def cross_gene_sequences(gene_a, gene_b):
atomised_a = gene_a.strip().split()
atomised_b = gene_b.strip().split()
new_action = random.choice([atomised_a.pop(0), atomised_b.pop(0)])
func_a = parse_func(atomised_a)
func_b = parse_func(atomised_b)
node_a = random.choice(func_a.as_list())
node_b = random.choice(func_b.as_list())
if node_a.parent is None:
return new_action + " " + str(func_b)
elif node_a.parent.left == node_a:
node_a.parent.left = node_b
else:
node_a.parent.right = node_b
root = node_a.parent
while root.parent is not None:
root = root.parent
return new_action + " " + str(root)
def cross_gene_sequences(gene_a, gene_b):
atomised_a = gene_a.strip().split()
atomised_b = gene_b.strip().split()
new_action = random.choice([atomised_a.pop(0), atomised_b.pop(0)])
func_a = parse_func(atomised_a)
func_b = parse_func(atomised_b)
node_a = random.choice(func_a.as_list())
node_b = random.choice(func_b.as_list())
if node_a.parent is None:
return new_action + " " + str(func_b)
elif node_a.parent.left == node_a:
node_a.parent.left = node_b
else:
node_a.parent.right = node_b
root = node_a.parent
while root.parent is not None:
root = root.parent
return new_action + " " + str(root)
def __init__(self, meta, sequence):
# Set up first the metagenome
meta_genes = [float(g) for g in meta.strip().split()]
self.colors = [
meta_genes[i:i + 3] + [1.0]
for i in range(*Genome.META_INDICES["colors"], 3)
]
self.fuzziness = Genome.meta_item(meta_genes, "fuzziness")
self.const_bounds = Genome.meta_item(meta_genes, "const_bounds")
self.fun_gen_depth = Genome.meta_item(meta_genes, "fun_gen_depth")
self.incr_range = Genome.meta_item(meta_genes, "incr_range")
self.mult_range = Genome.meta_item(meta_genes, "mult_range")
self.mute_rates = {}
self.mute_rates["mute"] = Genome.meta_item(meta_genes, "mute")
self.mute_rates["genome"] = Genome.meta_item(meta_genes, "genome")
self.mute_rates["gene_action"] = Genome.meta_item(
meta_genes, "gene_action")
self.mute_rates["struct_mod"] = Genome.meta_item(
meta_genes, "struct_mod")
self.mute_rates["leaf_type"] = Genome.meta_item(
meta_genes, "leaf_type")
self.mute_rates["genome_rel"] = dict(
zip(list(GenomeMutes), Genome.meta_item(meta_genes, "genome_rel")))
self.mute_rates["const_rel"] = dict(
zip(list(ConstMutes), Genome.meta_item(meta_genes, "const_rel")))
self.mute_rates["leaf_rel"] = dict(
zip(list(RefType), Genome.meta_item(meta_genes, "leaf_rel")))
self.mute_rates["enum_rel"] = dict(
zip(list(EnumMutes), Genome.meta_item(meta_genes, "enum_rel")))
self.mute_rates["struct_rel"] = dict(
zip(list(StructMutes), Genome.meta_item(meta_genes, "struct_rel")))
# Now handle the behavioural genes
gene_sequences = [s.strip().split() for s in sequence.split("|")]
self.genes = [Gene(None, None) for _ in range(len(gene_sequences))]
for i in range(len(self.genes)):
action = goomba.Action(int(gene_sequences[i].pop(0)))
func = parse_func(gene_sequences[i])
self.fuzzify(func)
self.genes[i].action = action
self.genes[i].function = func
self.link()
def __init__(self, meta, sequence):
# Set up first the metagenome
meta_genes = [float(g) for g in meta.strip().split()]
self.colors = [meta_genes[i:i+3] + [1.0] for i in range(*Genome.META_INDICES["colors"], 3)]
self.fuzziness = Genome.meta_item(meta_genes, "fuzziness")
self.const_bounds = Genome.meta_item(meta_genes, "const_bounds")
self.fun_gen_depth = Genome.meta_item(meta_genes, "fun_gen_depth")
self.incr_range = Genome.meta_item(meta_genes, "incr_range")
self.mult_range = Genome.meta_item(meta_genes, "mult_range")
self.mute_rates = {}
self.mute_rates["mute"] = Genome.meta_item(meta_genes, "mute")
self.mute_rates["genome"] = Genome.meta_item(meta_genes, "genome")
self.mute_rates["gene_action"] = Genome.meta_item(meta_genes, "gene_action")
self.mute_rates["struct_mod"] = Genome.meta_item(meta_genes, "struct_mod")
self.mute_rates["leaf_type"] = Genome.meta_item(meta_genes, "leaf_type")
self.mute_rates["genome_rel"] = dict(zip(list(GenomeMutes),
Genome.meta_item(meta_genes, "genome_rel")))
self.mute_rates["const_rel"] = dict(zip(list(ConstMutes),
Genome.meta_item(meta_genes, "const_rel")))
self.mute_rates["leaf_rel"] = dict(zip(list(RefType),
Genome.meta_item(meta_genes, "leaf_rel")))
self.mute_rates["enum_rel"] = dict(zip(list(EnumMutes),
Genome.meta_item(meta_genes, "enum_rel")))
self.mute_rates["struct_rel"] = dict(zip(list(StructMutes),
Genome.meta_item(meta_genes, "struct_rel")))
# Now handle the behavioural genes
gene_sequences = [s.strip().split() for s in sequence.split("|")]
self.genes = [Gene(None, None) for _ in range(len(gene_sequences))]
for i in range(len(self.genes)):
action = goomba.Action(int(gene_sequences[i].pop(0)))
func = parse_func(gene_sequences[i])
self.fuzzify(func)
self.genes[i].action = action
self.genes[i].function = func
self.link()