def get_eval_points(model, conn, rng):
if conn.eval_points is None:
return npext.array(
model.params[conn.pre_obj].eval_points, min_dims=2)
else:
return gen_eval_points(
conn.pre_obj, conn.eval_points, rng, conn.scale_eval_points)
def build_linear_system(model, conn, rng):
encoders = model.params[conn.pre_obj].encoders
gain = model.params[conn.pre_obj].gain
bias = model.params[conn.pre_obj].bias
if conn.eval_points is None:
eval_points = npext.array(model.params[conn.pre_obj].eval_points,
min_dims=2)
else:
eval_points = gen_eval_points(conn.pre_obj, conn.eval_points, rng,
conn.scale_eval_points)
z = encoders.T / dc.Decimal(conn.pre_obj.radius)
#z=np.array([dc.Decimal(p) for p in z])
x = np.dot(eval_points, z)
activities = conn.pre_obj.neuron_type.rates(x, gain, bias)
if np.count_nonzero(activities) == 0:
raise RuntimeError(
"Building %s: 'activites' matrix is all zero for %s. "
"This is because no evaluation points fall in the firing "
"ranges of any neurons." % (conn, conn.pre_obj))
if conn.function is None:
targets = eval_points[:, conn.pre_slice]
else:
targets = npext.castDecimal(np.zeros(
(len(eval_points), conn.size_mid)))
for i, ep in enumerate(eval_points[:, conn.pre_slice]):
targets[i] = conn.function(ep)
return eval_points, activities, targets
def build_linear_system(model, conn, rng):
encoders = model.params[conn.pre_obj].encoders
gain = model.params[conn.pre_obj].gain
bias = model.params[conn.pre_obj].bias
if conn.eval_points is None:
eval_points = npext.array(
model.params[conn.pre_obj].eval_points, min_dims=2)
else:
eval_points = gen_eval_points(
conn.pre_obj, conn.eval_points, rng, conn.scale_eval_points)
x = np.dot(eval_points, encoders.T / conn.pre_obj.radius)
activities = conn.pre_obj.neuron_type.rates(x, gain, bias)
if np.count_nonzero(activities) == 0:
raise RuntimeError(
"Building %s: 'activites' matrix is all zero for %s. "
"This is because no evaluation points fall in the firing "
"ranges of any neurons." % (conn, conn.pre_obj))
if conn.function is None:
targets = eval_points[:, conn.pre_slice]
else:
targets = np.zeros((len(eval_points), conn.size_mid))
for i, ep in enumerate(eval_points[:, conn.pre_slice]):
targets[i] = conn.function(ep)
return eval_points, activities, targets
def get_eval_points(model, conn, rng):
if conn.eval_points is None:
view = model.params[conn.pre_obj].eval_points.view()
view.setflags(write=False)
return view
else:
return gen_eval_points(conn.pre_obj, conn.eval_points, rng, conn.scale_eval_points)
def get_eval_points(model, conn, rng):
if conn.eval_points is None:
view = model.params[conn.pre_obj].eval_points.view()
view.setflags(write=False)
return view
else:
return gen_eval_points(conn.pre_obj, conn.eval_points, rng,
conn.scale_eval_points)
def build_ensemble(model, ens):
if isinstance(ens.neuron_type, nengo.Direct):
raise NotImplementedError("Direct neurons not implemented")
# Create random number generator
rng = np.random.RandomState(model.seeds[ens])
eval_points = gen_eval_points(ens, ens.eval_points, rng=rng)
# Set up encoders
if isinstance(ens.encoders, Distribution):
encoders = get_samples(ens.encoders,
ens.n_neurons,
ens.dimensions,
rng=rng)
else:
encoders = npext.array(ens.encoders, min_dims=2, dtype=np.float64)
if ens.normalize_encoders:
encoders /= npext.norm(encoders, axis=1, keepdims=True)
# Build the neurons
gain, bias, max_rates, intercepts = get_gain_bias(ens, rng,
model.intercept_limit)
block = LoihiBlock(ens.n_neurons, label='%s' % ens)
block.compartment.bias[:] = bias
model.build(ens.neuron_type, ens.neurons, block)
# set default filter just in case no other filter gets set
block.compartment.configure_default_filter(model.decode_tau, dt=model.dt)
if ens.noise is not None:
raise NotImplementedError("Ensemble noise not implemented")
# Scale the encoders
# we exclude the radius to keep scaling reasonable for decode neurons
scaled_encoders = encoders * gain[:, np.newaxis]
model.add_block(block)
model.objs[ens]['in'] = block
model.objs[ens]['out'] = block
model.objs[ens.neurons]['in'] = block
model.objs[ens.neurons]['out'] = block
model.params[ens] = BuiltEnsemble(eval_points=eval_points,
encoders=encoders,
intercepts=intercepts,
max_rates=max_rates,
scaled_encoders=scaled_encoders,
gain=gain,
bias=bias)
def build_lif(model, ens):
# Create a random number generator
rng = np.random.RandomState(model.seeds[ens])
# Get the eval points
eval_points = ensemble.gen_eval_points(ens, ens.eval_points, rng=rng)
# Get the encoders
if isinstance(ens.encoders, Distribution):
encoders = ens.encoders.sample(ens.n_neurons, ens.dimensions, rng=rng)
encoders = np.asarray(encoders, dtype=np.float64)
else:
encoders = npext.array(ens.encoders, min_dims=2, dtype=np.float64)
encoders /= npext.norm(encoders, axis=1, keepdims=True)
# Get maximum rates and intercepts
max_rates = ensemble.sample(ens.max_rates, ens.n_neurons, rng=rng)
intercepts = ensemble.sample(ens.intercepts, ens.n_neurons, rng=rng)
# Build the neurons
if ens.gain is None and ens.bias is None:
gain, bias = ens.neuron_type.gain_bias(max_rates, intercepts)
elif ens.gain is not None and ens.bias is not None:
gain = ensemble.sample(ens.gain, ens.n_neurons, rng=rng)
bias = ensemble.sample(ens.bias, ens.n_neurons, rng=rng)
else:
raise NotImplementedError(
"gain or bias set for {!s}, but not both. Solving for one given "
"the other is not yet implemented.".format(ens)
)
# Scale the encoders
scaled_encoders = encoders * (gain / ens.radius)[:, np.newaxis]
# Store all the parameters
model.params[ens] = BuiltEnsemble(
eval_points=eval_points,
encoders=encoders,
scaled_encoders=scaled_encoders,
max_rates=max_rates,
intercepts=intercepts,
gain=gain,
bias=bias
)
# Create the object which will handle simulation of the LIF ensemble. This
# object will be responsible for adding items to the netlist and providing
# functions to prepare the ensemble for simulation. The object may be
# modified by later methods.
model.object_operators[ens] = operators.EnsembleLIF(ens)
def get_eval_points(model, conn, rng):
if conn.eval_points is None:
view = model.params[conn.pre_obj].eval_points.view()
view.setflags(write=False)
assert view.dtype == rc.float_dtype
return view
else:
return gen_eval_points(
conn.pre_obj,
conn.eval_points,
rng,
conn.scale_eval_points,
dtype=rc.float_dtype,
)
def build_lif(model, ens):
# Create a random number generator
rng = np.random.RandomState(model.seeds[ens])
# Get the eval points
eval_points = ensemble.gen_eval_points(ens, ens.eval_points, rng=rng)
# Get the encoders
if isinstance(ens.encoders, Distribution):
encoders = ens.encoders.sample(ens.n_neurons, ens.dimensions, rng=rng)
encoders = np.asarray(encoders, dtype=np.float64)
else:
encoders = npext.array(ens.encoders, min_dims=2, dtype=np.float64)
encoders /= npext.norm(encoders, axis=1, keepdims=True)
# Get maximum rates and intercepts
max_rates = ensemble.sample(ens.max_rates, ens.n_neurons, rng=rng)
intercepts = ensemble.sample(ens.intercepts, ens.n_neurons, rng=rng)
# Build the neurons
if ens.gain is None and ens.bias is None:
gain, bias = ens.neuron_type.gain_bias(max_rates, intercepts)
elif ens.gain is not None and ens.bias is not None:
gain = ensemble.sample(ens.gain, ens.n_neurons, rng=rng)
bias = ensemble.sample(ens.bias, ens.n_neurons, rng=rng)
else:
raise NotImplementedError(
"gain or bias set for {!s}, but not both. Solving for one given "
"the other is not yet implemented.".format(ens))
# Scale the encoders
scaled_encoders = encoders * (gain / ens.radius)[:, np.newaxis]
# Store all the parameters
model.params[ens] = BuiltEnsemble(eval_points=eval_points,
encoders=encoders,
scaled_encoders=scaled_encoders,
max_rates=max_rates,
intercepts=intercepts,
gain=gain,
bias=bias)
# Create the object which will handle simulation of the LIF ensemble. This
# object will be responsible for adding items to the netlist and providing
# functions to prepare the ensemble for simulation. The object may be
# modified by later methods.
model.object_operators[ens] = operators.EnsembleLIF(ens)
def get_eval_points(model, conn, rng):
if conn.eval_points is None:
return npext.array(model.params[conn.pre_obj].eval_points, min_dims=2)
else:
return gen_eval_points(conn.pre_obj, conn.eval_points, rng,
conn.scale_eval_points)
def build_ensemble(model, ens):
if isinstance(ens.neuron_type, nengo.Direct):
raise NotImplementedError("Direct neurons not implemented")
# Create random number generator
rng = np.random.RandomState(model.seeds[ens])
eval_points = gen_eval_points(
ens, ens.eval_points, rng=rng, dtype=nengo.rc.float_dtype
)
# Set up encoders
if isinstance(ens.encoders, Distribution):
encoders = get_samples(ens.encoders, ens.n_neurons, ens.dimensions, rng=rng)
encoders = np.asarray(encoders, dtype=nengo.rc.float_dtype)
else:
encoders = npext.array(ens.encoders, min_dims=2, dtype=nengo.rc.float_dtype)
if ens.normalize_encoders:
encoders /= npext.norm(encoders, axis=1, keepdims=True)
if np.any(np.isnan(encoders)):
raise BuildError(
f"{ens}: NaNs detected in encoders. This usually means that you have "
"zero-length encoders; when normalized, these result in NaNs. Ensure all "
"encoders have non-zero length, or set `normalize_encoders=False`."
)
# Build the neurons
gain, bias, max_rates, intercepts = get_gain_bias(
ens, rng, intercept_limit=model.intercept_limit, dtype=nengo.rc.float_dtype
)
block = LoihiBlock(ens.n_neurons, label="%s" % ens)
block.compartment.bias[:] = bias
# build the neuron_type (see builders below)
model.build(ens.neuron_type, ens.neurons, block)
# set default filter just in case no other filter gets set
block.compartment.configure_default_filter(model.decode_tau, dt=model.dt)
if ens.noise is not None:
raise NotImplementedError("Ensemble noise not implemented")
# Scale the encoders
# we exclude the radius to keep scaling reasonable for decode neurons
scaled_encoders = encoders * gain[:, np.newaxis]
# add instructions for splitting
model.block_shapes[block] = model.config[ens].block_shape
model.add_block(block)
model.objs[ens]["in"] = block
model.objs[ens]["out"] = block
model.objs[ens.neurons]["in"] = block
model.objs[ens.neurons]["out"] = block
model.params[ens] = BuiltEnsemble(
eval_points=eval_points,
encoders=encoders,
intercepts=intercepts,
max_rates=max_rates,
scaled_encoders=scaled_encoders,
gain=gain,
bias=bias,
)
def tune_ens_parameters(ens, function=None, solver=None, rng=None, n=1000):
"""Find good ensemble parameters for decoding a particular function.
Randomly generate many sets of parameters and determine the decoding error
for each. Then set the ensemble parameters to those with the lowest
decoding error. The "ensemble parameters" are the encoders, gains, biases,
and evaluation points.
Parameters
----------
ens : Ensemble
The ensemble to optimize.
function : callable, optional
The target function to optimize for. Defaults to the identity function.
solver : nengo.solvers.Solver, optional
The solver to use for finding the decoders. Default: ``LstsqL2()``
rng : numpy.random.RandomState, optional
The random number generator to use. Default: ``np.random``
n : int, optional
The number of random combinations to test. Default: 1000
"""
from nengo.dists import Distribution
from nengo.neurons import Direct
from nengo.solvers import LstsqL2
from nengo.builder.connection import solve_for_decoders
from nengo.builder.ensemble import gen_eval_points
if solver is None:
solver = LstsqL2()
if rng is None:
rng = np.random
if isinstance(ens.neuron_type, Direct):
raise ValueError("Parameters do not apply to Direct mode ensembles")
sample = lambda dist, n, d=None: (dist.sample(n, d=d, rng=rng)
if isinstance(dist, Distribution) else np
.asarray(dist))
# use the same evaluation points for all trials
eval_points = gen_eval_points(ens, ens.eval_points, rng=rng)
targets = (np.array([function(ep) for ep in eval_points])
if function is not None else eval_points)
# --- try random parameters and record error
errors = []
for i in range(n):
# --- generate random parameters
if ens.gain is None and ens.bias is None:
max_rates = sample(ens.max_rates, ens.n_neurons)
intercepts = sample(ens.intercepts, ens.n_neurons)
gain, bias = ens.neuron_type.gain_bias(max_rates, intercepts)
elif ens.gain is not None and ens.bias is not None:
gain = sample(ens.gain, ens.n_neurons)
bias = sample(ens.bias, ens.n_neurons)
else:
raise NotImplementedError("Mixed gain/bias and rates/ints")
encoders = sample(ens.encoders, ens.n_neurons, ens.dimensions)
# --- determine residual
x = np.dot(eval_points, encoders.T / ens.radius)
decoders, info = solve_for_decoders(solver, ens.neuron_type, gain,
bias, x, targets, rng)
error = info['rmses'].mean()
errors.append((error, encoders, gain, bias, eval_points))
# --- set parameters to those with the lowest error
errors.sort(key=lambda x: x[0])
ens.encoders, ens.gain, ens.bias, ens.eval_points = errors[0][1:]
def tune_ens_parameters(ens, function=None, solver=None, rng=None, n=1000):
"""Find good ensemble parameters for decoding a particular function.
Randomly generate many sets of parameters and determine the decoding error
for each. Then set the ensemble parameters to those with the lowest
decoding error. The "ensemble parameters" are the encoders, gains, biases,
and evaluation points.
Parameters
----------
ens : Ensemble
The ensemble to optimize.
function : callable, optional
The target function to optimize for. Defaults to the identity function.
solver : nengo.solvers.Solver, optional
The solver to use for finding the decoders. Default: ``LstsqL2()``
rng : numpy.random.RandomState, optional
The random number generator to use. Default: ``np.random``
n : int, optional
The number of random combinations to test. Default: 1000
"""
from nengo.dists import Distribution
from nengo.neurons import Direct
from nengo.solvers import LstsqL2
from nengo.builder.connection import solve_for_decoders
from nengo.builder.ensemble import gen_eval_points
if solver is None:
solver = LstsqL2()
if rng is None:
rng = np.random
if isinstance(ens.neuron_type, Direct):
raise ValueError("Parameters do not apply to Direct mode ensembles")
sample = lambda dist, n, d=None: (
dist.sample(n, d=d, rng=rng) if isinstance(dist, Distribution)
else np.asarray(dist))
# use the same evaluation points for all trials
eval_points = gen_eval_points(ens, ens.eval_points, rng=rng)
targets = (np.array([function(ep) for ep in eval_points])
if function is not None else eval_points)
# --- try random parameters and record error
errors = []
for i in range(n):
# --- generate random parameters
if ens.gain is None and ens.bias is None:
max_rates = sample(ens.max_rates, ens.n_neurons)
intercepts = sample(ens.intercepts, ens.n_neurons)
gain, bias = ens.neuron_type.gain_bias(max_rates, intercepts)
elif ens.gain is not None and ens.bias is not None:
gain = sample(ens.gain, ens.n_neurons)
bias = sample(ens.bias, ens.n_neurons)
else:
raise NotImplementedError("Mixed gain/bias and rates/ints")
encoders = sample(ens.encoders, ens.n_neurons, ens.dimensions)
# --- determine residual
x = np.dot(eval_points, encoders.T / ens.radius)
decoders, info = solve_for_decoders(
solver, ens.neuron_type, gain, bias, x, targets, rng)
error = info['rmses'].mean()
errors.append((error, encoders, gain, bias, eval_points))
# --- set parameters to those with the lowest error
errors.sort(key=lambda x: x[0])
ens.encoders, ens.gain, ens.bias, ens.eval_points = errors[0][1:]
