| home | user guide | examples | reference | changelog | resources |
A socket for a _CMLINK(network, Neuron, neuron). See the _CMLINK(assembly, BaseSocket, socket section) for a general description of how items can be grouped and reused by plugging compound items into sockets.
A neuron in Catacomb 2 is simply a _CMLINK(assembly, BaseAssembly, compound assembly) which has been assigned the property of being a neuron. That is, calling things neurons adds conceptual information to a model, but does not change what gets computed. Note, however, that future versions may use this information to try various optimisations which may be applicable to neurons.
The various formats are as follows:
rows: one line for each row in the table, with values for each column 0 1 0 0 1 0 1 0 0 x-y-value: triplets of indexes giving per element values 1 1 0 note that the zero elements can be ommitted and the order 1 2 1 does not matter 1 3 0 2 2 1 3 1 1 x y: as above but only with including the non-zero elements 1 2 2 2 3 1 xml: an xml format - best explained by saving an example in this format and looking at the result auto: the system will try to guess what the file contains
A projection causing direct conductance changes in the post-synaptic population. Like the _CMLINK(network,SpikeProjection, spiking projections), the connectivity may be one-one, one-all or random. In the latter case, the probability of any tow items being connected should be specified as _TT(p_connect). Conductance projections may be made between any vector provider, including a population of cells, and a population of components able to receive conductance inputs. There are therefore two possible sorts of inputs: vectors which represent membrane potential, and vectors which are dimensionless, typically ranging between zero and one. For the latter type, the _TT(rescale) flag should be set. This causes vector elements which are zero to add a conductance _TT(G0) with reversal potential _TT(V0) to the receiving component. Elements of value one give conductance and potential _TT(G1) and _TT(V1) and other values are scaled linearly. One common use of this is to stimulate a population of cells with binary vectors. In this case, _TT(G0) might be zero, making _TT(V0) irrelevant. Non-zero elements of the vector would then open a conductance _TT(G1) at reversal potential _TT(V1).
A spiking projection between populations. This is a convenience object that is equivalent to two spike links with a relay in the middle and, optionally, a delay relay.
A box for capturing compound cell objects. See the _CMLINK(assembly, CaptureBox, capture box) for details. The only extra feature of this component is that once captured the compound is called a _CMLINK(network,Neuron,Neuron) so it can be used in populations of neurons.
A neuron in Catacomb 2 is simply a _CMLINK(assembly, BaseAssembly, compound assembly) which has been assigned the property of being a neuron. That is, calling things neurons adds conceptual information to a model, but does not change what gets computed. Note, however, that future versions may use this information to try various optimisations which may be applicable to neurons.
A socket for a _CMLINK(network, Neuron, neuron). See the _CMLINK(assembly, BaseSocket, socket section) for a general description of how items can be grouped and reused by plugging compound items into sockets.