RuntimeDevice class

(Shortest import: from brian2.devices import RuntimeDevice)

class brian2.devices.device.RuntimeDevice

Bases: Device

The default device used in Brian, state variables are stored as numpy arrays in memory.

Attributes

arrays

Mapping from Variable objects to numpy arrays (or DynamicArray objects).

Methods

add_array(var)	Add an array to this device.
fill_with_array(var, arr)	Fill an array with the values given in another array.
get_array_name(var[, access_data])	Return a globally unique name for var.
get_random_state()	Return a (pickable) representation of the current random number generator state.
get_value(var[, access_data])
init_with_arange(var, start, dtype)	Initialize an array with an integer range.
init_with_zeros(var, dtype)	Initialize an array with zeros.
resize(var, new_size)	Resize a DynamicArrayVariable.
resize_along_first(var, new_size)
seed([seed])	Set the seed for the random number generator.
set_random_state(state)	Reset the random number generator state to a previously stored state (see Device.get_random_state).
set_value(var, value)
spike_queue(source_start, source_end)	Create and return a new SpikeQueue for this Device.

Details

arrays

Mapping from Variable objects to numpy arrays (or DynamicArray objects). Arrays in this dictionary will disappear as soon as the last reference to the Variable object used as a key is gone

add_array(var)

Add an array to this device.

Parameters:

var : ArrayVariable

The array to add.

fill_with_array(var, arr)

Fill an array with the values given in another array.

Parameters:

var : ArrayVariable

The array to fill.

arr : ndarray

The array values that should be copied to var.

get_array_name(var, access_data=True)

Return a globally unique name for var.

Parameters:

access_data : bool, optional

For DynamicArrayVariable objects, specifying True here means the name for the underlying data is returned. If specifying False, the name of object itself is returned (e.g. to allow resizing).

Returns:

name : str

The name for var.

get_random_state()

Return a (pickable) representation of the current random number generator state. Providing the returned object (e.g. a dict) to Device.set_random_state should restore the random number generator state.

Returns:

state :

The state of the random number generator in a representation that can be passed as an argument to Device.set_random_state.

get_value(var, access_data=True)

init_with_arange(var, start, dtype)

Initialize an array with an integer range.

Parameters:

var : ArrayVariable

The array to fill with the integer range.

start : int

The start value for the integer range

dtype : dtype

The data type to use for the array.

init_with_zeros(var, dtype)

Initialize an array with zeros.

Parameters:

var : ArrayVariable

The array to initialize with zeros.

dtype : dtype

The data type to use for the array.

resize(var, new_size)

Resize a DynamicArrayVariable.

Parameters:

var : DynamicArrayVariable

The variable that should be resized.

new_size : int

The new size of the variable

resize_along_first(var, new_size)

seed(seed=None)

Set the seed for the random number generator.

Parameters:

seed : int, optional

The seed value for the random number generator, or None (the default) to set a random seed.

set_random_state(state)

Reset the random number generator state to a previously stored state (see Device.get_random_state).

Parameters:

state :

A random number generator state as provided by Device.get_random_state.

set_value(var, value)

spike_queue(source_start, source_end)

Create and return a new SpikeQueue for this Device.

Parameters:

source_start : int

The start index of the source group (necessary for subgroups)

source_end : int

The end index of the source group (necessary for subgroups)

Tutorials and examples using this

• Example frompapers/Kremer_et_al_2011_barrel_cortex