def BigGAN_render(class_vector, noise_vector, truncation):
if class_vector.shape[0] == 1:
class_vector = np.tile(class_vector, [noise_vector.shape[0], 1])
if noise_vector.shape[0] == 1:
noise_vector = np.tile(noise_vector, [class_vector.shape[0], 1])
class_vector = torch.from_numpy(class_vector.astype(np.float32)).to('cuda')
noise_vector = torch.from_numpy(noise_vector.astype(np.float32)).to('cuda')
with torch.no_grad():
output = model(noise_vector, class_vector, truncation)
imgs = convert_to_images(output.cpu())
return imgs
def get_image(class_vector, noise_vector, truncation=0.5):
class_vector = torch.from_numpy(class_vector)
noise_vector = torch.from_numpy(noise_vector)
# If you have a GPU, put everything on cuda
# noise_vector = noise_vector.to('cuda')
# class_vector = class_vector.to('cuda')
# model.to('cuda')
with torch.no_grad():
output = model(noise_vector, class_vector, truncation)
return convert_to_images(output)[0]
def biggan(inp: List[str], metadata):
truncation = 0.4
try:
class_vector = one_hot_from_names(inp, batch_size=len(inp))
noise_vector = truncated_noise_sample(truncation=truncation, batch_size=len(inp))
noise_vector = torch.from_numpy(noise_vector)
class_vector = torch.from_numpy(class_vector)
with torch.no_grad():
output = modelBG(noise_vector, class_vector, truncation)
except:
inp = ['cat']
class_vector = torch.from_numpy(one_hot_from_names(inp, batch_size=len(inp)))
noise_vector = torch.from_numpy(truncated_noise_sample(truncation=truncation, batch_size=len(inp)))
with torch.no_grad():
output = modelBG(noise_vector, class_vector, truncation)
return convert_to_images(output)[0]
def generate_image(thing="mushroom",
model_name="biggan-deep-512",
truncation=0.4):
"Generate an image of *thing* from the model, save it and return the path"
if model_name in ["waifu", "celeb"]:
return generate_waifu(model_name, truncation)
global img_i
model = get_model(model_name)
# Prepare a input
class_vector = one_hot_from_names([thing], batch_size=1)
noise_vector = truncated_noise_sample(truncation=truncation, batch_size=1)
# All in tensors
noise_vector = torch.from_numpy(noise_vector)
class_vector = torch.from_numpy(class_vector)
# If you have a GPU, put everything on cuda
noise_vector = noise_vector.to('cuda')
class_vector = class_vector.to('cuda')
model.to('cuda')
# Generate an image
with torch.no_grad():
output = model(noise_vector, class_vector, truncation)
# If you have a GPU put back on CPU
output = output.to('cpu')
img = convert_to_images(output)
out = img[0]
file_name = f"images/{img_i}.png"
img_i += 1
os.system("mkdir -p images/")
out.save(file_name, 'png')
print(
f"Generated an image of {thing} in file {file_name} with model {model_name}"
)
return file_name
pass
if (i + 1) * batch_size > len(class_vectors):
torch.cuda.empty_cache()
break
#get batch
noise_vector = noise_vectors[i * batch_size:(i + 1) * batch_size]
class_vector = class_vectors[i * batch_size:(i + 1) * batch_size]
# Generate images
with torch.no_grad():
output = model(noise_vector, class_vector, truncation)
output_cpu = output.cpu()
pil_imgs = convert_to_images(output_cpu)
#convert to image array and add to frames
for out in pil_imgs:
if args.scaling != 1:
newsize = (img_resolution * args.scaling,
img_resolution * args.scaling)
im = np.array(out.resize(newsize, Image.LANCZOS))
else:
im = np.array(out)
if use_disk == True:
imageio.imwrite(
os.path.join(tmp_folder_path,
str(counter) + ".png"), im)
counter = counter + 1
else:
saveFile += str('_LargeBatches/')
saveFile += str(str(classnam.capitalize()) + '/')
if super_jitter:
saveFile += str('SuperJitter' + str(minJ) + '/')
saveFile += str(classnam + '_' + str(x))
ensure_dir(saveFile)
save_as_images(output, saveFile)
#Save pngs as a .gif
imgs = convert_to_images(output)
saveGIF = saveFile + '/anim.gif'
ensure_dir(saveGIF)
imgs[0].save(saveGIF,
save_all=True,
append_images=imgs[1:],
duration=100,
loop=0)
noise_vector = truncated_noise_sample(truncation=truncation,
batch_size=batch_size)
#noise_vector = truncated_noise_sample(truncation=truncation, batch_size=1)
# All in tensors
#noise_vector = torch.from_numpy(np.ones([3, 128]).astype(np.float32)) #
noise_vector = torch.from_numpy(noise_vector)
class_vector = torch.from_numpy(class_vector)
# If you have a GPU, put everything on cuda
noise_vector = noise_vector.to('cuda')
class_vector = class_vector.to('cuda')
model.to('cuda')
# Generate an image
with torch.no_grad():
output = model(noise_vector, class_vector, truncation)
imgs = convert_to_images(output.cpu())
#%% 1d interpolation
truncation = 0.7
batch_size = 11
class_vector = one_hot_from_names(['mushroom'] * batch_size, batch_size=1)
noise_vector = truncated_noise_sample(truncation=truncation, batch_size=1)
scale_vec = np.arange(-1, 1.1, 0.2)
noise_vec_scale = scale_vec.reshape([-1, 1]) * noise_vector
imgs = BigGAN_render(class_vector, noise_vec_scale, truncation=truncation)
#%
figh = plt.figure(figsize=[25, 3])
gs = figh.add_gridspec(1, len(imgs)) # 1d interpolation
for i, img in enumerate(imgs):
plt.subplot(gs[i])
plt.imshow(img)
plt.axis('off')