def __init__(self, params):
super(GenerativeNetwork, self).__init__()
self.params = params
generative_layers = [
nn.Linear(in_features=self.params.latent_dim,
out_features=self.params.hidden_dim),
GELU(),
UnFlatten(self.params.channels[3], self.params.hidden_height,
self.params.hidden_width),
nn.ConvTranspose2d(in_channels=self.params.channels[3],
out_channels=self.params.channels[4],
kernel_size=(self.params.kernel_size[6],
self.params.kernel_size[7]),
stride=self.params.stride[3]),
nn.BatchNorm2d(self.params.channels[4]),
GELU(),
nn.ConvTranspose2d(in_channels=self.params.channels[4],
out_channels=self.params.channels[5],
kernel_size=(self.params.kernel_size[8],
self.params.kernel_size[9]),
stride=self.params.stride[4]),
nn.BatchNorm2d(self.params.channels[5]),
GELU(),
nn.ConvTranspose2d(in_channels=self.params.channels[5],
out_channels=self.params.channels[6],
kernel_size=(self.params.kernel_size[10],
self.params.kernel_size[11]),
stride=self.params.stride[5]),
nn.Sigmoid(),
]
self.decoder = nn.Sequential(*generative_layers)
initialize_weights(self)
def __init__(self, params):
super(InferenceNetwork, self).__init__()
self.params = params
inference_layers = [
nn.Conv2d(in_channels=self.params.channels[0],
out_channels=self.params.channels[1],
kernel_size=self.params.kernel_size[0],
stride=self.params.stride[0],),
GELU(),
nn.BatchNorm2d(self.params.channels[1]),
nn.Conv2d(in_channels=self.params.channels[1],
out_channels=self.params.channels[2],
kernel_size=self.params.kernel_size[1],
stride=self.params.stride[1]),
GELU(),
nn.BatchNorm2d(self.params.channels[2]),
nn.Conv2d(in_channels=self.params.channels[2],
out_channels=self.params.channels[3],
kernel_size=self.params.kernel_size[2],
stride=self.params.stride[2]),
GELU(),
nn.BatchNorm2d(self.params.channels[3]),
Flatten(),
]
self.encoder = nn.Sequential(*inference_layers)
self.activation_fn = GELU()
self.fc_latent = nn.Linear(in_features=self.params.hidden_dim, out_features=self.params.latent_dim)
initialize_weights(self)
def __init__(self, params):
super(GenerativeNetwork, self).__init__()
self.params = params
self.fc1 = nn.Linear(in_features=self.params.latent_dim, out_features=self.params.hidden_dim)
self.fc2 = nn.Linear(in_features=self.params.hidden_dim, out_features=self.params.input_dim)
self.activation_fn = GELU()
initialize_weights(self)
def __init__(self, params):
super(InferenceNetwork, self).__init__()
self.params = params
self.fc = nn.Linear(in_features=self.params.input_dim, out_features=self.params.hidden_dim)
self.fc_mu = nn.Linear(in_features=self.params.hidden_dim, out_features=self.params.latent_dim)
self.fc_logvar = nn.Linear(in_features=self.params.hidden_dim, out_features=self.params.latent_dim)
self.activation_fn = GELU()
initialize_weights(self)