import re
import glob
import data
import module
from os.path import join,basename,exists
from os import mkdir,makedirs
import os
os.environ["CUDA_DEVICE_ORDER"]="PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"]="3"
# ==============================================================================
# = param =
# ==============================================================================
# py.arg('--dataset', default='MR2CT')
py.arg('--datasets_dir', default='./datasets')
py.arg('--size', type=int, default=256) # load image to this size
# py.arg('--crop_size', type=int, default=256) # then crop to this size
py.arg('--batch_size', type=int, default=2)
py.arg('--epochs', type=int, default=225)
py.arg('--epoch_decay', type=int, default=25) # epoch to start decaying learning rate
py.arg('--lr', type=float, default=0.000001)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--adversarial_loss_mode', default='gan', choices=['gan', 'lsgan'])
# py.arg('--gradient_penalty_mode', default='none', choices=['none', 'dragan', 'wgan-gp'])
py.arg('--gradient_penalty_weight', type=float, default=10.0)
py.arg('--cycle_loss_weight', type=float, default=12.0)
py.arg('--identity_loss_weight', type=float, default=0.5)
py.arg('--output_dir',default='./Results')
py.arg('--pool_size', type=int, default=50) # pool size to store fake samples
args = py.args()
import numpy as np
import pylib as py
import tensorflow as tf
import tensorflow.keras as keras
import tf2lib as tl
import tf2gan as gan
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--dataset', default='horse2zebra')
py.arg('--output_index', default='')
py.arg('--load_size', type=int, default=286) # load image to this size
py.arg('--crop_size', type=int, default=256) # then crop to this size
py.arg('--batch_size', type=int, default=1)
py.arg('--epochs', type=int, default=200)
py.arg('--epoch_decay', type=int,
default=100) # epoch to start decaying learning rate
py.arg('--lr', type=float, default=0.0002)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--adversarial_loss_mode',
default='lsgan',
choices=['gan', 'hinge_v1', 'hinge_v2', 'lsgan', 'wgan'])
py.arg('--gradient_penalty_mode',
default='none',
choices=['none', 'dragan', 'wgan-gp'])
format='%(asctime)s %(levelname)s %(name)s %(threadName)s : %(message)s')
# imports for running test.py (UniGAN app)
import imlib as im
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# configuration
DEBUG = True
os.environ["CUDA_PATH"] = "/usr/local/cuda"
py.arg('--flask_path', default='/var/www/html/flaskapp_unigan')
py.arg('--img_dir', default='./data/zappos_50k/images')
py.arg('--test_label_path', default='./data/zappos_50k/test_label.txt')
py.arg('--test_int', type=float, default=2)
py.arg('--experiment_name', default='UniGAN_128')
args_ = py.args()
# output_dir
output_dir = os.path.join(args_.flask_path,
py.join('output', args_.experiment_name))
# save settings
args = py.args_from_yaml(py.join(output_dir, 'settings.yml'))
args.__dict__.update(args_.__dict__)
# others
# default_att_names = ['ancient', 'barren', 'bent', 'blunt', 'bright', 'broken', 'browned', 'brushed',
# 'burnt', 'caramelized', 'chipped', 'clean', 'clear', 'closed', 'cloudy', 'cluttered', 'coiled',
# 'cooked', 'cored', 'cracked', 'creased', 'crinkled', 'crumpled', 'crushed', 'curved', 'cut',
# 'damp', 'dark', 'deflated', 'dented', 'diced', 'dirty', 'draped', 'dry', 'dull', 'empty',
# 'engraved', 'eroded', 'fallen', 'filled', 'foggy', 'folded', 'frayed', 'fresh', 'frozen',
# 'full', 'grimy', 'heavy', 'huge', 'inflated', 'large', 'lightweight', 'loose', 'mashed',
# 'melted', 'modern', 'moldy', 'molten', 'mossy', 'muddy', 'murky', 'narrow', 'new', 'old',
# 'open', 'painted', 'peeled', 'pierced', 'pressed', 'pureed', 'raw', 'ripe', 'ripped', 'rough',
# 'ruffled', 'runny', 'rusty', 'scratched', 'sharp', 'shattered', 'shiny', 'short', 'sliced',
# 'small', 'smooth', 'spilled', 'splintered', 'squished', 'standing', 'steaming', 'straight',
# 'sunny', 'tall', 'thawed', 'thick', 'thin', 'tight', 'tiny', 'toppled', 'torn', 'unpainted',
# 'unripe', 'upright', 'verdant', 'viscous', 'weathered', 'wet', 'whipped',
# 'wide', 'wilted', 'windblown', 'winding', 'worn', 'wrinkled', 'young']
py.arg('--att_names', choices=data.ATT_ID.keys(), nargs='+', default=default_att_names)
py.arg('--img_dir', default='./data/CelebAMask-HQ/CelebA-HQ-img')
py.arg('--train_label_path', default='./data/CelebAMask-HQ/train_label.txt')
py.arg('--val_label_path', default='./data/CelebAMask-HQ/val_label.txt ')
py.arg('--load_size', type=int, default=256)
py.arg('--crop_size', type=int, default=256)
py.arg('--n_epochs', type=int, default=60)
py.arg('--epoch_start_decay', type=int, default=30)
py.arg('--batch_size', type=int, default=1)
py.arg('--learning_rate', type=float, default=2e-4)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--model', default='model_256', choices=['model_128', 'model_256', 'model_384'])
import imlib as im
import numpy as np
import pylib as py
import tensorflow as tf
import tf2lib as tl
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--experiment_dir')
py.arg('--batch_size', type=int, default=32)
test_args = py.args()
args = py.args_from_yaml(py.join(test_args.experiment_dir, 'settings.yml'))
args.__dict__.update(test_args.__dict__)
# ==============================================================================
# = test =
# ==============================================================================
# data
A_img_paths_test = py.glob(py.join(args.datasets_dir, args.dataset, 'testA'),
'*.jpg')
B_img_paths_test = py.glob(py.join(args.datasets_dir, args.dataset, 'testB'),
'*.jpg')
A_dataset_test = data.make_dataset(A_img_paths_test,
args.batch_size,
args.load_size,
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
default_att_names = [
'Bald', 'Bangs', 'Black_Hair', 'Blond_Hair', 'Brown_Hair',
'Bushy_Eyebrows', 'Eyeglasses', 'Male', 'Mouth_Slightly_Open', 'Mustache',
'No_Beard', 'Pale_Skin', 'Young'
]
py.arg('--att_names',
choices=data.ATT_ID.keys(),
nargs='+',
default=default_att_names)
py.arg(
'--img_dir',
default=
'./data/img_celeba/aligned/align_size(572,572)_move(0.250,0.000)_face_factor(0.450)_jpg/data'
)
py.arg('--train_label_path', default='./data/img_celeba/train_label.txt')
py.arg('--val_label_path', default='./data/img_celeba/val_label.txt')
py.arg('--load_size', type=int, default=143)
py.arg('--crop_size', type=int, default=128)
py.arg('--n_epochs', type=int, default=60)
py.arg('--epoch_start_decay', type=int, default=30)
py.arg('--batch_size', type=int, default=32)
import imlib as im
import numpy as np
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--flask_path', default='/var/www/html/flaskapp_unigan')
py.arg('--generator_pb', default='generator_unigan_gender_only_beta_0.5.pb')
py.arg('--img_dir', default='./data/zappos_50k/images')
py.arg('--test_label_path', default='./data/zappos_50k/test_label.txt')
py.arg('--test_att_name', choices=data.ATT_ID.keys(), default='Women')
py.arg('--test_int_min', type=float, default=-2)
py.arg('--test_int_max', type=float, default=2)
py.arg('--test_int_step', type=float, default=0.5)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = os.path.join(args_.flask_path, py.join('output', args_.experiment_name))
# output_dir = py.join('output', args_.experiment_name)
import imlib as im
import numpy as np
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--flask_path', default='/var/www/html/flaskapp_unigan')
py.arg('--img_dir', default='./data/zappos_50k/images')
py.arg('--test_label_path', default='./data/zappos_50k/test_label.txt')
py.arg('--test_att_names',
choices=data.ATT_ID.keys(),
nargs='+',
default=['Men', 'Women'])
py.arg('--test_ints', type=float, nargs='+', default=2)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = os.path.join(args_.flask_path,
py.join('output', args_.experiment_name))
# output_dir = py.join('output', args_.experiment_name)
import tensorflow as tf
import tensorflow.keras as keras
import tf2lib as tl
import tf2gan as gan
from PIL import Image
import tqdm
import matplotlib.pyplot as plt
import data
import random
import module
import tensorflow_datasets as tfds
# ==============================================================================
# = param =
# ==============================================================================
'''py.arg('--dataset', default='summer2winter_yosemite')
py.arg('--datasets_dir', default='dataset')
py.arg('--load_size', type=int, default=256) # load image to this size
py.arg('--crop_size', type=int, default=256) # then crop to this size
py.arg('--batch_size', type=int, default=1)
py.arg('--epochs', type=int, default=200)
py.arg('--epoch_decay', type=int, default=100) # epoch to start decaying learning rate
py.arg('--lr', type=float, default=0.0002)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--adversarial_loss_mode', default='lsgan', choices=['gan', 'hinge_v1', 'hinge_v2', 'lsgan', 'wgan'])
py.arg('--gradient_penalty_mode', default='none', choices=['none', 'dragan', 'wgan-gp'])
py.arg('--gradient_penalty_weight', type=float, default=10.0)
py.arg('--cycle_loss_weight', type=float, default=10.0)
py.arg('--identity_loss_weight', type=float, default=0.0)
py.arg('--pool_size', type=int, default=50) # pool size to store fake samples
args = py.args()'''
import tqdm
import data
import module
# ==============================================================================
# = environ =
# ==============================================================================
os.environ["CUDA_PATH"] = "/usr/local/cuda"
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--flask_path', default='/var/www/html/flaskapp')
py.arg(
'--img_dir',
default=
'./data/img_celeba/aligned/align_size(572,572)_move(0.250,0.000)_face_factor(0.450)_jpg/data'
)
py.arg('--test_label_path', default='./data/img_celeba/test_label.txt')
py.arg('--test_int', type=float, default=2)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = os.path.join(args_.flask_path,
py.join('output', args_.experiment_name))
import imlib as im
import numpy as np
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--img_dir', default='./data/zappos_50k/images')
py.arg('--test_label_path', default='./data/zappos_50k/test_label.txt')
py.arg('--test_int', type=float, default=2)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = py.join('output', args_.experiment_name)
# save settings
args = py.args_from_yaml(py.join(output_dir, 'settings.yml'))
args.__dict__.update(args_.__dict__)
# others
n_atts = len(args.att_names)
import tensorflow as tf
import Utils as utils
import cv2
import data
import module
from os.path import join,exists,basename
from os import makedirs,mkdir
import os
def normalize(arr,eps=0.000001):
return 2*((arr-np.min(arr))/(np.max(arr)-np.min(arr)+eps))-1
os.environ["CUDA_DEVICE_ORDER"]="PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"]="0"
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--datasets_dir', default='./datasets')
py.arg('--experiment_dir',default='./Results')
py.arg('--batch_size', type=int, default=1)
test_args = py.args()
args = py.args_from_yaml(join(test_args.experiment_dir, 'settings.yml'))
args.__dict__.update(test_args.__dict__)
# ==============================================================================
# = test =
# ==============================================================================
# data
A_img_paths_test = glob.glob(join(args.datasets_dir, 'MRI_test', '*.png'))
# print(len(A_img_paths_test))
# B_img_paths_test = py.glob(py.join(args.datasets_dir, args.dataset, 'CT_test'), '*.png')
import pylib as py
import imlib as im
import torch
import numpy as np
import torchlib
import module
import data
py.arg(
'--dataset',
default='fashion_mnist',
choices=['cifar10', 'fashion_mnist', 'mnist', 'celeba', 'anime', 'custom'])
# py.arg('--experiment_name', required=True)
py.arg('--checkpoint_name', default="Epoch_inter(50).ckpt")
py.arg('--z_dim', type=int, default=128)
py.arg('--num_samples', type=int, required=True)
py.arg('--batch_size', type=int, default=1)
py.arg('--experiment_names', nargs='+', type=str)
py.arg('--output_dir', type=str, default='generated_imgs')
args = py.args()
print(args.experiment_names)
experiment_names = args.experiment_names
use_gpu = torch.cuda.is_available()
device = torch.device("cuda" if use_gpu else "cpu")
torch.manual_seed(0)
if args.dataset in ['cifar10', 'fashion_mnist', 'mnist', 'imagenet']: # 32x32
output_channels = 3
import imlib as im
import numpy as np
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--img_dir', default='./data/zappos_50k/images')
py.arg('--test_label_path', default='./data/zappos_50k/test_label.txt')
py.arg('--test_att_names',
choices=data.ATT_ID.keys(),
nargs='+',
default=['Unisex', 'Athletics'])
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = py.join('output', args_.experiment_name)
# save settings
args = py.args_from_yaml(py.join(output_dir, 'settings.yml'))
args.__dict__.update(args_.__dict__)
import pylib as py
import imlib as im
import torch
import numpy as np
import torchlib
from torchvision import datasets, transforms
from torch.utils.data import DataLoader, Dataset, sampler
import torchvision.utils as vutils
import os
import module
import data
py.arg(
'--dataset',
default='fashion_mnist',
choices=['cifar10', 'fashion_mnist', 'mnist', 'celeba', 'anime', 'custom'])
py.arg('--out_dir', required=True)
py.arg('--num_samples_per_class', type=int, required=True)
py.arg('--batch_size', type=int, default=1)
py.arg('--num', type=int, default=-1)
py.arg('--output_dir', type=str, default='generated_imgs')
args = py.args()
use_gpu = torch.cuda.is_available()
device = torch.device("cuda" if use_gpu else "cpu")
py.mkdir(args.out_dir)
transform = transforms.Compose([
import pylib as py
import tensorflow as tf
import tflib as tl
import module
from tensorflow.python.framework import graph_util
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = py.join('output', args_.experiment_name)
# save settings
args = py.args_from_yaml(py.join(output_dir, 'settings.yml'))
args.__dict__.update(args_.__dict__)
# others
n_atts = len(args.att_names)
sess = tl.session()
sess.__enter__() # make default
# ==============================================================================
import imlib as im
import numpy as np
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--img_dir', default='./data/zappos_50k/images')
py.arg('--test_label_path', default='./data/zappos_50k/test_label.txt')
py.arg('--test_att_name', choices=data.ATT_ID.keys(), default='Unisex')
py.arg('--test_int_min', type=float, default=-2)
py.arg('--test_int_max', type=float, default=2)
py.arg('--test_int_step', type=float, default=0.5)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = py.join('output', args_.experiment_name)
# save settings
args = py.args_from_yaml(py.join(output_dir, 'settings.yml'))
args.__dict__.update(args_.__dict__)
import numpy as np
import pylib as py
import tensorflow as tf
import tensorflow.keras as keras
import tf2lib as tl
import tf2gan as gan
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--datasets_dir', default='datasets')
py.arg('--output_dir', default='output')
py.arg('--load_size', type=int, default=286) # load image to this size
py.arg('--crop_size', type=int, default=256) # then crop to this size
py.arg('--batch_size', type=int, default=1)
py.arg('--epochs', type=int, default=200)
py.arg('--epoch_decay', type=int,
default=100) # epoch to start decaying learning rate
py.arg('--lr', type=float, default=0.0002)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--norm_type',
default='instance_norm',
choices=['none', 'batch_norm', 'instance_norm', 'layer_norm'])
py.arg('--adversarial_loss_mode',
default='lsgan',
choices=['gan', 'hinge_v1', 'hinge_v2', 'lsgan', 'wgan'])
def weights_init(m):
classname = m.__class__.__name__
if classname.find('ConvTranspose') != -1:
m.weight.data.normal_(0.0, 0.02)
elif classname.find('Conv2d') != -1:
m.weight.data.normal_(0.0, 0.02)
elif classname.find('BatchNorm') != -1:
m.weight.data.normal_(1.0, 0.02)
m.bias.data.fill_(0)
# ==============================================================================
# = param =
# ==============================================================================
# command line
py.arg('--dataset', default='fashion_mnist',
choices=['cifar10', 'fashion_mnist', 'mnist', 'celeba', 'anime', 'custom', 'imagenet'])
py.arg('--batch_size', type=int, default=256)
py.arg('--epochs', type=int, default=100)
py.arg('--lr', type=float, default=0.0002)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--n_d', type=int, default=1) # # d updates per g update
py.arg('--z_dim', type=int, default=128)
py.arg('--adversarial_loss_mode', default='gan',
choices=['gan', 'hinge_v1', 'hinge_v2', 'lsgan', 'wgan'])
py.arg('--gradient_penalty_mode', default='none',
choices=['none', '1-gp', '0-gp', 'lp'])
py.arg('--gradient_penalty_sample_mode', default='line',
choices=['line', 'real', 'fake', 'dragan'])
py.arg('--gradient_penalty_weight', type=float, default=10.0)
py.arg('--experiment_name', default='none')
py.arg('--gradient_penalty_d_norm', default='layer_norm',
# CUDA_VISIBLE_DEVICES=0 python3 train.py --dataset=custom --custom_dataroot ./data/car_renderings_baseline --updown_sampling 6 --im_size 256 --epoch=200 --adversarial_loss_mode=gan --batch_size 150 --experiment_name car_dcgan
# CUDA_VISIBLE_DEVICES=1 python3 train.py --dataset=custom --custom_dataroot ./data/car_renderings_baseline --updown_sampling 6 --im_size 256 --epoch=200 --adversarial_loss_mode=lsgan --batch_size 150 --experiment_name car_lsgan
# CUDA_VISIBLE_DEVICES=2 python3 train.py --dataset=custom --custom_dataroot ./data/car_renderings_baseline --updown_sampling 6 --im_size 256 --epoch=200 --adversarial_loss_mode=wgan --batch_size 110 --gradient_penalty_d_norm instance_norm --gradient_penalty_mode 1-gp --n_d 5 --experiment_name car_wgangp
# CUDA_VISIBLE_DEVICES=3 python3 train.py --dataset=custom --custom_dataroot ./data/chair_renderings_baseline --updown_sampling 6 --im_size 256 --epoch=400 --adversarial_loss_mode=gan --batch_size 150 --experiment_name chair_dcgan
# CUDA_VISIBLE_DEVICES=4 python3 train.py --dataset=custom --custom_dataroot ./data/chair_renderings_baseline --updown_sampling 6 --im_size 256 --epoch=400 --adversarial_loss_mode=lsgan --batch_size 150 --experiment_name chair_lsgan
# CUDA_VISIBLE_DEVICES=5 python3 train.py --dataset=custom --custom_dataroot ./data/chair_renderings_baseline --updown_sampling 6 --im_size 256 --epoch=400 --adversarial_loss_mode=wgan --batch_size 110 --gradient_penalty_d_norm instance_norm --gradient_penalty_mode 1-gp --n_d 5 --experiment_name chair_wgangp
# ==============================================================================
# = param =
# ==============================================================================
# command line
py.arg(
'--dataset',
default='fashion_mnist',
choices=['cifar10', 'fashion_mnist', 'mnist', 'celeba', 'anime', 'custom'])
py.arg('--custom_dataroot',
default='./data',
help='the data root in custom dataset mode')
py.arg('--updown_sampling',
type=int,
default=6,
help='3 for 32x32, 4 for 64x64, etc')
py.arg('--im_size', type=int, default=256, help='image size')
py.arg('--batch_size', type=int, default=64)
py.arg('--epochs', type=int, default=25)
py.arg('--lr', type=float, default=0.0002)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--n_d', type=int, default=1) # # d updates per g update
py.arg('--z_dim', type=int, default=128)
import numpy as np
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg(
'--img_dir',
default=
'./data/img_celeba/aligned/align_size(572,572)_move(0.250,0.000)_face_factor(0.450)_jpg/data'
)
py.arg('--test_label_path', default='./data/img_celeba/test_label.txt')
py.arg('--test_att_name', choices=data.ATT_ID.keys(), default='Pale_Skin')
py.arg('--test_int_min', type=float, default=-2)
py.arg('--test_int_max', type=float, default=2)
py.arg('--test_int_step', type=float, default=0.5)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = py.join('output', args_.experiment_name)
# save settings
import model
import data_generator
import os
import pylib as py
import json
#Add code for last entry data file. (For systematic number of LOGS and CHECKPOINTS.
#Add code for parsing through JSON files to find already trained code.
#py.arg('--prev_train_data', default = "./Previous Train Data")
py.arg('--dataset', default="./Data")
py.arg('--batch_size', type=int, default=4)
py.arg('--epochs', type=int, default=100)
py.arg('--lr', type=float, default=0.002)
py.arg('--image_size', type=int, default=256)
py.arg('--n_channels', type=int, default=3)
py.arg('--shuffle_data', type=bool, default=True)
py.arg('--bottleneck_size', type=int, default=None)
py.arg('--loss_weight', type=float, default=0.8)
py.arg('--checkpoint_dir', default=None)
py.arg('--tensorboard_dir', default="./logs")
py.arg('--prev_checkpoint', default=None)
py.arg('--prev_tensorboard', default=None)
args = py.args()
##Finding previously trained model. [CODE NOT COMPLETED]
#json_directory = "args.prev_train_data"
#json_list = os.listdir(json_directory)
#for json_file in json_list:
# with open(os.path.join(json_directory, json_file)) as f:
# metadata = json.load(f)
config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)
#os.environ["CUDA_VISIBLE_DEVICES"] = '2,3'
tf.config.experimental.set_lms_enabled(True)
#neptune.set_project('Serre-Lab/paleo-ai')
#GPUS to be used
GPU = [2, 3]
# ==============================================================================
# = param =
# ==============================================================================
py.arg(
'--outdir',
default=
'/users/irodri15/data/irodri15/Fossils/Experiments/cyclegan/checkpoints/')
py.arg(
'--train_datasetA',
default=
'/users/irodri15/data/irodri15/Fossils/Experiments/datasets/gan_fossils_leaves_v1/fossils_train_oscar_processed.csv'
)
py.arg(
'--train_datasetB',
default=
'/users/irodri15/data/irodri15/Fossils/Experiments/datasets/gan_fossils_leaves_v1/leaves_train_oscar_processed.csv'
)
py.arg(
'--test_datasetA',
default=
'/users/irodri15/data/irodri15/Fossils/Experiments/datasets/gan_fossils_leaves_v1/fossils_test_oscar_processed.csv'
import pylib as py
import tensorflow as tf
import tensorflow.keras as keras
import tf2gan as gan
import tf2lib as tl
import tqdm
import matplotlib.pyplot as plt
import sys
import numpy as np
# ==============================================================================
# = param =
# ==============================================================================
# command line
py.arg('--dataset', default='PETCT', choices=['PETCT', 'cifar10', 'fashion_mnist', 'mnist', 'celeba', 'anime', 'custom'])
py.arg('--batch_size', type=int, default=5)
py.arg('--epochs', type=int, default=25)
py.arg('--lr', type=float, default=0.0002)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--n_d', type=int, default=1) # # d updates per g update
py.arg('--PETCT_dim', type=int, default=128)
py.arg('--adversarial_loss_mode', default='wgan', choices=['gan', 'hinge_v1', 'hinge_v2', 'lsgan', 'wgan'])
py.arg('--gradient_penalty_mode', default='none', choices=['none', 'dragan', 'wgan-gp'])
py.arg('--gradient_penalty_weight', type=float, default=10.0)
py.arg('--experiment_name', default='PETCT')
py.arg('--data_rate', type=float, default=0.0)
py.arg('--training_mode', type=int, default=0)
py.arg('--padding', default='same', choices=['same', 'valid', 'full'])
py.arg('--input_size',type=int, default=256)
args = py.args()
except RuntimeError as e:
print(e)
import tensorflow.keras as keras
import tf2lib as tl
import tf2gan as gan
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
# python train.py --dataset horse2zebra --epochs 10 --batch_size 500
py.arg('--dataset', default='horse2zebra')
py.arg('--datasets_dir', default='datasets')
py.arg('--load_size', type=int, default=286) # load image to this size
py.arg('--crop_size', type=int, default=256) # then crop to this size
py.arg('--batch_size', type=int, default=1000)
py.arg('--epochs', type=int, default=2)
py.arg('--epoch_decay', type=int,
default=100) # epoch to start decaying learning rate
py.arg('--lr', type=float, default=0.0002)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--adversarial_loss_mode',
default='lsgan',
choices=['gan', 'hinge_v1', 'hinge_v2', 'lsgan', 'wgan'])
py.arg('--gradient_penalty_mode',
default='none',
choices=['none', 'dragan', 'wgan-gp'])
import numpy as np
import pylib as py
import tensorflow as tf
import tensorflow.keras as keras
import tf2lib as tl
import tf2gan as gan
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg("--dataset", default="horse2zebra")
py.arg("--output_dir", default="horse2zebra")
py.arg("--datasets_dir", default="datasets")
py.arg("--load_size", type=int, default=286) # load image to this size
py.arg("--crop_size", type=int, default=256) # then crop to this size
py.arg("--channels", type=int, default=1) # 3 for RGB, 1 for grayscale
py.arg("--batch_size", type=int, default=1)
py.arg("--epochs", type=int, default=100)
py.arg("--epoch_decay", type=int,
default=50) # epoch to start decaying learning rate
py.arg("--lr", type=float, default=0.0002)
py.arg("--beta_1", type=float, default=0.5)
py.arg(
"--adversarial_loss_mode",
default="lsgan",
choices=["gan", "hinge_v1", "hinge_v2", "lsgan", "wgan"],
import imlib as im
import module
import pylib as py
import tensorflow as tf
import tensorflow.keras as keras
import tf2gan as gan
import tf2lib as tl
import tqdm
import matplotlib.pyplot as plt
# ==============================================================================
# = param =
# ==============================================================================
# command line
py.arg('--dataset', default='fashion_mnist', choices=['cifar10', 'fashion_mnist', 'mnist', 'celeba', 'anime', 'custom'])
py.arg('--batch_size', type=int, default=64)
py.arg('--epochs', type=int, default=25)
py.arg('--lr', type=float, default=0.0002)
py.arg('--beta_1', type=float, default=0.5)
py.arg('--n_d', type=int, default=1) # # d updates per g update
py.arg('--z_dim', type=int, default=128)
py.arg('--adversarial_loss_mode', default='gan', choices=['gan', 'hinge_v1', 'hinge_v2', 'lsgan', 'wgan'])
py.arg('--gradient_penalty_mode', default='none', choices=['none', 'dragan', 'wgan-gp'])
py.arg('--gradient_penalty_weight', type=float, default=10.0)
py.arg('--experiment_name', default='none')
args = py.args()
# output_dir
if args.experiment_name == 'none':
args.experiment_name = '%s_%s' % (args.dataset, args.adversarial_loss_mode)
import os
import imlib as im
import numpy as np
import pylib as py
import scipy
import tensorflow as tf
import tflib as tl
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg('--n_traversal', type=int, default=100)
py.arg('--n_traversal_point', type=int, default=17)
py.arg('--truncation_threshold', type=float, default=1.5)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = py.join('output', args_.experiment_name)
# save settings
args = py.args_from_yaml(py.join(output_dir, 'settings.yml'))
args.__dict__.update(args_.__dict__)
sess = tl.session()
import numpy as np
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg(
'--img_dir',
default=
'./data/img_celeba/aligned/align_size(572,572)_move(0.250,0.000)_face_factor(0.450)_jpg/data'
)
py.arg('--test_label_path', default='./data/img_celeba/test_label.txt')
py.arg('--test_att_names',
choices=data.ATT_ID.keys(),
nargs='+',
default=['Bangs', 'Mustache'])
py.arg('--test_ints', type=float, nargs='+', default=2)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = py.join('output', args_.experiment_name)
import numpy as np
import pylib as py
import tensorflow as tf
import tflib as tl
import tqdm
import data
import module
# ==============================================================================
# = param =
# ==============================================================================
py.arg(
'--img_dir',
default=
'./data/img_celeba/aligned/align_size(572,572)_move(0.250,0.000)_face_factor(0.450)_jpg/data'
)
py.arg('--test_label_path', default='./data/img_celeba/test_label.txt')
py.arg('--with_mask', default=False)
py.arg('--test_int', type=float, default=2)
py.arg('--experiment_name', default='default')
args_ = py.args()
# output_dir
output_dir = py.join('output', args_.experiment_name)
# save settings
args = py.args_from_yaml(py.join(output_dir, 'settings.yml'))
args.__dict__.update(args_.__dict__)