Merge branch 'develop'

.gitignore

@@ -29,6 +29,11 @@ smodels/lib/nllfast/nllfast-1.2/old/
 smodels/lib/nllfast/nllfast-2.1/fort.21
 smodels/lib/pythia6/fort.61
 smodels/lib/pythia6/pythia_card.dat
+smodels/lib/resummino/resummino-*/
+smodels/lib/resummino/resummino-*.zip
+smodels/lib/resummino/lhapdf/
+smodels/lib/resummino/LHAPDF-*/
+smodels/lib/resummino/resummino_install/
 debian/files
 debian/python-module-stampdir/
 debian/smodels.debhelper.log
@@ -53,6 +58,7 @@ docs/manual/source/recipes/compareUL.py
 docs/manual/source/recipes/runAsLibrary.py
 docs/manual/source/recipes/runWithParameterFile.py
 docs/manual/source/recipes/compute_likelihood.py
+docs/manual/source/recipes/lheLLPExample.py
 docs/manual/source/recipes/lo_xsecs_from_slha.py
 docs/manual/source/recipes/load_database.py
 docs/manual/source/recipes/lookup_efficiency.py
@@ -75,6 +81,7 @@ docs/manual/source/RunSModelS.rst
 docs/manual/source/SlhaChecker.rst
 docs/manual/source/ToolBox.rst
 docs/manual/source/XSecComputer.rst
+docs/manual/source/XSecResummino.rst
 docs/manual/source/FixPermissions.rst
 docs/manual/source/InteractivePlots.rst
 docs/manual/source/images/iplots_parameters.py

.readthedocs.yml

@@ -1,8 +1,15 @@
 # .readthedocs.yml
 
+version: 2
+
 build:
-  image: latest
+  os: ubuntu-22.04
+  tools:
+    python: "3.11"
+
+sphinx:
+  configuration: docs/manual/source/conf.py
 
 python:
-  version: 3.6
-  setup_py_install: true
+  install:
+  - requirements: docs/manual/requirements.txt

Makefile

@@ -3,7 +3,7 @@ VER=$(shell cat smodels/version)
 HAS_FC := $(shell smodels/lib/check_fortran_compiler.sh 2> /dev/null)
 HAS_CXX := $(shell command -v $(CXX) 2> /dev/null)
 
-all: resolve_deps externaltools
+all: resolve_deps # make all just resolves dependencies
 
 check_compilers: .PHONY
 ifndef HAS_FC
@@ -17,14 +17,16 @@ resolve_deps: ## resolve the deps via pip
 	@echo "try to resolve the python dependencies via pip"
 	smodels/installation.py -R
 
-smodels: all tidy
+smodels_externaltools: resolve_deps externaltools
 	@echo
 	@echo "done. you can now run the software directly from this source directory.\n"
 	@echo "Try e.g. \n\n ./runSModelS.py --help\n"
 	@echo "The latest SModelS documentation can be found at: http://smodels.readthedocs.io/en/latest/"
 	@echo "For this version documentation go to: https://smodels.readthedocs.io/en/v$(VER)"
 
-smodels_noexternaltools: resolve_deps tidy
+smodels: resolve_deps
+
+smodels_noexternaltools: resolve_deps
 	@echo
 	@echo "done. you can now run the software directly from this source directory.\n"
 	@echo "Try e.g. \n\n ./runSModelS.py --help\n"
@@ -47,6 +49,12 @@ pythia6:
 pythia8:
 	cd smodels/lib && make pythia8
 
+resummino:
+	cd smodels/lib && make resummino
+
+nllfast:
+	cd smodels/lib && make nllfast
+
 cpp: .PHONY
 	cd cpp && make
 
@@ -60,13 +68,13 @@ buildrpm:
 builddeb: buildrpm
 	cd dist && fakeroot alien smodels-$(VER)-1.x86_64.rpm
 
-pypi:
+pypi: clean
 	## pypi user is walten, repository is https://upload.pypi.org/legacy/
 	rm -rf dist
 	python3 setup.py sdist bdist_wheel
 	twine upload dist/smodels-*.tar.gz
 
-testpypi: 
+testpypi: clean
 	## testpypi user is smodels, repository is https://test.pypi.org/legacy/
 	# to install from testpypi: 
 	# pip3 install --user --upgrade --index-url https://test.pypi.org/simple/ smodels

ReleaseNotes

@@ -1,3 +1,12 @@
+Release v2.3.3, Tue 19 Dec 2023
+=======================================================
+
+  * added resummino cross section computer
+  * fixed bug in computation of error on muhat, for pyhf likelihoods
+    (affects mostly only numpy backend)
+  * small change in initialisation of gradient descent method for computation of
+    combined mu_hat, to increate robustness of method
+
 Release v2.3.2, Wed 30 Aug 2023
 =======================================================
 

docs/manual/requirements.txt

@@ -1,11 +1,13 @@
 ###### pythonic requirements for readthedocs.io ######
 jupyter
 nbconvert
-unum
-pyslha
-numpy
-scipy
 sphinx_rtd_theme
-plotly
-pyhf
-torch
+docutils==0.16
+numpy>=1.22.0
+scipy>=1.0.0
+unum>=4.0.0
+requests>=2.0.0
+pyslha>=3.1.0
+pyhf>=0.6.1
+jsonpatch>=1.25
+jsonschema>=3.2.0

docs/manual/rstFromHelpFiles.py

@@ -63,4 +63,5 @@ def runSModelS ():
 run ( "interactive-plots", "InteractivePlots" )
 run ( "toolbox", "ToolBox" )
 run ( "fixpermissions", "FixPermissions" )
+run ( "xsecresummino", "XSecResummino" )
 runSModelS()

docs/manual/source/Installation.rst

@@ -19,15 +19,16 @@ SModelS is a Python library that requires Python version 3.6 or later. It depend
 
 .. include:: dependencies.rst
 
-For speed reasons, we moreover recommend pytorch>=1.8.0 as backend for pyhf. This is, however, optional: if pytorch is not available, SModelS will use the default backend. 
+For performance reasons, we moreover recommend pytorch>=1.8.0 as backend for pyhf. This is, however, optional: if pytorch is not available, SModelS will use the default backend. 
 
-In addition, the :ref:`cross section computer <xsecCalc>` provided by :ref:`smodelsTools.py <smodelsTools>`
-requires:
+In addition, the :ref:`cross section computers <xsecCalc>` provided by :ref:`smodelsTools.py <smodelsTools>`
+require:
 
- * `Pythia 8.3 <https://arxiv.org/abs/1410.3012>`_ (requires a C++ compiler) or `Pythia 6.4.27 <http://arxiv.org/abs/hep-ph/0603175>`_ (requires gfortran)
- * `NLL-fast <http://pauli.uni-muenster.de/~akule_01/nllwiki/index.php/NLL-fast>`_ 1.2 (7 TeV), 2.1 (8 TeV), and 3.1 (13 TeV) (requires a fortran compiler)
+ * `Pythia 8.3 <https://arxiv.org/abs/1410.3012>`_ (needs a C++ compiler) or `Pythia 6.4.27 <http://arxiv.org/abs/hep-ph/0603175>`_ (needs gfortran)
+ * `NLL-fast <http://pauli.uni-muenster.de/~akule_01/nllwiki/index.php/NLL-fast>`_ 1.2 (7 TeV), 2.1 (8 TeV), and 3.1 (13 TeV) (needs a Fortran compiler)
+ * `Resummino <https://resummino.hepforge.org>`_ (needs a C++ compiler and gfortran). Moreover, in rpm-based Linux distributions, this tool needs boost, boost-devel, gsl and gsl-devel. For deb-based distributions, libboost-dev and libgsl-dev are required.
 
-These tools need not be installed separately, as the SModelS build system takes care of that. The current default is that both Pythia6 and Pythia8 are installed together with NLLfast.
+The tools themselves, i.e. Pythia6|8, NLL-fast, and/or Resummino need not be installed separately, as the SModelS build system takes care of that (see below). SModelS however expects the tools' dependencies (boost, gsl for Resummino, as well as the compilers) to be installed by the user. 
 Finally, the :ref:`database browser <databaseBrowser>` provided by :ref:`smodelsTools.py <smodelsTools>`
 requires `IPython <https://ipython.org/>`_, while the :ref:`interactive plotter <interactivePlots>` requires `plotly <https://plot.ly/python/>`_ and `pandas <https://pandas.pydata.org/>`_. 
 
@@ -43,27 +44,30 @@ Installation Methods
 
      make smodels
 
-   in the top-level directory. The installation will remove redundant folders, install the required 
-   dependencies (using pip install) and compile Pythia and NLL-fast. 
-   If the MSSM cross section computer is not needed, one can install SModelS without Pythia and NLL-fast. To this end, run::
+   in the top-level directory. SModelS will install the required dependencies (using pip install), but none of the external  tools (Pythia6|8, NLL-fast, Resummino). 
+   If the MSSM :ref:`cross section computers <xsecCalc>` are needed, one can directly install SModelS together with its external tools. To this end, run::
 
-     make smodels_noexternaltools
+     make smodels_externaltools
  
    instead of *make smodels*. 
-   In case the Python libraries can not be successfully
-   installed, the user can install them separately using his/her preferred method. Pythia and NLL-fast can also be compiled separately
+   In case the Python libraries cannot be successfully
+   installed, the user can install them separately using his/her preferred method. Pythia, NLL-fast and Resummino can also be compiled separately
    running **make externaltools**. In case the Fortran comiler isn't found, 
    try *make FCC=<path-to-gfortran> smodels* or *make FCC=<path-to-gfortran> externaltools*. 
 
- * If Python's *setuptools* is installed in your machine, SModelS and its dependencies
-   can also be installed without the use of pip.
+ * Every external tool can also be compiled individually, run e.g.::
+
+     make pythia6 pythia8 nllfast resummino
+
+   Remember, though, that the compilers as well as Resummino's dependencies (boost, gsl, see above) need to be installed already.
+
+ * Python's *setuptools*, if installed in your machine, can also be used for installing SModelS and its dependencies.
    After downloading the source from the `SModelS releases page <https://github.com/SModelS/smodels/releases>`_
    and extracting it, run::
 
-
      setup.py install
 
-   within the main smodels directory. If the python libraries are installed in a system folder (as is the default behavior),
+   within the main SModelS directory. If the python libraries are installed in a system folder (as is the default behavior),
    it will be necessary to run the install command with superuser privilege.
    Alternatively, one can run setup.py with the "--user" flag: ::
 
@@ -73,16 +77,16 @@ Installation Methods
    manually and then rerun setup.py.
    For Ubuntu, SL6 machines and other platforms, a recipe is given below.
 
-
-   Note that this installation method will install smodels into the default system or user directory (e.g. ~/.local/lib/python3/site-packages/).
+   Note that this installation method will install SModelS into the default system or user directory (e.g. ~/.local/lib/python3.10/site-packages/).
    Depending on your platform, the environment variables $PATH, $PYTHONPATH, $LD_LIBRARY_PATH
    (or $DYLD_LIBRARY_PATH) might have to be set appropriately.
 
+   Note also, that setup.py will *not* attempt at downloading and compiling the external tools (Pythia6|8, NLL-fast, Resummino) at install time. 
+   Instead, this will be done on the fly, at runtime, upon call of the :ref:`cross section computer(s) <xsecCalc>`. 
+   The external tools will also be located in the above smodels installation directory (<installdir>/lib/...).  
 
 
-
- * Finally, SModelS is `indexed on pypi <https://pypi.org/project/smodels/>`_. Thus, if *pip3* (or *pip*) is installed in your machine, it is also possible to install SModelS directly without the need for
-   downloading the source code: ::
+ * Finally, SModelS is `indexed on pypi <https://pypi.org/project/smodels/>`_. Thus, if *pip3* (or *pip*) is installed in your machine, it is possible to install SModelS  without downloading the source code: ::
 
      pip3 install smodels
 
@@ -92,14 +96,17 @@ Installation Methods
    
    for user-specific installations.
 
-
-   Note that this installation method will install smodels into the default system or user directory (e.g. ~/.local/lib/python3/site-packages/).
+   This installation method will install SModelS into the default system or user directory (e.g. ~/.local/lib/python3.10/site-packages/).
    Depending on your platform, the environment variables $PATH, $PYTHONPATH, $LD_LIBRARY_PATH
    (or $DYLD_LIBRARY_PATH) might have to be set appropriately.
    Be aware that the example files and the |parameters| discussed in the manual 
    will also be located in your default system or user directory. Furthermore the database
-   folder is not included (see :ref:`database installation <installingDB>` below).
-   This installation method is best suited for experienced python users.   
+   folder is not included (see :ref:`database installation <installingDB>` below). 
+
+   Moreover, pip will *not* attempt at downloading and compiling the external tools (Pythia6|8, NLL-fast, Resummino) at install time. Instead, this will be done on the fly, at runtime, upon call of the :ref:`cross section computer(s) <xsecCalc>`. 
+   The external tools will also be located in the above smodels installation directory (<installdir>/lib/...).  
+
+   Generally, this installation method is best suited for experienced python users.   
 
 
 There is also a diagnostic tool available: ::

docs/manual/source/ReleaseUpdate.rst

@@ -37,6 +37,15 @@ What's New
 ==========
 The major novelties of all releases since v1.0 are as follows:
 
+New in Version 2.3.3:
+^^^^^^^^^^^^^^^^^^^^^
+
+  * added :ref:`resummino cross section <xsecResummino>` computer
+  * fixed bug in computation of error on muhat, for `pyhf likelihoods <tools.html#tools.pyhfInterface.PyhfUpperLimitComputer.lmax>`_
+    (affects mostly only numpy backend)
+  * small change in initialisation of gradient descent method for computation of
+    `combined mu_hat <tools.html#tools.analysesCombinations.AnaCombLikelihoodComputer.lmax>`_, to increate robustness of method
+
 New in Version 2.3.2:
 ^^^^^^^^^^^^^^^^^^^^^
 

docs/manual/source/SModelSTools.rst

@@ -29,8 +29,8 @@ SModelS Tools
 
 Inside SModelS there is a number of tools that may be convenient for the user:
 
-* a :ref:`cross section calculator <xsecCalc>` based on `Pythia8 <http://pythia.org>`_ (or `Pythia6 <http://pythia6.hepforge.org>`_) and
-  `NLLfast <http://pauli.uni-muenster.de/~akule_01/nllwiki/index.php/NLL-fast>`_,
+* a :ref:`cross section calculator <xsecCalc>` based on `Pythia8 <http://pythia.org>`_ (or `Pythia6 <http://pythia6.hepforge.org>`_),
+  `NLLfast <http://pauli.uni-muenster.de/~akule_01/nllwiki/index.php/NLL-fast>`_, and one based on `Resummino <https://resummino.hepforge.org>`_,
 * :ref:`SLHA and LHE file checkers <fileChecks>` to check your input files for completeness and sanity,
 * a :ref:`database browser <databaseBrowser>` to provide easy access to the |database| of experimental results,
 * a plotting tool to make :ref:`interactive plots <interactivePlots>` based on `plotly <https://plot.ly/python/>`_ (v1.1.3 onwards),
@@ -39,15 +39,19 @@ Inside SModelS there is a number of tools that may be convenient for the user:
 
 .. _xsecCalc:
 
-Cross Section Calculator
-------------------------
+Cross Section Calculators
+-------------------------
 
-This tool computes LHC production cross sections for *MSSM particles*
-and writes them out in :ref:`SLHA convention <xsecblock>`. This can in particular be
+These tools compute LHC production cross sections for *MSSM particles*
+and write them out in :ref:`SLHA convention <xsecblock>`. This is particularly 
 convenient for adding cross sections to SLHA input files, see :doc:`Basic Input <BasicInput>`.
-The calculation is done at LO with `Pythia8 <http://pythia.org>`_ or `Pythia6.4 <http://pythia6.hepforge.org>`_ ; K-factors
-for colored particles are computed with `NLLfast <http://pauli.uni-muenster.de/~akule_01/nllwiki/index.php/NLL-fast>`_. Signal strength multipliers can optionally be supplied for each "mother" particle. However, use at your own risk! Make sure the
-output is sensible and contains all cross sections for all production mechanisms
+
+
+Pythia and NLLfast Cross Sections
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+In the basic "xseccomputer", the calculation is done at LO with `Pythia8 <http://pythia.org>`_ or `Pythia6.4 <http://pythia6.hepforge.org>`_ ; K-factors
+for colored particles are computed with `NLLfast <http://pauli.uni-muenster.de/~akule_01/nllwiki/index.php/NLL-fast>`_. Signal strength multipliers can optionally be supplied for each "mother" particle. However, use at your own risk! Make sure the output is sensible and contains all cross sections for all production mechanisms
 you are interested in! 
 
 **The usage of the cross section calculator is:**
@@ -83,8 +87,33 @@ This will compute 8 TeV and 13 TeV LO cross sections as above, but the cross sec
 Note that signal strength multipliers get applied only to LO cross sections. This means they are propagated to NLO and NLL level only if the LO cross sections are computed first and the NLO/NLL corrections added afterwards. In other words, if the xseccomputer is called with -n or -N argument but without -O (--LOfromSLHA), the --ssmultipliers argument will be ignored.
 
 
-* **The cross section calculation is implemented by the** `xsecComputer function <tools.html#tools.xsecComputer.XSecComputer>`_
+* **The cross section calculation is implemented by the** `XSecComputer class <tools.html#tools.xsecComputer.XSecComputer>`_
+
+.. _xsecResummino:
+
+Resummino Cross Sections
+^^^^^^^^^^^^^^^^^^^^^^^^
+
+For electroweak-ino and/or slepton production cross sections, another tool based on `Resummino <https://resummino.hepforge.org>`_ is available. This can calculate EW cross sections at LO, NLO, and NLL+NLO orders. No K-factors are used, all orders are calculated independently.
+
+**The usage of the Resummino cross section calculator is:**
+
+.. include:: XSecResummino.rst
+
+A typical
+usage example is: ::
+
+   smodelsTools.py xsecresummino -s 13 -p -n -f test/testFiles/resummino/resummino_example.slha -part 1000023 1000024 
+
+This will compute the (1000023,-1000024), (1000023, 1000024) and (-1000024,1000024) cross sections for sqrt(s)=13 TeV at NLO for the spectrum given in resummino_example.slha and append them to that SLHA file. 
+Additional settings, like the PDF sets to use, are taken from the Resummino configuration file, smodels/etc/resummino.py. There, also the default threshold for the minimum cross section is set.  
+
+Instead of providing a list of particles via the --part argument, one can also directly specify in the resummino.py  configuration file which production channels shall be considered. To this end one can either adapt the resummino.py file in the smodels/etc folder, or provide their own configuration file via the --conf argument. 
+Note that options set directly on the command line always take precedence over the settings in the 
+configuration file. 
+
 
+* **The resummino cross section calculation is implemented by the** `XSecResummino class <tools.html#tools.xsecResummino.XSecResummino>`_
 
 .. _fileChecks: