def call_gan_training(generator, adversary):
gan_model = VanillaGAN(
generator=generator,
adversary=adversary,
z_dim=z_dim,
x_dim=im_dim,
folder="./veganModels/GAN",
)
# gan_model.summary(save=True)
gan_model.fit(
X_train=X_train,
X_test=X_test,
batch_size=batch_size,
epochs=epochs,
steps={"Adversary": 5},
print_every="0.5e",
save_model_every=None,
save_images_every="0.5e",
save_losses_every=1,
enable_tensorboard=True,
)
samples, losses = gan_model.get_training_results(by_epoch=False)
utils.plot_images(images=samples.reshape(-1, *samples.shape[2:]),
show=False)
utils.plot_losses(losses=losses, show=True)
def _build_images(self, images, labels=None):
""" Build matplotlib figure containing all images.
Parameters
----------
images : torch.Tensor()
A torch Tensor containing the images to be plotted by matplotlib.
Returns
-------
fig, axs: plt.Figure, np.array
Objects containing the figure as well as all separate axis objects.
"""
images = images.detach().cpu().numpy()
if self.images_produced:
fig, axs = plot_images(images=images, labels=labels, show=False)
return fig, axs
return None, None
gan_model.summary(save=True)
gan_model.fit(X_train=X_train,
y_train=y_train,
X_test=X_test,
y_test=y_test,
batch_size=batch_size,
epochs=epochs,
steps=None,
print_every="0.2e",
save_model_every=None,
save_images_every="0.2e",
save_losses_every=10,
enable_tensorboard=False)
samples, losses = gan_model.get_training_results(by_epoch=False)
training_time = np.round(gan_model.total_training_time / 60, 2)
title = "Epochs: {}, z_dim: {}, Time trained: {} minutes\nParams: {}\n\n".format(
epochs, z_dim, training_time, gan_model.get_number_params())
fixed_labels = np.argmax(gan_model.get_fixed_labels(), axis=1)
fig, axs = utils.plot_images(images=samples.reshape(-1, 32, 32),
labels=fixed_labels,
show=False)
fig.suptitle(title, fontsize=12)
fig.tight_layout()
plt.savefig(gan_model.folder + "/generated_images.png")
fig, axs = utils.plot_losses(losses=losses, show=False)
fig.suptitle(title, fontsize=12)
fig.tight_layout()
plt.savefig(gan_model.folder + "/losses.png")
# gan_model.save()
raise NotImplementedError("{} no yet implemented in logical gate.".format(model.__name__))
gan_model.summary(save=True)
gan_model.fit(
X_train=X_train,
X_test=X_test,
batch_size=batch_size,
epochs=epochs,
steps=None,
print_every="0.2e",
save_model_every=None,
save_images_every="0.5e",
save_losses_every=10,
enable_tensorboard=True
)
samples, losses = gan_model.get_training_results(by_epoch=False)
training_time = np.round(gan_model.total_training_time/60, 2)
title = "Epochs: {}, z_dim: {}, Time trained: {} minutes\nParams: {}\n\n".format(
epochs, z_dim, training_time, gan_model.get_number_params()
)
fig, axs = utils.plot_images(images=samples.reshape(-1, 32, 32), show=False)
fig.suptitle(title, fontsize=12)
fig.tight_layout()
plt.savefig(gan_model.folder+"/generated_images.png")
fig, axs = utils.plot_losses(losses=losses, show=False)
fig.suptitle(title, fontsize=12)
fig.tight_layout()
plt.savefig(gan_model.folder+"/losses.png")
# gan_model.save()
ngpu=None, # Optional
folder=None, # Optional
secure=True # Optional
)
gan_model.summary()
gan_model.fit(
X_train=X_train,
y_train=y_train,
X_test=X_test, # Optional
y_test=y_test, # Optional
batch_size=32, # Optional
epochs=2, # Optional
steps=None, # Optional
print_every="0.2e", # Optional
save_model_every=None, # Optional
save_images_every=None, # Optional
save_losses_every=10, # Optional
enable_tensorboard=False # Optional
)
samples, losses = gan_model.get_training_results(by_epoch=False)
fixed_labels = np.argmax(gan_model.get_fixed_labels(), axis=1)
fig, axs = plot_images(images=samples, labels=fixed_labels, show=True)
test_labels = np.eye(N=10)
test_samples = gan_model.generate(y=test_labels)
fig, axs = plot_images(images=test_samples,
labels=np.argmax(test_labels, axis=1),
show=True)
"{} no yet implemented in logical gate.".format(
model.__name__))
gan_model.summary(save=True)
gan_model.fit(X_train=train_dataloader,
y_train=None,
X_test=None,
y_test=None,
batch_size=None,
epochs=epochs,
steps=None,
print_every=500,
save_model_every=None,
save_images_every="0.1e",
save_losses_every=10,
enable_tensorboard=False)
samples, losses = gan_model.get_training_results(by_epoch=False)
training_time = np.round(gan_model.total_training_time / 60, 2)
title = "Epochs: {}, z_dim: {}, Time trained: {} minutes\nParams: {}\n\n".format(
epochs, z_dim, training_time, gan_model.get_number_params())
fig, axs = utils.plot_images(images=samples, show=False)
fig.suptitle(title, fontsize=12)
fig.tight_layout()
plt.savefig(gan_model.folder + "/generated_images.png")
fig, axs = utils.plot_losses(losses=losses, show=False)
fig.suptitle(title, fontsize=12)
fig.tight_layout()
plt.savefig(gan_model.folder + "/losses.png")
# gan_model.save()
gan_model.summary(save=True)
gan_model.fit(X_train=X_train,
y_train=y_train,
X_test=X_test,
y_test=y_test,
batch_size=batch_size,
epochs=epochs,
steps=None,
print_every="0.2e",
save_model_every=None,
save_images_every="0.2e",
save_losses_every=10,
enable_tensorboard=False)
samples, losses = gan_model.get_training_results(by_epoch=False)
training_time = np.round(gan_model.total_training_time / 60, 2)
title = "Epochs: {}, z_dim: {}, Time trained: {} minutes\nParams: {}\n\n".format(
epochs, z_dim, training_time, gan_model.get_number_params())
fixed_labels = np.argmax(gan_model.get_fixed_labels(), axis=1)
fig, axs = utils.plot_images(images=samples,
labels=fixed_labels,
show=False)
fig.suptitle(title, fontsize=12)
fig.tight_layout()
plt.savefig(gan_model.folder + "/generated_images.png")
fig, axs = utils.plot_losses(losses=losses, show=False)
fig.suptitle(title, fontsize=12)
fig.tight_layout()
plt.savefig(gan_model.folder + "/losses.png")
# gan_model.save()
generator = loader.load_generator(x_dim=x_dim, z_dim=z_dim, y_dim=None)
discriminator = loader.load_adversary(x_dim=x_dim, y_dim=None)
gan = VanillaGAN(generator=generator,
adversary=discriminator,
z_dim=z_dim,
x_dim=x_dim,
folder=None)
gan.summary() # optional, shows architecture
# Training
gan.fit(X_train[:300], enable_tensorboard=False)
# Vizualise results
images, losses = gan.get_training_results()
utils.plot_images(images)
utils.plot_losses(losses)
# Sample new images, you can also pass a specific noise vector
samples = gan.generate(n=36)
utils.plot_images(samples)
elif mode == "supervised":
import torch
import numpy as np
import vegans.utils as utils
import vegans.utils.loading as loading
from vegans.GAN import ConditionalVanillaGAN
# Data preparation (Load your own data or example MNIST)
loader = loading.MNISTLoader()