def __init__(self, paths: list, width: int, height: int, z_size: int):
self.paths = paths
self.width = width
self.height = height
self.z_size = z_size
self.input_data = InputData(self.width, self.height, self.z_size)
super(MyDataset, self).__init__()
def test_input_data():
data = InputData(width=12, height=14, z_size=2)
x, z = data.as_batch()
# check shape
assert x.shape == (12 * 14, 3)
assert z.shape == (12 * 14, 2)
def gen_images():
parser = argparse.ArgumentParser(description="multiple images as single image")
parser.add_argument('--out', type=Path, default=Path('./tmp/out.png'))
parser.add_argument('--n_rows', type=int, default=5)
parser.add_argument('--n_cols', type=int, default=5)
parser.add_argument('--model_config_path', type=Path, default=Path('./conf/model.yaml'))
parser.add_argument('--gpu', type=int, default=-1)
parser.add_argument('--load', type=Path)
parser.add_argument('--size', type=int)
args = parser.parse_args()
batch_size = args.n_rows * args.n_cols
# init model
model_config = ModelConfig.load(args.model_config_path)
model = CPPN(model_config)
# override size of output
if args.size:
model_config.width = args.size
model_config.height = args.size
if args.load:
assert args.load.exists()
print(f"load model from {args.load}")
chainer.serializers.load_npz(args.load, model)
# model to gpu
if args.gpu >= 0:
chainer.cuda.get_device_from_id(args.gpu).use()
model.to_gpu()
# init x and z
input_data = InputData(width=model_config.width, height=model_config.height, z_size=model_config.z_size)
x, z = [], []
for _ in range(batch_size):
_x, _z = input_data.as_batch()
x.append(_x)
z.append(_z)
x = numpy.concatenate(x)
z = numpy.concatenate(z)
# to device
xp = model.xp
x = chainer.Variable(xp.asarray(x))
z = chainer.Variable(xp.asarray(z))
y = model.forward(x, z)
y = chainer.cuda.to_cpu(y.data)
# chainer variable [B, 1, W, H], float [0, 1] -> numpy array uint8 [0, 255]
y = post_process_output(y)
y = y.reshape((args.n_rows, args.n_cols, 1, input_data.height, input_data.width))
y = y.transpose((0, 3, 1, 4, 2))
y = y.reshape((args.n_rows * input_data.height, args.n_cols * input_data.width))
Image.fromarray(y).save(args.out)
def __init__(self, width: int, height: int, z_size: int):
self.width = width
self.height = height
self.z_size = z_size
self.input_data = InputData(self.width, self.height, self.z_size)
# use only train dataset
self.train_data, _ = chainer.datasets.get_mnist()
super(MnistDataset, self).__init__()
def __init__(self, width: int, height: int, z_size: int, data_path: Path):
self.width = width
self.height = height
self.z_size = z_size
self.input_data = InputData(self.width, self.height, self.z_size)
# load image data
with gzip.open(data_path, 'rb') as f:
self.data = numpy.frombuffer(f.read(), numpy.uint8,
offset=16).reshape(-1, 28 * 28)
super(EMnistDataset, self).__init__()
def test_conditional_forward():
width = 5
height = 7
z_size = 2
batch_size = 3
batch_size = 3
model = ConditionalCPPN(
ConditionalModelConfig(width=width,
height=height,
n_units_xyr=3,
n_hidden_units=[
10,
10,
],
z_size=z_size,
in_width=64,
in_height=64,
in_channel=1))
x, z = [], []
for _ in range(batch_size):
_x, _z = InputData(width=width, height=height, z_size=z_size).as_batch()
x.append(_x)
z.append(_z)
x = numpy.concatenate(x, axis=0)
z = numpy.concatenate(z, axis=0)
c = get_dammy_input(batch_size, 64, 64, 1) # init dammy conditional input
y = model.forward(x, z, c)
assert y.shape == (batch_size, 1, width, height)
class MnistDataset(chainer.dataset.DatasetMixin):
def __init__(self, width: int, height: int, z_size: int):
self.width = width
self.height = height
self.z_size = z_size
self.input_data = InputData(self.width, self.height, self.z_size)
# use only train dataset
self.train_data, _ = chainer.datasets.get_mnist()
super(MnistDataset, self).__init__()
def __len__(self):
return len(self.train_data)
def get_example(self, i: int):
""" Return batch of image [1, 1, S, S], where S = `size` """
x, z = self.input_data.as_batch()
c, _ = self.train_data[i]
c = c * -1 + 1.0 # flip [0, 1] -> [1, 0]
c = c.reshape((28, 28)) * 255.0
image = Image.fromarray(c.astype(numpy.uint8)).resize((self.width, self.height))
c = numpy.asarray(image).astype(numpy.float32) / 255.0 # 2D array
c = c.reshape((1, 1, self.width, self.height))
return x, z, c
class EMnistDataset(chainer.dataset.DatasetMixin):
def __init__(self, width: int, height: int, z_size: int, data_path: Path):
self.width = width
self.height = height
self.z_size = z_size
self.input_data = InputData(self.width, self.height, self.z_size)
# load image data
with gzip.open(data_path, 'rb') as f:
self.data = numpy.frombuffer(f.read(), numpy.uint8,
offset=16).reshape(-1, 28 * 28)
super(EMnistDataset, self).__init__()
def __len__(self):
return len(self.data)
def get_example(self, i: int):
""" Return batch of image [1, 1, S, S], where S = `size` """
x, z = self.input_data.as_batch()
c = self.data[i] # [0, 255]
c = c * -1 + 255 # flip
c = c.reshape((28, 28)).T
image = Image.fromarray(c).resize((self.width, self.height))
c = numpy.asarray(image).astype(numpy.float32) / 255.0 # 2D array
c = c.reshape((1, 1, self.width, self.height))
return x, z, c
class MyDataset(chainer.dataset.DatasetMixin):
def __init__(self, paths: list, width: int, height: int, z_size: int):
self.paths = paths
self.width = width
self.height = height
self.z_size = z_size
self.input_data = InputData(self.width, self.height, self.z_size)
super(MyDataset, self).__init__()
def __len__(self):
return len(self.paths)
def get_example(self, i):
""" returns x, z, c """
# get
x, z = self.input_data.as_batch()
# open image and convert to [1, 1, W, H]
c = Image.open(Path(self.paths[i]))
c = c.resize((self.width, self.height))
c = c.convert('L')
c = numpy.asarray(c, dtype=numpy.float32) / 255. # [0, 255] -> [0, 1]
c = c.reshape((1, 1, self.width, self.height)) # [1, 1, W, H]
return x, z, c
def gen_input_batch(batch_size, width, height, z_size):
# create inputs
inputs = {}
x, z = [], []
for idx in range(batch_size):
_x, _z = InputData(width=width, height=height,
z_size=z_size).as_batch()
_x = chainer.Variable(_x)
_z = chainer.Variable(_z)
x.append(_x)
z.append(_z)
inputs[idx] = (_x, _z)
x = F.concat(x, axis=0)
z = F.concat(z, axis=0)
return x, z, inputs
def test_unconditional_forward():
width = 5
height = 7
z_size = 2
batch_size = 3
model = CPPN(ModelConfig(width=width, height=height, n_units_xyrz=3, n_hidden_units=[5, 5], z_size=z_size))
x, z = [], []
for _ in range(batch_size):
_x, _z = InputData(width=width, height=height, z_size=z_size).as_batch()
x.append(_x)
z.append(_z)
x = numpy.concatenate(x, axis=0)
z = numpy.concatenate(z, axis=0)
y = model.forward(x, z)
assert y.shape == (batch_size, 1, width, height)
def interp_movie():
parser = argparse.ArgumentParser(description="Gen gif movie")
parser.add_argument('--out', type=Path, default=Path('./tmp/out.gif'))
parser.add_argument('--frames', type=int, default=10)
parser.add_argument('--z_points', type=int, default=10)
parser.add_argument('--batch_size', type=int, default=50)
parser.add_argument('--model_config_path',
type=Path,
default=Path('./conf/model.yaml'))
parser.add_argument('--gpu', type=int, default=-1)
parser.add_argument('--load', type=Path)
parser.add_argument('--size', type=int)
args = parser.parse_args()
# create directory to put result
args.out.parent.mkdir(exist_ok=True)
# init model
model_config = ModelConfig.load(args.model_config_path)
# override size of output
if args.size:
model_config.width = args.size
model_config.height = args.size
model = CPPN(model_config)
if args.load:
assert args.load.exists()
print(f"load model from {args.load}")
chainer.serializers.load_npz(args.load, model)
# model to gpu
if args.gpu >= 0:
chainer.cuda.get_device_from_id(args.gpu).use()
model.to_gpu()
# init x and z
input_data = InputData(width=model_config.width,
height=model_config.height,
z_size=model_config.z_size)
# gen frames
images = []
zs = []
for _ in range(args.z_points):
zs.extend(
interp_z(sample_z(model_config.z_size),
sample_z(model_config.z_size), args.frames))
for i in range(0, len(zs), args.batch_size):
begin_idx = i
end_idx = min(i + args.batch_size, len(zs) - 1)
print(f"{begin_idx} -> {end_idx}")
# make input batch
x = []
z = []
for _z in zs[begin_idx:end_idx]:
_x, _z = input_data.as_batch(z=_z)
x.append(_x)
z.append(_z)
if len(x) == 0:
break
x = numpy.concatenate(x)
z = numpy.concatenate(z)
# to device
xp = model.xp
x = chainer.Variable(xp.asarray(x))
z = chainer.Variable(xp.asarray(z))
y = model.forward(x, z)
y = chainer.cuda.to_cpu(y.data)
# chainer variable [B, 1, W, H], float [0, 1] -> numpy array uint8 [0, 255]
y = post_process_output(y)
for _y in y:
images.append(Image.fromarray(_y[0]))
# save as gif
images[0].save(str(args.out), save_all=True, append_images=images)