def create_objects(status):
status.Data_gen = GeneratorFromImage.GeneratorFromImage(status.Comp,
status.S,
cuda=status.cuda)
status.Data_gen.dataConv2d()
dataGen = status.Data_gen
x = dataGen.size[1]
y = dataGen.size[2]
max_layers = status.max_layer
max_filters = status.max_filter
def condition(DNA):
return max_filter(max_layer(DNA, max_layers), max_filters)
version = status.typos_version
"""
center=((-1,1,3,x,y),
(0,3, 4, x, y),
(1, 4, 2), (2,),(3,-1,0),(3,0,1),
(3,1,2))"""
center = ((-1, 1, 3, x, y), (0, 3, 15, 3, 3), (0, 18, 15, 3, 3),
(0, 33, 15, x, y), (1, 15, 2), (2, ), (3, -1, 0), (3, 0, 1),
(3, -1, 1), (3, 1, 2), (3, 0, 2), (3, -1, 2), (3, 2, 3), (3, 3,
4))
selector = status.Selector_creator(condition=condition, directions=version)
status.Selector = selector
creator = status.Creator
selector.update(center)
actions = selector.get_predicted_actions()
space = DNA_Graph(center, 1, (x, y), condition, actions, version, creator)
if status.Alai:
stream = TorchStream(status.Data_gen,
status.log_size,
min_size=status.min_log_size,
Alai=status.Alai)
else:
stream = TorchStream(status.Data_gen,
status.log_size,
min_size=status.min_log_size)
Phase_space = DNA_Phase_space(space, stream=stream)
Dynamics = Dynamic_DNA(space,
Phase_space,
status.dx,
Creator=creator,
Selector=selector,
update_velocity=velocity_updater,
update_space=space_updater,
version=version,
mutation_coefficient=status.mutation_coefficient,
clear_period=status.clear_period)
Phase_space.create_particles(status.n)
Phase_space.beta = status.beta
Phase_space.alpha = status.alpha
Phase_space.influence = status.influence
status.Dynamics = Dynamics
status.objects = Dynamics.objects
def create_objects(status):
status.Data_gen = GeneratorFromImage.GeneratorFromImage(status.Comp,
status.S,
cuda=False)
status.Data_gen.dataConv2d()
dataGen = status.Data_gen
x = dataGen.size[1]
y = dataGen.size[2]
def condition(DNA):
output = True
if DNA:
for num_layer in range(0, len(DNA) - 3):
layer = DNA[num_layer]
x_l = layer[3]
y_l = layer[4]
output = output and (x_l < x) and (y_l < y)
if output:
return max_filter(max_layer(DNA, 3), 40)
typos = []
print('The value of typos is')
print(status.typos)
#for element in status.typos:
# if type(element) == list:
# typos.append(tuple(element))
# else:
# typos.append(element)
#status.typos=tuple(typos)
#Number of filters
#center=((0, 3, 2, x, y), (1, 2, 2), (2,))
#space=DNA_Graph(center,status.dx,(x,y),condition,(0,(1,1,0,0)))
#Dimension of kernel
center = ((0, 3, 5, 3, 3), (0, 8, 8, 3, 3), (0, 11, 5, x, y), (1, 5, 2),
(2, ))
#center=((0, 3, 2, x, y), (1, 2, 2), (2,))
version = 'inclusion'
space = space = DNA_Graph(center, status.dx, (x, y), condition,
(0, (1, 0, 0, 0), (0, 0, 1, 1), (0, 1, 0, 0)),
version)
#space=space=DNA_Graph(center,1,(x,y),condition,(0,(1,0,0,0)),version)
Phase_space = DNA_Phase_space(space)
Dynamics = Dynamic_DNA(space, Phase_space, status.dx)
Phase_space.create_particles(status.n)
Phase_space.beta = status.beta
Phase_space.alpha = status.alpha
Phase_space.influence = status.influence
status.Dynamics = Dynamics
status.objects = Dynamics.objects
def initialize():
S = 100
Comp = 2
dataGen = GeneratorFromImage.GeneratorFromImage(Comp, S, cuda=False)
dataGen.dataConv2d()
x = dataGen.size[1]
y = dataGen.size[2]
ks = [2]
def condition(DNA):
return max_layer(DNA, 10)
center = ((0, 3, ks[0], x, y), (1, ks[0], 2), (2, ))
space = DNA_Graph(center, 10, (x, y), condition, (0, (1, 1, 0, 0)))
Phase_space = DNA_Phase_space(space)
return Phase_space
def __init__(self,DNA_graph,
Potential=None,Interaction=None,External=None):
self.DNA_graph = DNA_graph
self.objects = DNA_graph.objects
self.num_particles = None
self.beta=2
self.alpha=50
self.support=[]
dataGen = GeneratorFromImage.GeneratorFromImage(2, 200, cuda=False)
dataGen.dataConv2d()
self.stream=TorchStream(dataGen,25)
self.radius=10
self.influence=2
self.node_max_particles=None
self.max_changed=False
self.attach_balls()
def test_get_net():
dataGen = GeneratorFromImage.GeneratorFromImage(2, 100, cuda=False)
dataGen.dataConv2d()
stream = TorchStream(dataGen, 10)
x = stream.dataGen.size[1]
y = stream.dataGen.size[2]
DNA = create_DNA(2, dataGen)
network = create_network(2, dataGen)
stream.add_node(DNA)
stream.link_node(DNA, network)
log = stream.key2log(DNA)
stream.charge_nodes()
i = 0
while i < 100:
stream.charge_nodes()
print(stream.findCurrentvalue(DNA))
print(stream.get_net(DNA))
stream.pop()
i = i + 1
def test_charge():
dataGen = GeneratorFromImage.GeneratorFromImage(2, 100, cuda=False)
dataGen.dataConv2d()
stream = TorchStream(dataGen, 10)
x = stream.dataGen.size[1]
y = stream.dataGen.size[2]
DNA = create_DNA(2, dataGen)
network = create_network(2, dataGen)
stream.add_node(DNA)
stream.link_node(DNA, network)
log = stream.key2log(DNA)
log.signal = True
stream.charge_node(DNA)
i = 0
while i < 100:
stream.charge_nodes()
print(log.log)
print('With signal on')
print(len(stream.Graph.key2node))
stream.key2signal_on(DNA)
stream.pop()
i = i + 1
stream.remove_node(DNA)
network = create_network(2, dataGen)
stream.add_node(DNA)
stream.link_node(DNA, network)
log = stream.key2log(DNA)
stream.charge_nodes()
log.signal = False
i = 0
while i < 100:
print(log.log)
print('With signal off')
print(len(stream.Graph.key2node))
stream.charge_nodes()
stream.pop()
i = i + 1
inputs, labels = a[0] / 255, a[1]
for j in range(24000):
# zero the parameter gradients
optimizer.zero_grad()
# forward + backward + optimize
outputs = net(inputs)
loss = criterion(outputs, labels)
loss.backward()
optimizer.step()
# print statistics
running_loss += loss.item()
if j % 2000 == 1999: # print every 2000 mini-batches
print("Energy: ", running_loss, "i = ", j+1)
#net.delete_attribute("fc2")
running_loss = 0.0
dataGen = GeneratorFromImage.GeneratorFromImage(2, 15000, cuda=True)
dataGen.dataConv2d()
size = dataGen.size
x = size[1]
y = size[2]
k = 2
Test_dynamicNetwork(dataGen)
from DAO import GeneratorFromCIFAR, GeneratorFromImage, GeneratorFromImageNet dataGenImage = GeneratorFromImage.GeneratorFromImage(2, 120) #dataGenImageNet = GeneratorFromImageNet.GeneratorFromImageNet(5, 200) #dataGenCIFAR = GeneratorFromCIFAR.GeneratorFromCIFAR(5, 250) dataGenImage.dataConv3d() #dataGenImage.dataConv3d() #dataGenImage.dataNumpy() print(dataGenImage.data[0][0].shape) dataGenImage.dataNumpy() print(dataGenImage.data[0][0].shape)
def test_add_net():
dataGen = GeneratorFromImage.GeneratorFromImage(2, 100, cuda=False)
dataGen.dataConv2d()
stream = TorchStream(dataGen, 10)
DNA = create_DNA(2, dataGen)
stream.add_net(DNA)
def test_get_energy():
dataGen = GeneratorFromImage.GeneratorFromImage(2, 100, cuda=False)
dataGen.dataConv2d()
stream = TorchStream(dataGen, 10)
stream.findCurrentvalue((0, 0, 0))