighted events' ...
>
> If both weights are present ... they are in the end multiplied ...
>
> Cheers,
> Helge
>
>
>
>> - So are these weight different? if yes they refers to what? if not, should
>> we apply the weight only in one of these methods?
>>
>> Regards
>>
>>
>> ------------------------------------------------------------------------------
>>
>> _______________________________________________
>> TMVA-users mailing list
>> TMV...@li...
>> https://lists.sourceforge.net/lists/listinfo/tmva-users
>>


-- 

Cheers,
Betty

[TMVA-users] undefined reference to `TMVA::Tools::Instance()'

[Betty C. <bet...@ce...>]

Dear experts,

I want to compile the TMVAClassification() in my main.cc[1] using a 
Makefile [2] but I git the error message [3]. Is some TMVA library 
missing in my Makefile?

Regards

[1]
http://calpas.web.cern.ch/calpas/main.cc

[2]
http://calpas.web.cern.ch/calpas/Makefile

[3]
In function `TMVAClassification(TString)':
main.cc:(.text+0x979): undefined reference to `TMVA::Tools::Instance()'
collect2: error: ld returned 1 exit status
Makefile:10: recipe for target 'do.exe' failed
make: *** [do.exe] Error 1