FAQ

========================== Frequently asked questions

1. How are the neurons/synapses addressed ?

---

Each neuron/synapse has three different numbers associated to it.

• Human readable address -> AddrGroup.addr
• Physical address ie The AER address -> AddrGroup.paddr
• Logical address ie (Neuron)(decimal)(synapse) eg 128.025 is neuron 128 , synappse corresponding to 0.025 -> AddrGroup.laddr

The logical address is what you see when you plot raster plots. So given a population all these addresses can be accesses as follows (logical addresses shown in the below example).

Soma

pop.soma.laddr

Synapses

pop.synapses['synapsetype'].laddr

2. How do I find out the available connection types aka 'fashions' available to me ?

---

Unfortunately there is no easy way to find out which types of connection 'fashions' you have available to you.

The easiest way to inspect which fashions you have available to you is described below.

Once you have initialized the NeuroSetup for your experimental setup in ipython do the following

$ setup.mapping.__connect_<tab>
  pyNCS.Mapping.__connect_all2all__                pyNCS.Mapping.__connect_by_probability_matrix__   pyNCS.Mapping.__connect_shuffle_all2all__
  pyNCS.Mapping.__connect_by_arbitrary_matrix__    pyNCS.Mapping.__connect_one2one__                
  pyNCS.Mapping.__connect_by_binary_matrix__       pyNCS.Mapping.__connect_random_all2all__ 

The above are the functions that are internally available when you use any given fashion. For eg. when you want to use fashion='one2one', internally the function connect_one2one is called. To know any further arguments that you need to pass to this function through fashion_kwargs, you can do the following:

$ setup.mapping.__connect_<fashion>__ ?
   Type:       instancemethod
String Form:<unbound method Mapping.__connect_by_arbitrary_matrix__>
File:       /usr/local/lib/python2.7/dist-packages/pyNCS/mapping.py
Definition: pyNCS.Mapping.__connect_by_arbitrary_matrix__(self, groupsrc, groupdst, M, resize_method='resample', expand=True, hide=False)
Docstring:  <no docstring>

The arguments you need to pass to your Connection for a given fashion through fashion_kwargs are the arguments you see in the above function call. For the above example the arguments that you need to send would be

fashion_kwargs = {'M':[<two dimensional matrix describing your connectivity>],
                            'resize_method':'resample',}