val_data,
batch_size=batch_size,
shuffle=True,
num_workers=num_dataloader_workers,
pin_memory=gpu,
)
test_dataloader = torch.utils.data.DataLoader(
test_data,
batch_size=batch_size,
shuffle=True,
num_workers=num_dataloader_workers,
pin_memory=gpu,
)
# Creating a sample set to visualize the model's training
sample = data.get_samples_from_data(val_data, 16)
################################################################################
##################################### Model ####################################
################################################################################
# Create Model
model = autoencoder.ConvolutionalAE(
max_filters=max_filters,
num_layers=num_layers,
input_image_dimensions=image_size,
latent_dim=latent_dim,
small_conv=small_conv,
)
model.to(device)
val_fusions,
batch_size=batch_size,
shuffle=True,
num_workers=num_dataloader_workers,
pin_memory=gpu,
)
test_fusion_dataloader = torch.utils.data.DataLoader(
test_fusions,
batch_size=batch_size,
shuffle=True,
num_workers=num_dataloader_workers,
pin_memory=gpu,
)
# Creating a sample set that we visualize every epoch to show the model's training
sample = data.get_samples_from_data(val_data, 16, fusion=False)
test_sample = data.get_samples_from_data(test_data, 16, fusion=False)
fusion_sample = data.get_samples_from_data(val_fusions, 4, fusion=True)
fusion_test_sample = data.get_samples_from_data(test_fusions, 4, fusion=True)
################################################################################
##################################### Model ####################################
################################################################################
# Create Model
model = models.ConvolutionalVAE(
max_filters=max_filters,
num_layers=num_layers,
input_image_dimensions=image_size,
latent_dim=latent_dim,
small_conv=small_conv,
