Skip to content

Exercises combining brain states and soft winner takes all

Notifications You must be signed in to change notification settings

APE-group/WTA-states

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

WTA-states

Exercises combining brain states and soft winner takes all

Acknowledgemt of fundings

This work has been cofunded by the European Next Generation EU grants Italian MUR CUP I53C22001400006 (FAIR PE0000013 PNRR Project).

TRY ON EBRAINS

USER GUIDE (preliminary)

see file AA-UserGuide.md

LOG OF MAIN EVENTS (from recent to past)

2024-1113 added a preliminary draft of User Guide 2024-1113-plastic-on-off release

  • (better provenance tracking) solved issue #57 by copying neural parameters yaml in output dirs

2024-1004 to 2024-1018 solved the follwing issues:

  • merged pull request #58 to solve enhancement issue 'copy neu params yaml in output dir' #57
  • solved enhancement issue #55: added a directory neu_models_catalogue hosting interesting neural configurations
  • solved bug #51 furter restringing the analysis window in _crop_and_plot.yaml
  • removed unnecessarily tracke overwrite_output directory
  • merged pull request #50 solving issue #45 add-actions-to-switch-on-plasticity-set-learning-parrameters-and-disconnect-synapses.
  • merged pull request #48 solving BUG issue #47: rates and spectral analysis for short analysis durations #47
  • merged pull request #46 solving BUG issue #32: measure of firing rate when a pop has zero spikes in the time window.
  • #43 specify plasticity default parameters in config file
  • #39 Add first approximation of background recurrency (can set inter_assemblies connectivity to existing (True) or non existing (False)) and attibute initial values for the weights in ...general_config.yaml files
  • #40 introduce flexible sparsity in synaptic matrices, added pairwise_bernolli connectivity, with programmable p_conn_exc and p_conn_inh values (defaulting to 1.0, i.e. all_to_all)

Release 2024-1002 solving issue #37 "Several issues related to Ca-AdEx calibration and plasticity management": addition of adaptation conductance, list of receptor names, learning rate settings, naming kinds of synapses and solving issue create dictionary book keeping the present kind of intra pop synapses #34

Release 2024-0912 solving issue #26 use recurrency false

Release 2024-0822 solving issue #19 switch on plasticity and #23: Added capability of disconnecting intra assemblies connections at programmable time, and what target exc pop to be disconnected

Release 2024-0820b config and plots can be saved in directories that can be specified using the file "directories_and_list_of_yamls.yaml". Use there the "overwrite_dir" file name if you want to send last runs always to the same directory.

Release 2024-0820 Solution of issue #16. It is now possible to specify a list of events during the simulation. The first supported action is the saving of intra assembly synaptic matrices at times that can be defined in the basic_ and tune_sim_and_recording_times.yaml files

Release 2024-0817 It Solves both issue #11 and issue #12. Now the main launches a single activity analysis function. All parameters for the analysis and the plot of the activity can be set in the basic_ and tune_ crop_and_plot.yaml files.

Commit 2024-0817 Solving issue #12 create crop_and_plot yaml files

Release 2024-0816 Solving issue #9: "encapsulate the first part of the main in a nest_parameters_preparation() function"

Release 2024-0808 Solving issue #5: "Add basic contextual signal reaching a single assembly with a start and stop time"it is the product of branch: "#5-add-basic-contextual-signal..."

Release 2024-0807 introduced multicompartment support solving issue#3: "Support for runs with both single and multi-compartment neurons"

Release 2024-0510 Added capability of spectral analysis with error bands, by pull request #1 that solves issue #2: "Need for spectral analysis with error band and detection of maxima"

2024-0430: novel features:

  • all parameters stored in reference yaml files
  • all parameters can be tuned reading additional "tune_..." yaml files
  • nest network creation and simulation in separate file

2024-0428 Initial commit