|
| 1 | +========================= |
| 2 | +PyNN 0.10.1 release notes |
| 3 | +========================= |
| 4 | + |
| 5 | +October 13th 2022 |
| 6 | + |
| 7 | +Welcome to PyNN 0.10.1! |
| 8 | + |
| 9 | + |
| 10 | +NEST 3.3, NEURON 8.1 |
| 11 | +-------------------- |
| 12 | + |
| 13 | +PyNN now supports the latest version of NEST_, and NEURON_ v8.1. |
| 14 | + |
| 15 | +New neuron model |
| 16 | +---------------- |
| 17 | + |
| 18 | +We have added a new standard neuron model, :class:`IF_curr_delta`, for which the synaptic response is |
| 19 | +a step change in the membrane voltage. |
| 20 | + |
| 21 | +Recording with irregular sampling intervals |
| 22 | +------------------------------------------- |
| 23 | + |
| 24 | +Where a simulator supports recording with varying sampling intervals (e.g. BrainScaleS, NEURON with CVode), |
| 25 | +PyNN can now handle such data, using the :class:`IrregularlySampledSignal` class from Neo. |
| 26 | + |
| 27 | +More efficient handling of spike trains |
| 28 | +--------------------------------------- |
| 29 | + |
| 30 | +Some simulators record spike trains as a pair of arrays, one containing neuron identifiers, |
| 31 | +the other spike times. |
| 32 | +PyNN can now retain this representation for internal data handling or for export, |
| 33 | +using the :class:`SpikeTrainList` class recently introduced in Neo, |
| 34 | +which avoids the overhead of always converting to a list of :class:`SpikeTrains`, one per neuron. |
| 35 | + |
| 36 | +Dependencies |
| 37 | +------------ |
| 38 | + |
| 39 | +PyNN now requires NumPy > 1.18.5, Neo > 0.11.0 or later. |
| 40 | + |
| 41 | +Bug fixes |
| 42 | +--------- |
| 43 | + |
| 44 | +A `small number of bugs`_ have been fixed. |
| 45 | + |
| 46 | +.. _`small number of bugs`: https://github.com/NeuralEnsemble/PyNN/issues?q=milestone%3A0.10.1+is%3Aclosed |
| 47 | +.. _NEST: https://www.nest-simulator.org |
| 48 | +.. _NEURON: https://neuron.yale.edu/neuron/ |
| 49 | +.. _`Brian 2`: https://brian2.readthedocs.io |
0 commit comments