https://github.com/vitay/ANNarchy

sort by:
Revision Author Date Message Commit Date
19475cf Version 4.5.7 23 August 2016, 17:22:23 UTC
5b76a3b Updated changelog 23 August 2016, 16:17:31 UTC
0d05da6 Issue #33: connect_one_to_one now accepts any kind of popview (even with random indices), as long as they have the same size (otherwise a warning is emitted). 23 August 2016, 16:05:30 UTC
dc44c75 Merge branch 'master' into develop 15 August 2016, 18:37:30 UTC
6d3a18b Merge branch 'master' of https://ai.informatik.tu-chemnitz.de/gogs/annarchy/annarchy 15 August 2016, 18:02:23 UTC
825322d Issue #33: bugfix in RecordGenerator related to spike events and PopulationViews. 15 August 2016, 17:58:36 UTC
34889df Merge branch 'master' into develop ( issue #32 ) 11 August 2016, 12:50:26 UTC
f1a7780 Merge branch 'master' of https://ai.informatik.tu-chemnitz.de/gogs/annarchy/annarchy 11 August 2016, 12:39:20 UTC
18cb241 Issue #32: bugfix in Monitor error messages. 11 August 2016, 09:20:07 UTC
bc92852 Bugfix: Normal(mean, sigma) distribution works now correctly. Before values for default (mean=0.0, sigma=1.0) were drawn. 19 July 2016, 13:16:55 UTC
a5ea8e8 Refractoriness: improved the doc 19 July 2016, 12:21:26 UTC
3dc0c58 Fixed refractoriness example to have the same noise as in the Brian example. 19 July 2016, 12:16:50 UTC
723ed9c Prepared the exponential numerical method to take out the step size of the for loop. TODO 15 July 2016, 08:33:16 UTC
6de0dac delays are now rounded to the next multiple of dt, not cast into a int (i.e. 1.99 is now 2, not 1 with dt=1.0). 15 July 2016, 08:06:10 UTC
8ace709 SpikeSourceArray now only uses long ints internally for the spike times (less errors). 15 July 2016, 08:04:49 UTC
7f730de SpikeSourceArray totally failed when the rounding of spike times lead to the same time (waiting forever for a spike in the past). 15 July 2016, 06:34:44 UTC
7edd055 The spike_times attribute of SpikeSourceArray can now be changed after compilation. 14 July 2016, 13:59:13 UTC
7f0086d Merge branch 'master' into develop 14 July 2016, 06:55:54 UTC
49a858c Variable delays with spikes: bug in the code generation. 14 July 2016, 06:54:26 UTC
8a22912 parser: spike condition and reset now also store the dependencies. For the spike condition it is in 'spike_cond_dependencies', for each reset equation it is in 'dependencies'. Beware: for each reset equation, the assigned variable is not part of it. 12 July 2016, 14:29:51 UTC
ebb83a1 Issue #30: PopulationViews had a bug hen modifying variables on a multidimensional population. 11 July 2016, 11:37:45 UTC
626e94d Sympy 1.0 introduced a regression on the != operator. 07 July 2016, 13:35:09 UTC
352d4b1 Merge branch 'master' into develop 06 July 2016, 13:20:39 UTC
53f91fe Bugfix: recording of weighted sums of inputs on CUDA 06 July 2016, 13:14:00 UTC
96cc7d0 Reformatted the doc of raster_plot() 06 July 2016, 09:47:36 UTC
0b417a6 Started verson 4.5.7. 05 July 2016, 20:52:40 UTC
1184583 Documented the recording. 05 July 2016, 20:37:54 UTC
93fba05 Rate-coded populations can now record weighted sums of inputs such as sum(exc). 05 July 2016, 17:52:19 UTC
0e07f21 PopulationViews now also have access to the weighted sums 05 July 2016, 14:41:33 UTC
f2d05b6 Populations now define a sum() method that allows to access the weighted sums. 05 July 2016, 12:39:07 UTC
a1a4596 Critical bug in parser: when a synapse uses pre/post variables with one name containing the other (e.g. r and r_mean), the temporary replacement fails. Used regex instead 01 July 2016, 16:17:08 UTC
45e7806 Non-uniform delays generated with Uniform(0.0, x) crashed. rd.min is set to dt. 01 July 2016, 12:11:04 UTC
6890038 bugfix: assigning random distributions to global synaptic variables was not working. 30 June 2016, 14:39:46 UTC
819fb27 Spiking networks can now use variable delays, at the cost of extra computations. 27 June 2016, 14:33:09 UTC
96bdde2 First dummy version for variable delays in spikes 27 June 2016, 09:19:11 UTC
a14f022 Started version 4.5.6. 26 June 2016, 16:31:41 UTC
291070a All examples are now added to the manifest. 26 June 2016, 16:30:18 UTC
5667163 Version 4.5.5 released 26 June 2016, 14:04:49 UTC
ec47091 Network.get() now accepts and returns a list of objects. 26 June 2016, 12:54:58 UTC
f944803 The glag pop.enabled was not set correctly when calling enable() or disable() 21 June 2016, 18:08:21 UTC
0cc6025 Network.add(Population) now also copies the enabled state. 21 June 2016, 14:33:23 UTC
f7a8e04 Populations have now an "enabled" attribute to read their status. enables()'s doc was buggy. 21 June 2016, 13:42:55 UTC
4ab48c2 parallel_run() with numbers now checks that compile() has been called. 14 June 2016, 12:47:04 UTC
54e5d2b Added an ``ite(cond, statement1, statement2)`` conditional function replicating ``if cond: statement1 else: statement2``, but which can be combined:: r = 1.0 + ite(sum(exc) > 1.0, sum(exc), 0.0) + ite(sum(inh) > 1.0, -sum(inh), 0.0) 10 June 2016, 17:16:03 UTC
46505b0 Merge branch 'develop' of https://ai.informatik.tu-chemnitz.de/gogs/annarchy/annarchy into develop 09 June 2016, 14:49:25 UTC
d3e3db1 Added ite(a,b,c) as a default function. 09 June 2016, 14:49:04 UTC
90dab11 save() method stores now the names of pre and post synaptic populations and the target for each projection. 09 June 2016, 13:38:21 UTC
4996e32 Cleaned up Compiler method. 08 June 2016, 11:42:02 UTC
7cbe6e4 Running multiple networks does not force recompilation. The files are now generated in subfolders (net0) and the makefiles are specific to each network. 08 June 2016, 11:38:08 UTC
7eb52d1 parallel_run()now accepts additional arguments. 08 June 2016, 08:56:10 UTC
a0e9310 python-config may not be shipped with python in virtualenv. 07 June 2016, 14:44:22 UTC
f57e90d Issue #25: Network has a get_population method. Doc updated to forbid accessing global simulation methods in parallel-run. 07 June 2016, 13:43:33 UTC
1fce550 Network: monitors now access time with their net_id 07 June 2016, 13:07:16 UTC
2f48206 Added ``Population.clear()`` to clear all spiking events (also delayed) without resetting the network. ``Population.reset()`` and ``Projection.reset()`` now accept a list of attributes to be reset, instead of resetting all of them. 03 June 2016, 11:15:27 UTC
b52166a Population implements a clear() methods which clears all spiking events (last spike, history of spikes) but leaves all variables intact, contrary to reset(). 02 June 2016, 16:19:25 UTC
b8bada8 STarted implemented ternary conditionals. Not ready yet! 02 June 2016, 15:49:11 UTC
87da77d bugfix: min/max bounds on g_target was wrongly analysed when depending on a parameter. 23 May 2016, 13:42:33 UTC
618e695 Issue #23: Added a warning when setup() is called after objects are created. Changing dt after delays are created does not change their value! 19 May 2016, 09:29:13 UTC
b8c14b2 Added a comment to connect_with_func() 17 May 2016, 14:51:06 UTC
851ee01 connect_from_file has problems with delays generated with connect_with_func() 17 May 2016, 13:50:56 UTC
c499bdb connect_from_file() has troubles with delays 17 May 2016, 13:35:40 UTC
e2d362a Windows line endings... 17 May 2016, 13:05:40 UTC
641da23 Refactored CPU unittests, callable now by python coverage tool (coverage.py) 02 May 2016, 14:54:30 UTC
a6143d5 reformatted README to be more Markdown like 29 April 2016, 15:57:37 UTC
7d58a9a Added travis CI's badge 29 April 2016, 15:51:05 UTC
4777db7 doc: no CSR should not throw an error 29 April 2016, 15:44:50 UTC
340fa0c doc: mocking submodules of scipy too 29 April 2016, 15:42:49 UTC
197ee5b doc: as scipy is now a dependency, readthedoc does not compile the doc... using mock to mimic scipy 29 April 2016, 15:39:02 UTC
69b65d1 setup.py does not import ANNarchy anymore 29 April 2016, 15:29:14 UTC
a7e176b Errors now automatically exit cleanly with the right error code. 29 April 2016, 15:18:30 UTC
0c0d891 travis: test 29 April 2016, 12:56:56 UTC
8ef2d00 Fixed test_global_operations for python 3. 29 April 2016, 12:19:36 UTC
a647549 travis: added cython 29 April 2016, 11:23:20 UTC
f3feb0b travis: pip install does not work 29 April 2016, 11:01:42 UTC
e2a769d travis: Forgot sympy 29 April 2016, 11:00:56 UTC
293e4d5 travis: try a correct numpy 29 April 2016, 10:59:04 UTC
390cdeb travis: reorganized the tests 29 April 2016, 10:46:30 UTC
84631b3 Cleaned the unittests 28 April 2016, 16:18:02 UTC
92ff670 Cleaned unnecessary tests 28 April 2016, 16:15:03 UTC
9886d7e travis: cython 28 April 2016, 14:53:47 UTC
d5884e2 travis: cython is way too old 28 April 2016, 14:51:18 UTC
12cb20e travis: added mpmath 28 April 2016, 14:46:21 UTC
80483c2 travis: last trial with sympy 28 April 2016, 14:41:19 UTC
d3b9c09 travis: sympy as dependency 28 April 2016, 14:30:27 UTC
e458190 travis: Perhaps better. 28 April 2016, 14:27:05 UTC
e00aa7b travis: trying the apt-get way 28 April 2016, 14:23:46 UTC
8d3160e travis: trying not installing the requirements... 28 April 2016, 14:11:31 UTC
1f7070a travis: got to pip install everything with pip2 and pip3... 28 April 2016, 14:08:00 UTC
bcaa5fd travis: install python 2.7 and 3.5before check 28 April 2016, 13:58:37 UTC
f20656c travis: preinstall miniconda to avoid missing dependencies with numpy 28 April 2016, 13:51:17 UTC
61a347c travis: Install requirement before installing ANNarchy 28 April 2016, 13:45:47 UTC
3998516 Forgot to add... 28 April 2016, 13:42:57 UTC
c1d4760 Added .travis.yml for continuous integration 28 April 2016, 13:39:55 UTC
4b16c0f doc: added HomogeneousCorrelatedSpikeTrains 28 April 2016, 12:02:38 UTC
7c86e6f CudaCheck files are now added to setup.py, Should allow pip install to use CUDA 28 April 2016, 11:47:53 UTC
398450a docstring for HomogeneousCorrelatedSpikeTrains 28 April 2016, 08:19:44 UTC
6f90b4a PopulationGenerator: global variables and RNG as in commit 46a13b7 RecordTemplate: bugfix record global variables (CUDA) 27 April 2016, 11:51:08 UTC
46a13b7 Global population variables can now use RNGs. TODO: CUDA 27 April 2016, 09:12:11 UTC
5948b88 python2-config does not always exist (anaconda). Using sys.prefix to find which python is used. 25 April 2016, 07:29:46 UTC
0a386d3 Anaconda does not have python2-config, only python-config 21 April 2016, 14:58:52 UTC
back to top