Source code for nn.recurrent

from neat.graphs import required_for_output
from neat.six_util import itervalues, iteritems

[docs]class RecurrentNetwork(object): def __init__(self, inputs, outputs, node_evals): self.input_nodes = inputs self.output_nodes = outputs self.node_evals = node_evals self.values = [{}, {}] for v in self.values: for k in list(inputs) + list(outputs): v[k] = 0.0 for node, ignored_activation, ignored_aggregation, ignored_bias, ignored_response, links in self.node_evals: v[node] = 0.0 for i, w in links: v[i] = 0.0 = 0
[docs] def reset(self): self.values = [dict((k, 0.0) for k in v) for v in self.values] = 0
[docs] def activate(self, inputs): if len(self.input_nodes) != len(inputs): raise RuntimeError("Expected {0:n} inputs, got {1:n}".format(len(self.input_nodes), len(inputs))) ivalues = self.values[] ovalues = self.values[1 -] = 1 - for i, v in zip(self.input_nodes, inputs): ivalues[i] = v ovalues[i] = v for node, activation, aggregation, bias, response, links in self.node_evals: node_inputs = [ivalues[i] * w for i, w in links] s = aggregation(node_inputs) ovalues[node] = activation(bias + response * s) return [ovalues[i] for i in self.output_nodes]
[docs] @staticmethod def create(genome, config): """ Receives a genome and returns its phenotype (a RecurrentNetwork). """ genome_config = config.genome_config required = required_for_output(genome_config.input_keys, genome_config.output_keys, genome.connections) # Gather inputs and expressed connections. node_inputs = {} for cg in itervalues(genome.connections): if not cg.enabled: continue i, o = cg.key if o not in required and i not in required: continue if o not in node_inputs: node_inputs[o] = [(i, cg.weight)] else: node_inputs[o].append((i, cg.weight)) node_evals = [] for node_key, inputs in iteritems(node_inputs): node = genome.nodes[node_key] activation_function = genome_config.activation_defs.get(node.activation) aggregation_function = genome_config.aggregation_function_defs.get(node.aggregation) node_evals.append((node_key, activation_function, aggregation_function, node.bias, node.response, inputs)) return RecurrentNetwork(genome_config.input_keys, genome_config.output_keys, node_evals)