Header only framework for data analysis in massively parallel platforms.
hydra::Decays
have became range-idiomatic. In particular, the Unweight(...) methods now return a range representing sub-set of the stored sample in hydra::Decays.Minor corrections on code and version macros.
hydra::convolute
and hydra::ConvolutionFunctor
hydra::spiline
and hydra::SpilineFunctor
New functors and implementations (hydra/functions
):
Orthogonal polynomials (hydra/functions/Math.h
)
New Parameter::Create("name")
method overload.
Wrappers around thrust algorithms using range semantics:
hydra/Algorithms.h
Predefined ranges:
hydra::random_gauss_range(...)
,hydra::random_exp_range(...)
,hydra::random_flat_range(...)
,hydra::range(...)
hydra::constant_range(...)
hydra::phase_space_range(...)
Collecting range: hydra::collect
to reorder a range of values according to a indexing scheme.
Introduction of hydra::make_loglikelihood_fcn
overloads supporting range semantics.
Introduction of hydra::make_loglikelihood_fcn
overloads for binned datasets.
Implementation of hydra::begin
, hydra::end
, hydra::begin
, hydra::rend
free functions.
Range semantics for hydra::PhaseSpace.
Many other performance and interface additions and improvements accross the tree.
multiarray::insert(...)
multivector::insert(...)
hydra::reduce
hydra::make_loglikelihood_fcn
overloads supporting weighted datasets.Release notes of the version v2.1.0
This is a major release of Hydra. Many new features was added and large parts of the framework have been rewritten. The main changes are summarized below.
New features:
__hydra_host__
, __hydra_device__
, __hydra_dual__
OMP
, TBB
, CPP
and CUDA
.hydra::omp::sys
, hydra::cpp::sys
, hydra::cuda::sys
, hydra::tbb::sys
.hydra::multiarray
and hydra::multivector
.hydra::Decays
and hydra::Chains
.Hydra based scripts can now be compiled and invoked from inside ROOT for execution of algorithms in parallel using TBB and CPP backends.
Many other minor infrastructure improvements, bug fixes and features.
Documentation: