Ninja

• Ninja - version 1.2.0
• Author: Tiziano Peraro

Ninja implements the Laurent series expansion method for the computation of one-loop integrals.

It is based on:

• P. Mastrolia, E. Mirabella and T. Peraro, "Integrand reduction of one-loop scattering amplitudes through Laurent series expansion," JHEP 1206 (2012) 095 arXiv:1203.0291 [hep-ph].

• T. Peraro, "Ninja: Automated Integrand Reduction via Laurent Expansion for One-Loop Amplitudes," Comput. Phys. Commun. 185 (2014) 2771 arXiv:1403.1229 [hep-ph]

Installing Ninja

Install from source

Ninja can be compiled and installed with the commands

./configure
make
make install

In order to compile and link the built-in interface to the OneLoop library use --with-avholo[=FLAGS] during configuration e.g.

./configure --with-avholo
make
make install

if OneLoop is installed in a standard directory, or

./configure --with-avholo='-L/path/to/avh_olo/lib -lavh_olo' \
    FCINCLUDE=-I/path/to/avh_olo/include
make
make install

if it is installed in /path/to/avh_olo/lib and the corresponding Fortran90 module avh_olo is installed in /path/to/avh_olo/include.

In order to compile the built-in interface to the LoopTools library (requires LoopTools-2.9 or later) use --with-looptools[=FLAGS] during configuration e.g.

./configure \
    --with-looptools='fldflags -L/path/to/looptools/lib -looptools' \
    CPPFLAGS=-I/path/to/looptools/include
make
make install

where /path/to/looptools/ is the directory where LooTools is installed and fldflags stands for the additional flags the library requires for static linking (you can find them e.g. in the file bin/fcc in the main installation directory of LoopTools). You can omit the FLAGS argument of --with-looptools if you do not wish to compile the examples.

For a full list of available options type

./configure --help

See the file INSTALL for more information about the installation procedure.

If you specified the --prefix option for the configure scrip, you should also add the directory where the library is installed to your LD_LIBRARY_PATH (and DYLD_LIBRARY_PATH on Mac OS) environment variable, if not already present. This is typically a sub-directory lib/ or lib64/ of the one you specified with --prefix.

The examples which come with the distribution can be compiled, after installation, using the command

make examples

either in the top directory or in the examples/ directory. Examples using POSIX threads can be compiled (if supported) with

make thread-examples

Install from conda-forge

Ninja is packaged and distributed on conda-forge for the following platforms:

To install and add Ninja to a project with pixi, from the project directory run

pixi add ninja-hep-ph

and to install into a particular conda environment with conda, in the activated environment run

conda install --channel conda-forge ninja-hep-ph

Installing the package NinjaNumGen

The Python package NinjaNumGen can be used both as a Python module and as a script in order to generate the input needed for Ninja. It can be installed using the command

python setup.py install

in the utils/ directory. The command

python setup.py --help

will give a list of installation options. If you specified the --prefix option during installation, update your PATH and PYTHONPATH environment variables accordingly.

The package NinjaNumGen requires Form-4 or later.