def create(input_width, input_height, input_channels=1):
""" Vel factory function """
def instantiate(**_):
return DoubleNatureCnn(input_width=input_width,
input_height=input_height,
input_channels=input_channels)
return ModelFactory.generic(instantiate)
def create(input_length, hidden_layers, activation='tanh', normalization=None):
""" Vel factory function """
def instantiate(**_):
return MLP(input_length=input_length,
hidden_layers=hidden_layers,
activation=activation,
normalization=normalization)
return ModelFactory.generic(instantiate)
def create(input_width, input_height, input_channels=1, output_dim=512):
""" Vel factory function """
def instantiate(**_):
return NatureCnn(input_width=input_width,
input_height=input_height,
input_channels=input_channels,
output_dim=output_dim)
return ModelFactory.generic(instantiate)
def create(blocks, mode='basic', inplanes=16, divisor=4, num_classes=1000):
""" Vel factory function """
block_dict = {
'basic': BasicBlock,
'bottleneck': Bottleneck
}
def instantiate(**_):
return ResNetV2(block_dict[mode], blocks, inplanes=inplanes, divisor=divisor, num_classes=num_classes)
return ModelFactory.generic(instantiate)
def create(input_shape):
""" Vel factory function """
if isinstance(input_shape, numbers.Number):
input_shape = (input_shape, )
elif not isinstance(input_shape, tuple):
input_shape = tuple(input_shape)
def instantiate(**_):
return NormalizeObservations(input_shape)
return ModelFactory.generic(instantiate)
def create(alphabet_size: int,
output_dim: int,
pretrained: bool = False,
frozen: bool = False,
source: TextData = None):
""" Vel factory function """
def instantiate(**_):
return EmbeddingInput(alphabet_size,
output_dim,
pretrained=pretrained,
frozen=frozen,
source=source)
return ModelFactory.generic(instantiate)
def create(input_block: ModelFactory,
rnn_type: str,
hidden_layers: typing.List[int],
output_dim: int,
dropout=0.0):
""" Vel factory function """
def instantiate(**_):
return MultilayerRnnSequenceModel(input_block.instantiate(),
rnn_type=rnn_type,
hidden_layers=hidden_layers,
output_dim=output_dim,
dropout=dropout)
return ModelFactory.generic(instantiate)
def create(input_width,
input_height,
input_channels=1,
output_dim=512,
initial_std_dev=0.4,
factorized_noise=True):
""" Vel factory function """
def instantiate(**_):
return NoisyNatureCnn(input_width=input_width,
input_height=input_height,
input_channels=input_channels,
output_dim=output_dim,
initial_std_dev=initial_std_dev,
factorized_noise=factorized_noise)
return ModelFactory.generic(instantiate)
def create(input_width,
input_height,
input_channels=1,
rnn_type='lstm',
cnn_output_dim=512,
hidden_units=128):
""" Vel factory function """
def instantiate(**_):
return NatureCnnRnnBackbone(input_width=input_width,
input_height=input_height,
input_channels=input_channels,
rnn_type=rnn_type,
cnn_output_dim=cnn_output_dim,
hidden_units=hidden_units)
return ModelFactory.generic(instantiate)
def create(blocks,
mode='basic',
inplanes=64,
cardinality=4,
image_features=64,
divisor=4,
num_classes=1000):
""" Vel factory function """
block_dict = {
# 'basic': BasicBlock,
'bottleneck': ResNeXtBottleneck
}
def instantiate(**_):
return ResNeXt(block_dict[mode],
blocks,
inplanes=inplanes,
image_features=image_features,
cardinality=cardinality,
divisor=divisor,
num_classes=num_classes)
return ModelFactory.generic(instantiate)
def create(input_block: ModelFactory,
rnn_type: str,
output_dim: int,
rnn_layers: typing.List[int],
rnn_dropout: float = 0.0,
bidirectional: bool = False,
linear_layers: typing.List[int] = None,
linear_dropout: float = 0.0):
""" Vel factory function """
if linear_layers is None:
linear_layers = []
def instantiate(**_):
return MultilayerRnnSequenceClassification(
input_block=input_block.instantiate(),
rnn_type=rnn_type,
output_dim=output_dim,
rnn_layers=rnn_layers,
rnn_dropout=rnn_dropout,
bidirectional=bidirectional,
linear_layers=linear_layers,
linear_dropout=linear_dropout)
return ModelFactory.generic(instantiate)
def create(fc_layers=None, dropout=None, pretrained=True):
""" Vel factory function """
def instantiate(**_):
return Resnet34(fc_layers, dropout, pretrained)
return ModelFactory.generic(instantiate)
def create(alphabet_size: int):
""" Vel factory function """
def instantiate(**_):
return OneHotEncodingInput(alphabet_size)
return ModelFactory.generic(instantiate)
def create(img_rows, img_cols, img_channels, num_classes):
""" Vel factory function """
def instantiate(**_):
return Net(img_rows, img_cols, img_channels, num_classes)
return ModelFactory.generic(instantiate)
def create():
""" Vel factory function """
return ModelFactory.generic(ImageToTensor)
def create():
""" Vel factory function """
def instantiate(**_):
return Identity()
return ModelFactory.generic(instantiate)
