NNPDF · felixhekhorn · Jan 26, 2024 · Jan 18, 2024 · Jan 18, 2024 · Jan 19, 2024
diff --git a/docs/source/dev-tools/bench-coverage.csv b/docs/source/dev-tools/bench-coverage.csv
@@ -1,7 +1,7 @@
-"feature \\ reference"    APFEL                 QCDNUM                  xspace-bench
-projectiles               |T|                   ""                      |T|
-EM                        |T|                   |T|                     |T|
-NC                        |T|                   |T|                     |T|
-CC                        |T|                   ""                      |T|
-scale-variatons           |T|                   |T|                     ""
-target-mass-corrections   |T|                   ""                      ""
+"feature \\ reference"    APFEL                 QCDNUM                  xspace-bench                  APFEL++
+projectiles               |T|                   ""                      |T|                           |T|
+EM                        |T|                   |T|                     |T|                           |T|
+NC                        |T|                   |T|                     |T|                           |T|
+CC                        |T|                   ""                      |T|                           |T|
+scale-variatons           |T|                   |T|                     ""                            |T|
+target-mass-corrections   |T|                   ""                      ""                            |T|
diff --git a/docs/source/dev-tools/bench-fns.csv b/docs/source/dev-tools/bench-fns.csv
@@ -1,4 +1,4 @@
-"|FNS| \\ reference" APFEL                QCDNUM                    xspace-bench
-|FFNS|               |T|                  |T|                       "|T| [#f1]_"
-|ZM-VFNS|            |T|                  |T|                       |T|
-FONLL                "|T| [#f2]_"         ""                        "|T| [#f2]_"
+"|FNS| \\ reference" APFEL                QCDNUM                    xspace-bench                  APFEL++
+|FFNS|               |T|                  |T|                       "|T| [#f1]_"                  |T|
+|ZM-VFNS|            |T|                  |T|                       |T|                           |T|
+|FFN0|                |T|                  ""                        |T|                         |T|
diff --git a/docs/source/dev-tools/benchmarks.rst b/docs/source/dev-tools/benchmarks.rst
@@ -30,7 +30,6 @@ also incoming projectiles, different type of currents, SV and TMC.
    :align: center
 
 .. [#f1] Only for NfFF=3
-.. [#f2] Only for charm threshold (FFNS3 to FFNS4 interpolation)
 
 .. csv-table:: Banchmarks coverage
    :file: ./bench-coverage.csv
@@ -131,5 +130,23 @@ In |xspace-bench| |SF| are defined as follows:
 
 Given these definitions, benchmarks with `yadism` are possible only in the region
 :math:`m^2_{charm} < Q^2 < m^2_{bottom}` selecting  either |ZM-VFNS| with :math:`F_X^{total}`
-or |FFNS| with NfFF=3 and FONLL with NfFF=4.
+or |FFNS| with NfFF=3.
 FONLL is implemented in the so called `scheme A` with and without damping factor.
+
+APFEL++
+-------
+
+
+|APFEL++| is a C++ rewriting of the Fortran 77 evolution code |APFEL|. However, |APFEL++| is based on a completely new code design and
+guarantees a better performance along with a more optimal memory management.
+|APFEL++| is suitable for a wide range of tasks: from the solution of the DGLAP evolution equations to the computation of DIS
+and single-inclusive-annihilation cross sections.
+
+|APFEL++| is also currently interfaced to PARTONS, a software dedicated to the phenomenology of Generalised Parton Distributions (GPDs) and TMDs,
+and to xFitter.
+
+
+SF in APFEL++
+~~~~~~~~~~~~~
+
+The |APFEL++| definitions follow the ones of |APFEL|
diff --git a/docs/source/dev-tools/metrics.rst b/docs/source/dev-tools/metrics.rst
diff --git a/docs/source/implementation/ESF.rst b/docs/source/implementation/ESF.rst
@@ -20,7 +20,7 @@ belong to a :class:`StructureFunction` instance of the respective kind.
 
 The kinds are defined by choosing one option for each of the following classes:
 
-- ``F2`` or ``FL``
+- ``F2``, ``FL``, ``F3``, ``g1``, `gL``, `g4``
 - ``light``, ``charm``, ``bottom``, ``top``, or ``total``
 
 Sometimes an instance of a different kind it is needed to complete the

diff --git a/docs/source/implementation/input.rst b/docs/source/implementation/input.rst
diff --git a/docs/source/implementation/kernels.rst b/docs/source/implementation/kernels.rst
@@ -16,8 +16,8 @@ The actual physical combination are not as many as flavors, indeed |QCD| is
 flavor blind (in the massless case, and the only difference in the massive case
 are the value of the masses themselves).
 
-In particular there are 3 relevant combinations, as described in
-:doc:`../theory/coeff-funcs`.
+In particular there are 3 relevant combinations, for polarized and 
+unpolarized |DIS| as described in :doc:`../theory/coeff-funcs`.
 
 .. todo::
 
@@ -37,11 +37,8 @@ In particular there are 3 relevant combinations, as described in
      (partons, operator, operator-error) by convolutions
 
 
-Rearrange Singlet & Non-singlet
--------------------------------
+Singlet & Non-singlet
+---------------------
 
-.. todo::
-
-   describe redefinition of the singlet -> pure singlet from Vogt notation
-
-   essentially include the already written note
+For the distinction about singlet and non-singlet coefficient function, 
+see :doc:`../theory/nonsinglet`.
diff --git a/docs/source/implementation/runner.rst b/docs/source/implementation/runner.rst
@@ -59,10 +59,6 @@ Some **handlers objects** are used to dispatch some isolated services. They are
 mostly not defined internally in :mod:`yadism` package, but mainly imported from
 :mod:`eko`.
 
-Another common trait that characterize these handlers is the presence of a
-method :func:`from_dict`, with which the object can be created loading the
-required options directly from a suitable dictionary (that in a :class:`Runner`
-object is always either `theory` or `observable` dictionary).
 
 Interpolator Dispatcher
 ~~~~~~~~~~~~~~~~~~~~~~~

diff --git a/docs/source/implementation/structure.rst b/docs/source/implementation/structure.rst
@@ -29,30 +29,28 @@ Essentially the flow of an execution is the following:
 1. (**user** initiated) a :class:`~yadism.runner.Runner` is instantiated and it
    is passed the theory configuration, and the requested observables to compute
    (together with related configurations)
-2. a check is performed on the user input (by an
-   :class:`~yadism.input.inspector.Inspector`)
-3. the relevant global *service providers* are initialized and stored by the
+2. the relevant global *service providers* are initialized and stored by the
    :class:`~yadism.runner.Runner` (like the :math:`\alpha_s` evolution, or the
    interpolation dispatcher, or the couplings computer)
-4. the requested observables are scanned, and they are assigned to the
+3. the requested observables are scanned, and they are assigned to the
    respective :class:`~yadism.sf.StructureFunction` /
    :class:`~yadism.xs.CrossSection` (acting as manager and caching storage)
    according to their kind and heavyness (but multiple kinematics will belong to
    the same :class:`~yadism.sf.StructureFunction` /
    :class:`~yadism.xs.CrossSection`) each kinematic point will correspond to an instance of
    :class:`~yadism.esf.esf.EvaluatedStructureFunction` / :class:`~yadism.esf.exs.EvaluatedCrossSection`
 
-5. (**user** initiated) output is requested
-6. the request is propagated to the managers, and then to all the required
+4. (**user** initiated) output is requested
+5. the request is propagated to the managers, and then to all the required
    `ESF` objects
-7. the `ESF` issues a request to the
+6. the `ESF` issues a request to the
    :class:`~yadism.coefficient_functions.Combiner` for the relevant
    :class:`~yadism.coefficient_functions.kernels.Kernel`
-8. all the :class:`~yadism.coefficient_functions.kernels.Kernel` are numerically
+7. all the :class:`~yadism.coefficient_functions.kernels.Kernel` are numerically
    convoluted with the |PDF| interpolation polynomials
-9. all the results are collected in an :class:`~yadism.output.Output` object and
+8. all the results are collected in an :class:`~yadism.output.Output` object and
    returned to the user
-10. (**user** initiated) the :class:`~yadism.output.Output` object might be
+9. (**user** initiated) the :class:`~yadism.output.Output` object might be
     dumped on disk in one of the available formats
 
 Elements
@@ -62,7 +60,6 @@ Elements
    :maxdepth: 2
 
    runner
-   input
    kernels
    SF
    ESF

diff --git a/docs/source/index.rst b/docs/source/index.rst
@@ -52,7 +52,7 @@ In particular:
    :hidden:
 
    overview/features.rst
-   overview/tutorials/index.rst
+   overview/tutorials/tutorials.rst
 
 .. toctree::
    :maxdepth: 1
@@ -84,9 +84,8 @@ In particular:
    :caption: Dev Tools
    :hidden:
 
-   dev-tools/tests.rst
+   dev-tools/benchmarks.rst
    dev-tools/yadmark.rst
-   dev-tools/metrics.rst
    dev-tools/code-todos.rst
 
 .. toctree::

diff --git a/docs/source/overview/features.rst b/docs/source/overview/features.rst
@@ -28,7 +28,7 @@ according to their :ref:`heavyness <theory/intro:Heavyness>` (e.g.
 
 Moreover the following features are all available:
 
-1. :doc:`flavor number schemes <../theory/fns>`: |FFNS|, |ZM-VFNS|, |FONLL|
+1. :doc:`flavor number schemes <../theory/fns>`: |FFNS|, |ZM-VFNS|, |FFN0|
 2. :doc:`scale variations <../theory/scale-variations>`: independent variations
    of factorization scale :math:`\mu_F` and renormalization scale :math:`\mu_R`
 3. :ref:`target mass corrections <theory/misc:target mass corrections>`

diff --git a/docs/source/overview/tutorials/.gitignore b/docs/source/overview/tutorials/.gitignore
@@ -0,0 +1,2 @@
+*.pineappl.lz4
+*.dat
diff --git a/docs/source/overview/tutorials/benchmarks.ipynb b/docs/source/overview/tutorials/benchmarks.ipynb