Dendrite
Dendrite(self, proj, post_rank, idx)Sub-group of a Projection for a single post-synaptic neuron.
It can not be created directly, only through a call to Projection.dendrite(rank):
dendrite = proj.dendrite(6)Attributes
| Name | Description |
|---|---|
| post_rank | Rank of the post-synaptic neuron. |
| proj | Parent projection. |
| size | Number of synapses reaching the post-synaptic neuron. |
| pre_ranks | List of ranks of pre-synaptic neurons. |
| synapses | Iteratively returns the synapses corresponding to this dendrite. |
Methods
| Name | Description |
|---|---|
| synapse | Returns the synapse coming from the corresponding presynaptic neuron. |
| set | Sets the value of a parameter/variable on all synapses in the dendrite. |
| get | Returns the value of a parameter/variable. |
| receptive_field | Returns the given variable as a receptive field. |
| create_synapse | Creates a single synapse for this dendrite with the given pre-synaptic neuron. |
| create_synapses | Creates a set of synapses for this dendrite with the given pre-synaptic neurons. |
| prune_synapse | Removes the synapse with the given pre-synaptic neuron from the dendrite. |
| prune_synapses | Removes the synapses which belong to the provided pre-synaptic neurons from the dendrite. |
synapse
synapse(pos)Returns the synapse coming from the corresponding presynaptic neuron.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| pos | int | tuple[int] | can be either the rank or the coordinates of the presynaptic neuron | required |
set
set(value)Sets the value of a parameter/variable on all synapses in the dendrite.
dendrite.set( { 'tau' : 20, 'w'= Uniform(0.0, 1.0) } )Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| value | dict | a dictionary containing the parameter/variable names as keys. | required |
get
get(name)Returns the value of a parameter/variable.
dendrite.get('w')Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| name | str | name of the parameter/variable. | required |
receptive_field
receptive_field(variable='w', fill=0.0)Returns the given variable as a receptive field.
A numpy array of the same geometry as the pre-synaptic population is returned. Non-existing synapses are replaced by zeros (or the value fill).
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| variable | str | name of the variable (default = ‘w’) | 'w' |
| fill | float | value to use when a synapse does not exist (default: 0.0). | 0.0 |
create_synapse
create_synapse(rank, w=0.0, delay=0)Creates a single synapse for this dendrite with the given pre-synaptic neuron.
The configuration key 'structural_plasticity' must be set to True before compile() for this method to work.
net = ann.Network()
net.config(structural_plasticity=True)
net.compile()
try:
proj.dendrite(10).create_synapse(rank=20, w=0.1, delay=0.0)
except Exception as e:
print(e)If the synapse already exists, an error is thrown, so make sure to catch the exception.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| rank | int | rank of the pre-synaptic neuron | required |
| w | float | synaptic weight. | 0.0 |
| delay | float | synaptic delay in milliseconds that should be a multiple of dt. | 0 |
create_synapses
create_synapses(ranks, weights=None, delays=None)Creates a set of synapses for this dendrite with the given pre-synaptic neurons.
The configuration key 'structural_plasticity' must be set to True before compile() for this method to work.
net = ann.Network()
net.config(structural_plasticity=True)
net.compile()
try:
proj.dendrite(10).create_synapses(ranks=[20, 30, 40], weights=0.1, delay=0.0)
except Exception as e:
print(e)If the synapses already exist, an error is thrown, so make sure to catch the exception.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| ranks | list[int] | list of ranks for the pre-synaptic neurons. | required |
| weights | float | list[float] | list of synaptic weights (default: 0.0). | None |
| delays | float | list[float] | list of synaptic delays (default = dt). | None |
prune_synapse
prune_synapse(rank)Removes the synapse with the given pre-synaptic neuron from the dendrite.
The configuration key 'structural_plasticity' must be set to True before compile() for this method to work.
net = ann.Network()
net.config(structural_plasticity=True)
net.compile()
try:
proj.dendrite(10).prune_synapse(rank=20)
except Exception as e:
print(e)If the synapse does not exist, an error is thrown, so make sure to catch the exception.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| rank | int | rank of the pre-synaptic neuron | required |
prune_synapses
prune_synapses(ranks)Removes the synapses which belong to the provided pre-synaptic neurons from the dendrite.
The configuration key 'structural_plasticity' must be set to True before compile() for this method to work.
net = ann.Network()
net.config(structural_plasticity=True)
net.compile()
try:
proj.dendrite(10).prune_synapses(ranks=[20, 30, 40])
except Exception as e:
print(e)If the synapses do not exist, an error is thrown, so make sure to catch the exception.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| ranks | list[int] | list of ranks of the pre-synaptic neurons. | required |