VERSION RELEASES

[salvadord]

# Version 0.9.2

https://github.com/Neurosim-lab/netpyne ... tag/v0.9.2

- Support for saving simulation at intervals

- Improved Python cell template import so can read section names from a Python list or dict of sections

- Fixed replicability issue across different number of cores when using 'connList' connectivity rules

- Improved speed performance of 'connList' connectivity rules

- Updated RxD calcium buffering example

- Added netClamp example to re-simulate activity of single cell from network data

- Added support for recording and plotting variable time step

- Replaced spectrogram FFT method with Morlet wavelet method

- Updated installation instructions

- Added 'fontSize' argument to several plots

- Fixed issue completing batches in Python 3

- Fixed bug saving evolutionary optimization batch output file

- Fixed bug in plotShape()

- Fixed bug in plotSpikeStats()

- Fixed bug loading files with colons

- Fixed issue sorting raster by population and y coord

[salvadord]

# Version 0.9.3

https://github.com/Neurosim-lab/netpyne ... ag/v.0.9.3

- Improved conversion from SONATA format, including 300 biophys cell example

- Added saveCellConns, pt3dRelativeToCellLocation and invertedYCoord cfg options

- Added lineWidth as argument of plotRatePSD() and plotLFP()

- Modified plotLFP and plotRatePSD to use Morlet wavelet for PSD calculation

- Added lineWidth argument to some raster and LFP plots

- Added shuffling test to granger causality functions

- Fixed bug plotting traces with one fig per trace

- Fixed bug in granger functions - required conversion to int

- Fixed bug in plotShape when cells sections not available

- Return meaningful cell info via 'repr' and 'str' for pointCell, compartCell

- Fixed spelling typos in documentation

[salvadord]

# Version 0.9.4

https://github.com/Neurosim-lab/netpyne ... tag/v0.9.4

- Added option to avoid creating connections with weight=0

- Made recording of dipoles optional with attribute cfg.recordDipoles

- Rotated conn matrix x-axis labels so can fit large matrices

- Added 'removeWeightNorm' argument to plotConn() function

- Added cfg option 'oneSynPerNetcon' to conn rules to create one synapse object per Netcon

- Added first version of Contributors Guide

- Fixed file write permissions for evol optim batches

- Fixed bug calculating normally distributed rhythmic inputs - was using variance instead of std

- Fixed bug in iplotDipole when smoothing window size = 0

- Fixed bug in plotSpikeStats pop order

- Fixed bug is plots with scalebars

- Fixed bug in grid lines of plotConn

- Fixed bug in subConnParams by sorting conns to ensure reproducibility on different number of cores

- Fixed bug so uses 'Agg' backend if no display env variable e.g. in clusters

[salvadord]

# Version 0.9.5

https://github.com/Neurosim-lab/netpyne ... tag/v0.9.5

- Print start and end date/time

- Avoid removing batch folder so can rerun and complete batch simulations

- Added loadBalance option to print individual node computer times

- Initialize hoc events recursivley to reduce event queue overhead

- Added plotRateSpectrogram analysis function

- Added option to save at intervals

- Fixed bug: wrong indentation for initializing fixedInterval in batch.py

- Fixed bug in _distributeSynsUniformly function

- Fixed bug in subConnParams grouped synapses

- Fixed bug in analysis.granger figure save name

- Fixed bug in printRunTime function

[joe.w.graham]

# Version 0.9.6

https://github.com/Neurosim-lab/netpyne ... ag/v0.9.6

- Added useful hover information to iplots

- Improvements to iplotTraces

- Added interactive plot for connectivity (iplotConn)

- Added 'dynamicRates' option for NetStim populations

- Added option to have a uniform distribution of rates for Vecstim pulses

- Added function for distributed saving at intervals

- Added issue templates for GitHub

- Overhaul of contribution guide (CONTRIBUTING.md)

- Improved documentation

- Fixed bugs and improved exception handling in plotConn

- Fixed loadSave V1 example model

- Fixed bug calculating popRates when no spikes

- Fixed bug in _distributeSynsUniformly()

- Fixed bug in saveCellParamsRule()

- Fixed bug in batch.py: initCfg params were not being updated for evol optim

- Fixed bug in subconn.py: missing initialization of newWeightNorm

- Fixed bug in run.py: print run time with 2 significant figures

- Fixed output stat filenames of evolutionary optim: .cvs to .csv

- Fixed bug in interval saving

- Fixed bug in spikes.py in plotSpikeStats

[joe.w.graham]

We have released NetPyNE version 0.9.7, which includes some major improvements:

- All analyses functions can now generate interactive HTML plots (via Bokeh package)
- Online documentation and tutorials have been updated and expanded (see http://www.netpyne.org/)
- Simplified cell type specification to facilitate model building
- New parameter optimization methods: Adaptive Stochastic Descent and Optuna (see examples)

See github.com/Neurosim-lab/netpyne/blob/v0.9.7/CHANGES.md for a complete list of improvements.
See netpyne.org/install.html for instructions on installation or updating.

Additionally, we have released our new NetPyNE GUI (see fig below) with a more intuitive interface, better performance and new features. Try out the new NetPyNE GUI without installing any software at: netpyne.org/gui

To learn more please check out our Tutorial and Workshop at CNS'2020 (register for free at http://cnsorg.org/cns-2020):

Tutorial: "Building mechanistic multiscale models, from molecules to networks, using NEURON and NetPyNE", July 18; website: https://sites.google.com/neurosim.downs ... 0-tutorial

Workshop: "Tools and resources for developing and sharing models in computational neuroscience", July 21-22; website: https://neuralensemble.github.io/Networks_SIG/CNS2020

[joe.w.graham]

NetPyNE v0.9.8 is now available.

For a complete list of changes and bug fixes see: https://github.com/Neurosim-lab/netpyne ... tag/v0.9.8

See here for instructions to install or update to the latest version: http://www.netpyne.org/install.html

NetPyNE Version 0.9.8

New features

- analysis.plotLFP can now be applied to external data

- Evolutionary and Optuna optimization can now read results from .pkl files

- Enabled recording of stimulus variables (e.g. SEClamp i)

- Added options in load func to avoid instantiating cells, conns, stims and/or rxd

- Analysis and plotting of rate vs current (f-I) curve for single cell tuning

- Recording and plotting of current dipoles in arbitrary subset of populations

- Added option to plot LFP spectrogram with log y-axis

- Added option to plot the log of connectivity in plotConn

- Enabled random distribution of multiple synapses with synsPerConn>1 and connRandomSecFromList=True

- Added lineWidth option to plot2Dnet (thanks Eric!)

- Updated Travis Continuous Integration to use NEURON 7.8.1 (from 7.6.2)

- Removed trailing white space from all Python files in the repository


Bug fixes

- Fixed bug when using index with recordTraces

- Fixed axes values in plotRxDConcentration and iplotRxDConcentration

- Added links to necessary mod files in the importing cell models tutorial

- Added aux_fun.inc to doc/source/code/mod as it's needed for some examples

- Blocked terminal output from Git when no repository is found (i.e. with pip install netpyne)

- Avoid adding section-based weightNorm values to point neurons (was affecting tut_import.py)

- Prevent error when loading pointCell params from json by converting Dict to dict

- Fixed bug in plotShape (thanks Eric!)

- Updated setup.py so PyPI website renders the README as markdown

- Fixed expected numSpikes in test of M1detailed example

- Merged GUI branch into development branch

- Fixed bug in iplotConn

- Fixed bug and coloring in iplotSpikeStats with different themes