ostaphep / ostap Goto Github PK

BernsteinEven is broken

1. prepare .zenodo.json file
2. prepare CITATION.CFF

I have follow the direction to install ostap on my Ubuntu machine. However, when I execute

cmake .. -DCMAKE_INSTALL_PREFIX=..

I have the following error. I am wondering if you could tell me how to fix it. Thank you.

........
........
-- Configuring done (3.8s)
CMake Error at source/CMakeROOT_6_23.cmake:225 (add_dependencies):
The dependency target "ROOT::MathMore" of target "ostapDict" does not
exist.
Call Stack (most recent call first):
source/CMakeLists.txt:14 (include)

CMake Error at source/CMakeROOT_6_23.cmake:194 (target_link_libraries):
Target "ostap" links to:

ROOT::MathMore

but the target was not found. Possible reasons include:

* There is a typo in the target name.
* A find_package call is missing for an IMPORTED target.
* An ALIAS target is missing.

Call Stack (most recent call first):
source/CMakeLists.txt:14 (include)

CMake Error at source/CMakeROOT_6_23.cmake:226 (target_link_libraries):
Target "ostapDict" links to:

ROOT::MathMore

but the target was not found. Possible reasons include:

* There is a typo in the target name.
* A find_package call is missing for an IMPORTED target.
* An ALIAS target is missing.

Call Stack (most recent call first):
source/CMakeLists.txt:14 (include)

-- Generating done (0.0s)
CMake Generate step failed. Build files cannot be regenerated correctly.

Current thisostap.sh works only for bash

• versions for zsh/ksh and csh/tcsh are needed
• version for fish coulb be nice

Seems TMinuit works not correctly as it fails if one tries to follow examples. (Ostap: 1.4.5.0, Python 3.8.0 | packaged by conda-forge | (default, Nov 22 2019, 19:11:38) )

In [1]: m = ROOT.TMinuit(9)
In [2]: help(m)
In [3]: m.addPar ( 'ququ' , 10 , 0.1 )

TypeError Traceback (most recent call last)
/home/adzyuba/miniconda3/envs/ostap-env/bin/ostap in
----> 1 m.addPar ( 'ququ' , 10 , 0.1 )

/home/adzyuba/miniconda3/envs/ostap-env/lib/python3.8/site-packages/ostap/fitting/minuit.py in mn_add_par(self, name, start, step, low, high)
225 ipar = len ( self )
226 ierr = ROOT.Long ( 0 )
--> 227 self.mnparm ( ipar , name , start , step , low , high , ierr )
228 #
229 return ierr

TypeError: void TMinuit::mnparm(int k, TString cnamj, double uk, double wk, double a, double b, int& ierflg) =>
could not convert argument 7 (use ctypes.c_int for pass-by-ref of ints)

Restore the initial state for PDF.wilks* and similar methods.

It would have a sense to use parallel ipython (or ipyparallel ) in ostap, see here

Output prints " ... mean = XXX +- YYY ...", which is a bit confusing, as YYY is an individual uncertainty of single entry, but not an uncertainty of the mean, which is also random value. Better switch to ""...mean=XXX rms=YYY...".

In principle, estimates of uncertainties for mean and rms random valuers (under the Gaussian hypothesis ) are also worth to be printed.

Let me alsothsnk authors for the incredible tool!!

From Feb 21 cmp_fit does not for ROOT 6.27/1

PyCallable needs to be fixed....

Ostap::Math::PhaseSpace3 and Ostap::Math::PhaeSpace3s do not agree

import ROOT, math
from   ostap.core.pyrouts  import Ostap
from   ostap.logger.logger import getLogger

logger = getLogger ("ChicPi")

PS3 = Ostap.Math.PhaseSpace3  ## <-- HERE 
PS3 = Ostap.Math.PhaseSpace3s ## <-- HERE  
GeV = 1.0
MeV = 0.001 * GeV


m_chic1 = 3510.67 * MeV
m_K     = 493.677 * MeV
m_pi    = 139.570 * MeV
m_p     = 938.272 * MeV
m_Lb    = 5619.60 * MeV


ps_pi = PS3 ( m_chic1 , m_pi , m_p )
ps_K  = PS3 ( m_chic1 , m_K  , m_p )

logger.info ( "PS(pi)                       : %s "  % ps_pi ( m_Lb ) )
logger.info ( "PS(K)                        : %s "  % ps_K  ( m_Lb ) )
r = ps_pi ( m_Lb ) / ps_K ( m_Lb )

theta_C = 13.02 * math.pi / 180.0
tc2     = math.tan (  theta_C ) **2
logger.info ( "PS(pi)/PS(K)                 : %s "  % r    )
logger.info ( "tan^2 theta_C                : %s "  % tc2  )
logger.info ( "PS(pi)/PS(K) * tan^2 theta_C : %s "  % ( r  * tc2 ) )

Bad interference for makePlots and calls for ROC-curves for tmva/chopping
certains standard TMVA macros produced segfaults for various ROOT versions