spinnakermanchester / pynnexamples_old Goto Github PK
View Code? Open in Web Editor NEWExamples of Neural Networks using the PyNN API
Examples of Neural Networks using the PyNN API
It would be useful to have more comments/explanation in this script of the contents of the packet
that is printed out. Even better would be to include code for a full packet decode - for example extracting the neuron id from the packet.
Dear all,
When I try to use live_spikes_connection_receive.add_receive_callback( "pop_forward", receive_spikes) to add a call back function for population "pop_forward" in spike_io.py, the call back function "receive_spikes" is never called upon the receival of spikes, although the injection of spikes is successful. I am unclear that whether there is something wrong with configuration of database communication in spike_io.py or parameters settings in spynnaker.cfg in the root directory. As I use the spike_io.py from github for a trial, I just list the spynnaker.cfg settings to see if you could give me some help,
[Machine]
machineName = 192.168.240.253
version = 3
[Database]
create_database = True
wait_on_confirmation = True
create_routing_info_to_neuron_id_mapping = True
send_start_notification = True
listen_port = 19996
notify_port = 19999
notify_hostname = localhost
[Reports]
reportsEnabled = True
writeTextSpecs = False
writePartitionerReports = True
writePlacerReports = True
writeRouterReports = True
writeRouterInfoReport = True
writeRouterDatReport = False
writeTransceiverReport = True
writeProvanceData = True
writeTagAllocationReports = True
outputTimesForSections = True
writeReloadSteps = True
defaultReportFilePath = DEFAULT
defaultApplicationDataFilePath = DEFAULT
max_reports_kept = 10
max_application_binaries_kept = 10
[Simulation]
spikes_per_second = 30
ring_buffer_sigma = 5
[Routing]
algorithm = BasicDijkstra
generate_graphs = False
graphs_output_file = tmp
[Placer]
algorithm = Radial
[TagAllocator]
algorithm = Basic
[Partitioner]
algorithm = PartitionAndPlace
[KeyAllocator]
algorithm = MallocBased
[SpecExecution]
specExecOnHost = True
[MasterPopTable]
generator = BinarySearch
In PyNNExamples/examples/stdp_example.py, "stdp_model = sim.STDPMechanism(
timing_dependence=sim.SpikePairRule(tau_plus=10.0, tau_minus=10.0
,nearest=True),
weight_dependence=sim.AdditiveWeightDependence(w_min=0, w_max=1,
A_plus=0.001, A_minus=0.0012))"
There are A_plus and A_minus as parameters in weight_dependence, however, when I check pyNN neuron models for class STDPMechanism http://neuralensemble.org/docs/PyNN/_modules/pyNN/standardmodels/synapses.html#AdditiveWeightDependence, parameters A_plus and A_minus are in timing_dependence instead of weight_dependence, is there some difference in STDPMechanism in spinnaker?
And my second question is that, what is the length of the LTP and LTD time window in STDPMechanism? This piece of info is important for the settings of simulation parameters.
the pynn example of using the push bot fails to execute correctly.
Untill such time as this works correctly with the master branches of the tool chain, the code cannot be pulled into the system.
This issue is to keep track of the problem, whilst allwoing us to close the pull request, for metric purposes.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.