Learning rules¶
Build Nengo learning rules into the TensorFlow graph.
-
class
nengo_dl.learning_rule_builders.SimBCMBuilder(ops, signals)[source]¶ Build a group of
SimBCMoperators.-
build_step(signals)[source]¶ This function builds whatever computations need to be executed in each simulation timestep.
Parameters: - signals :
signals.SignalDict Mapping from
Signaltotf.Tensor(updated by operations)
Returns: - list of ``tf.Tensor``, optional
If not None, the returned tensors correspond to outputs with possible side-effects, i.e. computations that need to be executed in the TensorFlow graph even if their output doesn’t appear to be used
- signals :
-
-
class
nengo_dl.learning_rule_builders.SimOjaBuilder(ops, signals)[source]¶ Build a group of
SimOjaoperators.-
build_step(signals)[source]¶ This function builds whatever computations need to be executed in each simulation timestep.
Parameters: - signals :
signals.SignalDict Mapping from
Signaltotf.Tensor(updated by operations)
Returns: - list of ``tf.Tensor``, optional
If not None, the returned tensors correspond to outputs with possible side-effects, i.e. computations that need to be executed in the TensorFlow graph even if their output doesn’t appear to be used
- signals :
-
-
class
nengo_dl.learning_rule_builders.SimVojaBuilder(ops, signals)[source]¶ Build a group of
SimVojaoperators.-
build_step(signals)[source]¶ This function builds whatever computations need to be executed in each simulation timestep.
Parameters: - signals :
signals.SignalDict Mapping from
Signaltotf.Tensor(updated by operations)
Returns: - list of ``tf.Tensor``, optional
If not None, the returned tensors correspond to outputs with possible side-effects, i.e. computations that need to be executed in the TensorFlow graph even if their output doesn’t appear to be used
- signals :
-
-
nengo_dl.learning_rule_builders.build_pes(model, pes, rule)[source]¶ Builds a
PESobject into a model.A re-implementation of the Nengo PES rule builder, so that we can avoid slicing the encoders.
See
build_pes().Parameters: - model :
Model The model to build into.
- pes :
PES Learning rule type to build.
- rule :
LearningRule The learning rule object corresponding to the neuron type.
- model :