v0.10.0

v0.10.0 (08.08.2024)

Major Changes since v0.9.0

• Automatic network generation from high-level specifications.
• Units at the user interface including scaling and conversion.
• Morphologies are loaded through a unified interface; which produces a bundle of morphology, metadata, and segment tree.

Internal Updates

• Documentation overhaul
• Python bindings offer type stubs; giving auto-completion and inline documentation in some editors/IDEs.
• Performance improvements
  • Label resolution uses hashes instead of strings, saving memory, network traffic, and time.
  • Spike delivery is up to 30% faster leading to 10% end-to-end improvements in some cases (e.g. example/brunel)
  • Load balancing is faster and doesn't rely on MPI, leading to shorter setup times for large networks.
• Fixes
  • Better support for source builds on MacOS (aarch64)
  • modcc no longer allows internal variables (v, celsius, ...) as ASSIGNED
  • Better support for raw (C++) mechanisms
• MC cells renamed cable cells.

Breaking changes

• Return values of all morphology loaders have changed.
• Raw (segment tree) loaders removed.
• Support for Python 3.8 removed.
• C++: Removed simulation::inject_events, use a generator instead.

New Contributors

• @ErbB4 made their first contribution in #2271

Full Changelog: v0.9.0...v0.10.0

v0.10.0

v0.10.0 (08.08.2024)

Major Changes since v0.9.0

• Automatic network generation from high-level specifications.
• Units at the user interface including scaling and conversion.
• Morphologies are loaded through a unified interface; which produces a bundle of morphology, metadata, and segment tree.

Internal Updates

• Documentation overhaul
• Python bindings offer type stubs; giving auto-completion and inline documentation in some editors/IDEs.
• Performance improvements
  • Label resolution uses hashes instead of strings, saving memory, network traffic, and time.
  • Spike delivery is up to 30% faster leading to 10% end-to-end improvements in some cases (e.g. example/brunel)
  • Load balancing is faster and doesn't rely on MPI, leading to shorter setup times for large networks.
• Fixes
  • Better support for source builds on MacOS (aarch64)
  • modcc no longer allows internal variables (v, celsius, ...) as ASSIGNED
  • Better support for raw (C++) mechanisms
• MC cells renamed cable cells.

Breaking changes

• Return values of all morphology loaders have changed.
• Raw (segment tree) loaders removed.
• Support for Python 3.8 removed.
• C++: Removed simulation::inject_events, use a generator instead.

New Contributors

• @ErbB4 made their first contribution in #2271

Full Changelog: v0.9.0...v0.10.0

v0.9.0

v0.9.0

** 2023 08 09 **

After much more delay than anticipated, we are very happy to present a new Arbor release. Nearly 8 months of work is in it, much of which focussed on speed, optimisation, fixes and build system changes. This release includes Python 3.12 wheels as probably one of the first packages on PyPI 😄.

Major new features

• External Connectivity: Arbor can now interface with other simulators or processes through MPI. Contexts now accept a second MPI communicator and Recipes can implement a external_connections_on method to specify how simulations might be interacting. See documentation for more.
• Arbor now supports checkpointing and resume from a previously stored checkpoint. Useful when a certain point in time needs to be explored in multiple directions, when you want to rewind to a previous state, etc. See documentation for more.
• More painted parameters are now scalable through iexpr: temperature, capacitance, resistivity, membrane potential and the following ionic parameters: internal and external concentration, diffusivity, and reversal potential. See documentation.
• A set of ANN activation functions for NMODL have been added: sigmoid(x), relu(x) and tanh(x). Control volume area is exposed through NMODL.
• A revamped backend for ARM CPU's that support Scalable Vector Extension (SVE). Arbor and modcc are now fully compatible, so users who have access to A64FX-based HPC can take full advantage of that hardware.

Breaking changes since v0.8.1

• It is no longer possible to set binning and sampling policies, due to Arbor now having a fixed timestep. Removing exact delivery increases the speed of the simulation due to elimination of small steps, makes the numerics independent of presence of sampling, and also leads to a number of code simplifications.
• Contexts are now constructed kwargs-only. So, arbor.context(12, None) is now arb.context(threads=12, gpu=None). In the case no arguments are supplied, the context initialized to default_allocation, which means it will use the total number of threads and first GPU that are locally available. arbor.simulation constructed without an explicit arbor.context also initializes a default_allocation.
• Version bumps:
  • CUDA 11 or higher is required for GPU builds.

Full commit log

Neuroscience, documentation

• Emit better warning. by @thorstenhater in #2071
• ANN activation functions for NMODL: sigmoid(x), relu(x) and tanh(x) by @llandsmeer in #2052
• 📊 Make profiler output a bit nicer. by @thorstenhater in #2078
• Merge GUI tutorial by @brenthuisman in #2088
• Busyring example by @boeschf in #2100
• Feature: External Connectivity by @thorstenhater in #2001
• FEATURE: fixed dt by @boeschf in #2053
• mention Matrix, GUI, Playground in docs by @brenthuisman in #2113
• Document presets by @thorstenhater in #2126
• Doc updates by @brenthuisman in #2119
• CV area exposed to NMODL. by @thorstenhater in #2110
• Fix enumeration in docs by @thorstenhater in #2140
• Docs: corrections. by @brenthuisman in #2135
• Fix documentation by @boeschf in #2152
• Address missing logo on all but front page by @brenthuisman in #2155
• Document features by @brenthuisman in #2158
• Make (most) painted parameters scalable. by @thorstenhater in #2141
• Add better (helpful?) errors on mechanism semantic misuse. by @thorstenhater in #2171
• Checkpoint/Resume by @thorstenhater in #2081

Core

• Better handling of powers in modcc. by @thorstenhater in #2061
• Silence spurious unused variable warnings in generated code by @bcumming in #2073
• BluePyOpt API updates by @lukasgd in #2045
• Add some missing #include directives by @musicinmybrain in #2080
• Remove redundant sample events structure. by @thorstenhater in #2079
• 🧹 Re-factor FVM lowered cell implementation and shared state by @thorstenhater in #2082
• Take less locks for shorter times when sampling. by @thorstenhater in #2093
• Fix LIF oversampling. by @thorstenhater in #2091
• 📊 Pin threads by @thorstenhater in #2094
• Add zero'ing by @thorstenhater in #2107
• Add an error on NET_RECEIVE w/ > 1 args. by @thorstenhater in #2127
• Update busyring to generate rings by @bcumming in #2144
• Apply some polish to model construction. by @thorstenhater in #2134
• SVE backend by @boeschf in #2148
• Exit sup::path stage left; enter std::filesystem::path stage right by @thorstenhater in #2165
• gtest dependency by @boeschf in #2163

Build, testing, CI

• v0.8.1 by @brenthuisman in #2072
• v0.8.1 postrelease by @brenthuisman in #2074
• Enable Spack tests by @brenthuisman in #2098
• Python/pip strikes again: execute pip install in venvs by @brenthuisman in #2111
• Scikit build core by @brenthuisman in #2121
• Spack: add dev version by @brenthuisman in #2150
• Break out pip and doc tests, exclude doc-change from eating up precious runner cycles by @brenthuisman in #2160
• rename test-pip.yml by @boeschf in #2166
• cscs-ci by @boeschf in #2149
• gtest in CI containers by @boeschf in #2170
• Spack-solvable versions, add tests to Spack workflow by @brenthuisman in #2156
• Fudge version in pyproject.toml for ci by @brenthuisman in #2200
• enable remote example by @boeschf in #2173

Fixes, optimization

• [BUGFIX] a-b-c: actually set compiler, improved default by @brenthuisman in #2051
• 🐙 Optimise PPACK by @thorstenhater in #2067
• Silence flake8 and appease black. by @thorstenhater in #2086
• BUGFIX: add missing functions to sve backend by @boeschf in #2096
• BUGFIX gpu ppack declarations by @boeschf in #2087
• BUGFIX: gpu mechanism init by @boeschf in #2095
• The Initialisation Optimisation by @thorstenhater in #2097
• Fix crash in FindId Visitor. by @thorstenhater in #2115
• Add LTO and M1 arch to CMake by @thorstenhater in #2106
• Fix crash on probing non-existent mechanism state. by @thorstenhater in #2112
• Python/concurrency defaults by @brenthuisman in #1979
• Add faster merge events. by @thorstenhater in #2108
• Fix a typo in python bindings. by @thorstenhater in #2125
• Fix incorrect mention of attribute catalogue in single_cell_model by @Helveg in #2122
• Avoid crash when integrating over empty branch. by @thorstenhater in #2124
• fix lint pipleine by @boeschf in #2151
• Ward against probing the void. by @thorstenhater in #2139
• Fix GPU/NVC problems by @thorstenhater in #2143
• fix ODR violation by @boeschf in #2154

New Contributors

• @musicinmybrain made their first contribution in #2080

Full Changelog: v0.8.1...v0.9.0

v0.8.1

v0.8.1

** 2022 12 22 **

A 🎄 holiday release! Not much has changed in a month, but we'd like to share it all the same. Notably, the Arbor GUI is co-released as of Arbor v0.8, and v0.8.1 will be no different.

Major new features

• Voltage Processes: add the VOLTAGE_PROCESS mechanism kind to modcc, allowing for direct writing to the membrane voltage (#2033)
• Spack gpu option: added conditional variant for cuda builds to enable GPU-based random number generation (#2043)
• SDE Tutorial (#2044)

Breaking changes since v0.7

• None 💃!

Bug fixed

• Fix ornstein_uhlenbeck example on gpu (#2039)
• Setting ARB_MODCC was broken and nunfunctional. Fixed. (#2029)
• The --cxx flag in arbor-build-catalogue is now properly used; falls back to c++. (#2051)

Full commit log

• Post release: add Zenodo entry, add Spack entry, update docs and scripts by @brenthuisman in #2036
• BUGFIX: add ARB_CUDA flag to example catalogue by @boeschf in #2039
• Additional builtin functions to Arbor's NMODL dialect by @boeschf in #2035
• Throw better errors when we cannot look up ion diffusivity by @thorstenhater in #2040
• ⚡ Voltage Processes by @thorstenhater in #2033
• simplify make catalogue by @boeschf in #2042
• spack gpu option by @boeschf in #2043
• Remove deprecated spike_detector. by @thorstenhater in #2041
• make ARB_MODCC functional again by @brenthuisman in #2029
• Yank v0.5 from Spack file, it does not build due to change in setting arch by @brenthuisman in #2037
• Use pugixml instead of libxml2 by @thorstenhater in #2048
• [BUGFIX] a-b-c: actually set compiler, improved default by @brenthuisman in #2051
• Sde Tutorial by @boeschf in #2044
• Set Python to known version for all CI workflows by @brenthuisman in #2058
• consistent mechanism ids by @boeschf in #2057
• Fix docs about exprelr. by @thorstenhater in #2064
• Allow __ in profiler names. by @thorstenhater in #2065
• CI fixes for wheel building, prep for musllinux, Spack by @brenthuisman in #2069

Full Changelog: v0.8...v0.8.1

v0.8

v0.8

** 2022 11 15 **

Welcome to another installment of the Arbor simulator!

In this release we add a CHANGELOG, where major new features and breaking changes will be mentioned specifically.

Major new features

• Stochastic Differential Equations. Introduces sources of white noise to arbor (and nmodl). Both point and density mechanisms may now use white noise as part of the state updates, turning the ODEs effectively into SDEs. The noise sources are provided per connection end point (point mech.) or control volume (density mech.) and are uncorrelated (spatially, temporally and across different mechanism instantiations). #1884
• Mutable connection table. Add functionality, docs, and examples on editing the connection table. This is a first, small PR
  on the topic, further functionality will come as requested. #1919
• Allow editing morphologies. Supported operations: join_at, split_at, equivalence, equality, apply isometry. #1957
• Arbor cable cell exporter and backend support in BluePyOpt. #1959
• Make LIF cells probeable. #2021

Breaking changes since v0.7:

• A change in decor API: arbor.cable_cell has the labels and decor arguments swapped. I.e.: (tree, labels, decor)
  -> (tree, decor, label). Labels are now optional. #1978
• Remove the generate-catalogue script. modcc accepts now a list of NMODL files and is able to spit out a catalogue.cpp file. #1975
• Mechanism ABI version is bumped to 0.3.1. #1884
• Rename spike detector -> threshold detector. #1976
• Remove access to time t in NMODL. #1967
• Major dependency version bumps:
  • GCC: 9 and up
  • CUDA: 11 and up
  • Clang: 10 and up

Full commit log

Neuroscience, documentation

• Convenience: shorter code to build decor. by @thorstenhater in #1903
• Fix doc error. by @thorstenhater in #1934
• probe id -> probeset id, and clarification in docs by @brenthuisman in #1898
• Elaborate on mpi4py. by @thorstenhater in #1940
• Add all_closest to place_pwlin. by @thorstenhater in #1952
• Add more visibility to decor. by @thorstenhater in #1953
• add spike source docs. by @thorstenhater in #1939
• 🦑 Excise time by @thorstenhater in #1967
• 🐍 Rename spike detector -> threshold detector by @thorstenhater in #1976
• Mutable connection table by @thorstenhater in #1919
• Allow multiple schedules per source_cell. by @thorstenhater in #1963
• Allow editing morphologies by @thorstenhater in #1957
• generate-catalogue is gone by @thorstenhater in #1975
• SDE by @boeschf in #1884
• Discuss q10 pattern in NMODL docs. by @thorstenhater in #1995

Core

• modcc: Allow redundant, but correct READ declaration. by @thorstenhater in #1936
• Deny WATCH statements. by @thorstenhater in #1942
• Excise fvm type aliases by @thorstenhater in #1938
• Add stack information to arbor_exception by @thorstenhater in #1945
• Expose find_private_gpu to python by @thorstenhater in #1943
• Make mpoints hashable. by @thorstenhater in #1950
• Arborio reads from Stream objects now. by @thorstenhater in #1937
• Remove explicit generator by @thorstenhater in #1962
• Clean up plasticity by @thorstenhater in #1985
• Heaviside step by @llandsmeer in #1989
• Do not restrict SWC record identifier by @schmitts in #1996
• Make decor mandatory and labels optional. by @thorstenhater in #1978
• Add virtual dtors to recipe components by @thorstenhater in #2000
• cleanup documentation by @boeschf in #2007
• Arbor cable cell exporter and backend support in BluePyOpt by @lukasgd in #1959
• Add a plethora of config options to a-b-c. by @thorstenhater in #1958
• Make LIF cells probeable. by @thorstenhater in #2021
• Added fixture dev docs. Made fixtures more robust. by @Helveg in #2025
• Make LIF cells LIF cells again. by @thorstenhater in #2026
• update cray documentation by @boeschf in #2022

Build, testing, CI

• Spack cache change, bump versions by @brenthuisman in #1926
• Expand docs on testing. by @thorstenhater in #1944
• Fix cmake paths so we can use Arbor as a sub-project. by @thorstenhater in #1948
• bump pybind11 for py3.11 compat by @brenthuisman in #1955
• Add spike counts to pre-commit tests. by @thorstenhater in #1965
• Move pybind11 to /ext by @brenthuisman in #1968
• Add black config to pyproject.toml. by @thorstenhater in #1971
• Bump CI workflow and GoogleTest by @boeschf in #2003
• Remove cscs-ci badges by @brenthuisman in #1999
• fix cmake: ninja + ExternalProject by @boeschf in #2008
• Use HTTPS access instead of ssh/git for gtest. by @thorstenhater in #2009
• Add memory sanitizer by @thorstenhater in #2013
• Try to use OpenMPI as of brew/apt. by @thorstenhater in #2016
• Bump versions by @thorstenhater in #2017
• Include ubenches into CI by @boeschf in #2014
• Speed up CI by @boeschf in #2019
• Update pyproject.toml for Py3.11 by @brenthuisman in #2018
• Check internal invariants in CI by @thorstenhater in #2023

Fixes, optimization

• Disable inconsistent rule. by @thorstenhater in #1928
• Action must change VERSION except on version tag. by @brenthuisman in #1931
• NMDOL: default catalogue clean up by @thorstenhater in #1935
• PANIC! Forgot to fix fvm type in GPU!!!. by @thorstenhater in #1946
• Remove abuse of arg_v. by @thorstenhater in #1954
• Catch symdiff errors. by @thorstenhater in #1964
• Bug fix: point mechs applying weights to ionic concentrations by @thorstenhater in #1960
• Fix some issues found by PVS Studio. by @thorstenhater in #1974
• load_asc Python fix by @lukasgd in #1977
• Ensure proper Pybind11 is found. by @thorstenhater in #1961
• missing std namespace by @boeschf in #1994
• Optimize CPU-side solvers by @thorstenhater in #1992
• 🦑 modcc now generates GPU mechs again. by @thorstenhater in #1988
• 🦑 Never call a procedure again. by @thorstenhater in #1972
• compiler warnings by @boeschf in #2006
• guard for zero random variables by @boeschf in #2031
• fix some compiler warning by @boeschf in #2034

New Contributors

• @lukasgd made their first contribution in #1977

Full Changelog: v0.7...v0.8-rc

v0.7

What's Changed

Neuroscience/documentation

• prep extracellular potentials tutorial. Updated corresponding example by @espenhgn in #1825
• Add documentation on faster NMODL. by @thorstenhater in #1840
• Add tstop doc in recipe API doc by @schmitts in #1852
• Use list comprehension to speed up creation of connections by @schmitts in #1864
• Support markers in neurolucida ascii files by @bcumming in #1867
• simplified create_polygon function in lfpykit example by @espenhgn in #1881
• Better key for the lfpykit intersphinx mapping by @Helveg in #1878
• Sort (in python) the mechanism names (for convenience). by @thorstenhater in #1882
• Always emit weight expression. by @thorstenhater in #1875
• Add support for epoch callbacks by @bcumming in #1873
• Add introspection for global properties. by @thorstenhater in #1890
• Add point probes to demo by @schmitts in #1891
• Tut/use avail threads by @brenthuisman in #1900
• Allen tutorial by @brenthuisman in #1781
• New policy 'round_robin_halt' by @jlubo in #1868
• Axial Diffusion by @thorstenhater in #1729
• Add some convenience to simulation creation. by @thorstenhater in #1904
• Predefine SWC Regions. by @thorstenhater in #1911
• Diffusion Example Improvements (and a bit of clean-up) by @thorstenhater in #1914
• Inhomogeneous parameters by @AdhocMan in #1887
• Fix line numbers in tutorials and assorted doc corrections by @brenthuisman in #1917

Core

• 🐍 Be more lenient when accepting args to file I/O by @thorstenhater in #1819
• modcc: generate missing node_index read needed for reading time t in the mechanisms by @noraabiakar in #1866
• Add Developer Documentation by @thorstenhater in #1639
• Isolate external catalogues from libarbor.a. by @thorstenhater in #1837

Build, testing, CI

• Build Python 3.10 binary wheels. Add v0.6 to spackfile. by @brenthuisman in #1817
• export API by @boeschf in #1824
• export doc by @boeschf in #1849
• Include CMAKE+CUDA iff NVCC is needed. by @thorstenhater in #1855
• Bit more on Spack, fix in tutorial by @brenthuisman in #1838
• json submodule added by @brenthuisman in #1871
• Fix a bug where Debian/Ubuntu's Python malfunctions by @brenthuisman in #1894
• Have dependency version policy by @brenthuisman in #1865
• random123 submodule added by @brenthuisman in #1872
• Fix tool installation paths. by @thorstenhater in #1905
• Adopt Black for Python. by @thorstenhater in #1906
• add cmake checks for non-bundled random123 by @bcumming in #1907
• Temporarily disable A64FX CI by @bcumming in #1910
• Adopt flake8 by @thorstenhater in #1908
• Move Python build to pyproject.toml, bump Python minver to 3.7, fix macos wheel generation by @brenthuisman in #1916
• Weekly CI Python wheel build pushes to Test.PyPI.org by @brenthuisman in #1921

Fixes, optimization

• Users may not give dt < 0. by @thorstenhater in #1821
• Fix ubenches compilation errors by @noraabiakar in #1828
• death test: wrong signal by @boeschf in #1858
• Make brunel.py setup faster. by @thorstenhater in #1854
• add a better error. by @thorstenhater in #1846
• Fix load_component for label-dict by @noraabiakar in #1859
• Found out the hard way that this is still needed :/ by @thorstenhater in #1860
• Elide memcpy where not needed by @thorstenhater in #1863
• Bug fix: Fix voltage vector size in threshold_watcher contstructor by @noraabiakar in #1820
• remove test statements for move ctor by @boeschf in #1899
• Code Quality: PVS Studio Finds by @thorstenhater in #1901
• Two context decomp swaps were forgotten by @brenthuisman in #1912

New Contributors

• @jlubo made their first contribution in #1868
• @AdhocMan made their first contribution in #1887

Full Changelog: v0.6...v0.7

v0.6

What's Changed

Core API:

• Add S-Exp for CV Policies by @thorstenhater in #1511
• Clean-up and extend mech cat iterator interface. by @thorstenhater in #1564
• Mech ABI: The final step. by @thorstenhater in #1452
• Add dim3 to gridDim constructor in generated mechanism code by @noraabiakar in #1630
• Added an empty catalogue by @Helveg in #1677
• Allow to limit Poisson schedule with a stop time (closes #1617) by @schmitts in #1684
• Expose profiler to Python by @Helveg in #1688
• Added a debug mode to build-catalogue by @Helveg in #1686
• Write build-catalogue (c)make errors to stdout and stderr by @Helveg in #1679
• Makejobs arg by @Helveg in #1673
• Feature/python clean memory by @max9901 in #1670
• Feature/label dict by @apeyser in #1711
• Add locset expressions for proximal and distal translation of locations by @bcumming in #1671
• Gap Junction mechanisms by @noraabiakar in #1682
• Solve non-linear systems that are not kinetic schemes. by @noraabiakar in #1724
• Add cos and sin SVE implementations by @noraabiakar in #1744
• Deal with zero radius points in a morphology by @halfflat in #1719
• Simplify default proc_allocation generation. by @halfflat in #1725
• Find morphology location from a coordinate by @bcumming in #1751
• resurrect lmorpho by @schmitts in #1746
• Expose information about the CV discretization to the public inteface by @noraabiakar in #1758
• Allow the use of string s-expression CV-policies in pyarb by @noraabiakar in #1799
• Construct domain_decomposition given a list of group_descriptions. by @noraabiakar in #1788
• Profile externally loaded mechanisms by @noraabiakar in #1691
• Have option to set thread count to local maximum by @brenthuisman in #1716

CI/build/testing:

• Python unit test clean-up by @noraabiakar in #1595
• Add CI for the in-repo Spack package by @schmitts in #1544
• Ci/sanitize by @thorstenhater in #1521
• Update spack package to include fmt for [email protected]: by @bcumming in #1609
• Cache Spack for GHA by @schmitts in #1619
• Run examples via scripts (closes #1566) by @schmitts in #1631
• Build a catalogue in CI by @schmitts in #1632
• Switch to PyPA's cibuildwheel action repo's. by @Helveg in #1703
• Automatic test discovery sans boilerplate by @Helveg in #1693
• build external mechanism catalogues in release mode by @noraabiakar in #1713
• make VERBOSE=1 in verbose mode by @Helveg in #1715
• Bump pybind to 2.8.1 by @brenthuisman in #1742
• Better error reporting in python context interface by @bcumming in #1732
• Better error diagnostic in python when invalid cell properties are provided. by @bcumming in #1743
• Test separately built catalogues by @bcumming in #1748
• Weekly cron job checking for submodule updates by @brenthuisman in #1741
• Added local Python testsuite instructions by @Helveg in #1809
• CMake CUDA clean-up. by @thorstenhater in #1804
• Add dry run benchmark cell model for testing communication scaling by @bcumming in #1627
• v0.6-rc by @brenthuisman in #1814
• More robust build-catalogue by @thorstenhater in #1784

Documentation:

• arbor.mechanism_catalogue.extend() documentation added. by @brenthuisman in #1597
• zenodo 0.5.2 entry by @brenthuisman in #1598
• mpi.py updated for new synapse labelling by @brenthuisman in #1602
• Tutorial structure: remove deduplication, includify by @brenthuisman in #1575
• Docs mechabi by @brenthuisman in #1610
• Improved documentation about weights by @brenthuisman in #1620
• Adapt to changes in Pandas requiring unique indices by @brenthuisman in #1625
• Add implies default license by @kanzl in #1635
• Update Slack to Gitter by @brenthuisman in #1637
• Better Python installation instructions by @brenthuisman in #1622
• Add example based on LFPykit example Example_Arbor_swc.ipynb by @espenhgn in #1652
• Name and describe arguments to constructors where ambiguous. by @brenthuisman in #1678
• improved Github Issue templates by @brenthuisman in #1698
• Add some comments about units to network ring examples by @schmitts in #1697
• Fix link (closes #1664) by @schmitts in #1701
• Fix class members in cable example by @schmitts in #1641
• Correctify and seabornize dendrite tutorial by @brenthuisman in #1706
• Polish GJ example readme by @thorstenhater in #1726
• Completed the docs with more config() flags by @Helveg in #1708
• Make connectivity example more complete by @brenthuisman in #1721
• Add description of arbor and NEURON's nernst application rules to the docs. by @noraabiakar in #1638
• Show docs version info more prominently on RTD by @brenthuisman in #1737
• Update/fix tinyopt README.md by @halfflat in #1747
• Add python gap junction example by @kanzl in #1750
• Documentation: add expected nmodl units by @noraabiakar in #1756
• Fixed broken link from get_probes to probes by @Helveg in #1763
• Removed double preposition by @Helveg in #1762
• Fixed MD to RST typo by @Helveg in #1760
• Fix broken links in probe_sample.rst by @Helveg in #1761
• Docs/contrib updates by @brenthuisman in #1736
• Add Release procedure to docs by @brenthuisman in #1738
• Example and tutorial for two cells connected via a gap junction by @schmitts in #1771
• Have single source for citation info, compatible with Github citation UI by @brenthuisman in #1791
• Advertise GH Discussions for modelling by @brenthuisman in #1793
• Add release cycle info to docs by @brenthuisman in #1798
• Added a warning when users are reading latest docs. by @Helveg in #1800

Fixes/optimization:

• Fix GPU compile. by @thorstenhater in #1601
• Compile all arbor source files with -fvisibility=hidden by @noraabiakar in #1599
• Bug/assorted static analysis by @thorstenhater in #1615
• Fix ambiguous Region/Locset expressions by @thorstenhater in #1629
• Clean-up AVX routine and global def [EOM] by @thorstenhater in #1618
• Categorise ASSIGNED RANGE variables as STATE-ish. by @thorstenhater in #1624
• Check that mechanisms have the right kind by @noraabiakar in #1633
• Resolve uninitialised values. by @thorstenhater in #1616
• Correctly parse diameter in ASCII files by @bcumming in #1640
• Attempt fixing a64fx build by @bcumming in #1636
• Fix simulation::add_sampler cell_group dispatch by @noraabiakar in #1654
• Fix on-components on empty cable bug by @halffla...

Arbor Library v0.5.2

This release fixes an error in the CI generated Python wheels, which are as of this release available on PyPI. Other than those fixes, this release is identical to v0.5.1.

Arbor Library v0.5.1

Since v0.5 there have been some major features, and many small fixes and improvements.

Core API features:

• [C++/Python] Labels instead of indices for placeable item identification.
• [C++/Python] Morphology file format support: Arbor Cable-Cell Format.
• [C++/Python] Morphology file format support: Neurolucida ASCII format.
• [C++/Python] Morphology file format improvements: SWC.
• [C++/Python] Simplified connections and junctions.
• [C++/Python] Enable simulation resume/restart.
• [C++/Python] Add post events functionality to support models with STDP synapses
• [C++/Python] Allow dynamically creating and loading of mechanism catalogue

Documentation:

• Documentation URL changed to docs.arbor-sim.org
• New Python examples and tutorials
• Ever more complete documentation
• Added Code of Conduct

Build / CI:

• Spack distribution
• CI generated binary Python wheels
• Apple M1 compatibility
• CI moved from Travis to Github Actions
• Improved Python and C++ unit testing
• ARM CI
• Sanitizer CI

Contributions by, in no specific order, @haampie, @clinssen, @espenhgn, @Helveg, @brenthuisman, @noraabiakar,
@thorstenhater, @halfflat, @schmitts and @bcumming

Arbor Library v0.5

Since v0.4 there have been some major features, and many small fixes and improvements.

Core API features:

• [C++/Python] Numerous small bug fixes, optimizations and improvements.
• [C++/Python] Refactor cable cell interface to be read only, and be constructed from
  descriptions of morphology, labels, and decorations.
• [C++/Python]Expose diverse probes and rich interface for describing where and
  what to sample on cable cells.
• [C++/Python] Support for querying names in mechanism catalogues
• [Python] Wrapper for existing C++ pw_lin functionality
• [C++] Improved validation of recipe definitions during model building

Documentation:

• Added new Python examples
• Many small fixes for links, spelling, grammar and clarity.
• Add extensive guide for contributions and coding polices.

Build:

• Allow CMake configuration to use system copies of C++ dependencies
  (nlohmann/json and pybind11), and makes this the default option.
• Added GitHub Actions support for automated testing of a wider range of tests
  and features than are run on our Travis CI (which will be removed soon)
• More robust Python detection and consistent use of the same Python
  interpreter in CMake configure and build steps.

Contributions by, in no specific order, @brenthuisman, @noraabiakar,
@thorstenhater, @halfflat, @schmitts and @bcumming