def preprocess():
"""Preprocess."""
print("Transform images")
client = storage.Client(PROJECT)
read_bucket = client.get_bucket(RAW_BUCKET)
write_bucket = client.get_bucket(BUCKET)
transformations = transforms.Compose([Rescale(256), RandomCrop(224)])
metadata_df = (
pd.read_csv(f"gs://{RAW_BUCKET}/{RAW_DATA_DIR}/metadata.csv",
usecols=["filename", "view"
]).query("view == 'PA'") # taking only PA view
)
start_time = time.time()
for filename in metadata_df["filename"].tolist():
image = gcs_imread(read_bucket,
os.path.join(RAW_DATA_DIR, "images", filename))
proc_image = transformations(image)
proc_image = (proc_image * 255).astype(np.uint8)
gcs_imwrite(write_bucket,
os.path.join(BASE_DIR, PREPROCESSED_DIR, filename),
filename, proc_image)
print(f" Time taken = {time.time() - start_time:.2f}s")
print(f" Number of images processed = {metadata_df.shape[0]}")
def process_image(image):
"""Process image."""
seed_torch(seed=42)
proc_image = transforms.Compose([
Rescale(256),
RandomCrop(224),
ToTensor(),
])(image)
proc_image = proc_image.unsqueeze(0).to(DEVICE)
return proc_image
def __init__(self):
self.cfg_json = "config/openi_progressive_hiach_test.json"
self.cfg = self.pare_cfg(self.cfg_json)
self.exp_name = self.cfg["EXPER_NAME"]
self.test_csv = self.cfg["TEST_CSV"]
self.encoder_resume = self.cfg["RESUME_ENCODER"]
self.decoder_resume_F = self.cfg["RESUME_DECODER_F"]
self.decoder_resume_L = self.cfg["RESUME_DECODER_L"]
self.word_dict = self.cfg["DICTIONARY"]
self.text_csv = self.cfg["TEXT_CSV"]
self.img_csv = self.cfg["IMG_CSV"]
self.data_root = self.cfg["DATA_ROOT"]
self.image_size = tuple(self.cfg["IMAGE_SIZE"])
self.name = self.cfg["EXPER_NAME"]
self.save_img_dir = './save_image/OPENI_HIAv3'
if os.path.exists(self.save_img_dir) == False:
os.mkdir(self.save_img_dir)
self.ENCODERS = {"HAttnEncoder": HAttnEncoder}
self.DECODERS = {
"baseDECODER": baseDecoder,
"baseDECODERv2": baseDecoderv2,
"baseDECODERv3": baseDecoderv3
}
self.P_DECODER = {
"PDECODER": PDecoder,
"PDECODERv2": PDecoderv2,
"PDECODERv3": PDecoderv3
}
self.DISCRIMINATOR = {
"baseDISCRIMINATOR": baseDiscriminator,
"noconDISCRIMINATOR": noCon_Discriminator,
"Patch": PatchDiscriminator,
"SNDiscriminator": SNDiscriminator,
"ResDISCRIMINATOR": ResDiscriminator,
"PDISCRIMINATOR": PDiscriminator
}
self.dataset = {
"OPENI": OpeniDataset2_Hiachy,
"MIMIC-CXR": MIMICDataset2_Hiachy
}
##################################################
################# Dataset Setup ##################
##################################################
self.t2i_dataset = self.dataset[self.cfg["DATASET"]](
csv_txt=self.text_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose(
[Rescale(self.image_size),
Equalize(), ToTensor()]))
self.testset = self.dataset[self.cfg["DATASET"]](
csv_txt=self.test_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose(
[Rescale(self.image_size),
Equalize(), ToTensor()]))
self.testset = self.t2i_dataset
self.device = torch.device(
"cuda" if torch.cuda.is_available() else "cpu")
s_gpus = self.cfg["GPU_ID"].split(',')
self.gpus = [int(ix) for ix in s_gpus]
self.num_gpus = len(self.gpus)
self.test_dataloader = DataLoader(self.testset,
batch_size=12,
shuffle=False,
num_workers=8,
drop_last=True)
self.base_size = 32
self.P_ratio = int(np.log2(self.image_size[0] // self.base_size))
self.base_ratio = int(np.log2(self.base_size))
print("Number of Decoders", self.P_ratio + 1)
print("Number of Discriminator", self.P_ratio + 1)
#########################################
############ Network Init ###############
#########################################
self.define_nets()
if self.num_gpus > 1:
self.encoder = nn.DataParallel(self.encoder, device_ids=self.gpus)
self.decoder_L = nn.DataParallel(self.decoder_L,
device_ids=self.gpus)
self.decoder_F = nn.DataParallel(self.decoder_F,
device_ids=self.gpus)
def __init__(self):
self.cfg_json = "config/MIMIC_wgan.json"
self.cfg = self.pare_cfg(self.cfg_json)
self.exp_name = self.cfg["EXPER_NAME"]
self.max_epoch = self.cfg["MAX_EPOCH"]
self.encoder_checkpoint = self.cfg["CHECKPOINT_ENCODER"]
self.decoder_checkpoint = self.cfg["CHECKPOINT_DECODER"]
self.D_checkpoint = self.cfg["CHECKPOINT_D"]
self.check_create_checkpoint()
self.encoder_resume = self.cfg["RESUME_ENCODER"]
self.decoder_resume_F = self.cfg["RESUME_DECODER_F"]
self.decoder_resume_L = self.cfg["RESUME_DECODER_L"]
self.D_resume = self.cfg["RESUME_D"]
self.train_csv = self.cfg["TRAIN_CSV"]
self.val_csv = self.cfg["VAL_CSV"]
self.test_csv = self.cfg["TEST_CSV"]
self.img_csv = self.cfg["IMG_CSV"]
self.data_root = self.cfg["DATA_ROOT"]
self.batch_size = self.cfg["BATCH_SIZE"]
self.image_size = tuple(self.cfg["IMAGE_SIZE"])
self.name = self.cfg["EXPER_NAME"]
self.pix_loss_ratio = self.cfg["PIXEL_LOSS_RATIO"]
self.adv_loss_ratio = self.cfg["ADV_LOSS_RATIO"]
self.checkpoint_epoch = self.cfg["CHECKPOINT_EPOCH"]
self.beta1 = self.cfg["beta1"]
self.word_dict = self.cfg["DICTIONARY"]
self.writer = SummaryWriter(os.path.join("runs", self.exp_name))
self.ENCODERS = {
"HAttnEncoder": HAttnEncoder,
}
self.DECODERS = {
"baseDECODER": baseDecoder,
"baseDECODERv2": baseDecoderv2,
"baseDECODERv3": baseDecoderv3
}
self.DISCRIMINATOR = {
"baseDISCRIMINATOR": baseDiscriminator,
"noconDISCRIMINATOR": noCon_Discriminator,
"Patch": PatchDiscriminator,
"SNDiscriminator": SNDiscriminator,
"ResDISCRIMINATOR": ResDiscriminator,
"PDISCRIMINATOR": PDiscriminator
}
self.dataset = {"OPENI": OpeniDataset2, "MIMIC-CXR": MIMICDataset2}
##################################################
################# Dataset Setup ##################
##################################################
self.trainset = self.dataset[self.cfg["DATASET"]](
csv_txt=self.train_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose(
[Rescale(self.image_size),
Equalize(), ToTensor()]))
self.valset = self.dataset[self.cfg["DATASET"]](
csv_txt=self.val_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose(
[Rescale(self.image_size),
Equalize(), ToTensor()]))
self.sia_dataset = self.dataset[self.cfg["DATASET"]][1](
csv_txt=self.train_csv,
csv_img=self.img_csv,
root=self.data_root,
transform=transforms.Compose(
[Rescale(self.image_size),
Equalize(), ToTensor()]))
self.device = torch.device(
"cuda" if torch.cuda.is_available() else "cpu")
s_gpus = self.cfg["GPU_ID"].split(',')
self.gpus = [int(ix) for ix in s_gpus]
self.num_gpus = len(self.gpus)
#########################################
############ Loss Function ##############
#########################################
content_losses = {"L2": nn.MSELoss(), "L1": nn.L1Loss()}
self.G_criterion = content_losses[self.cfg["CONTENT_LOSS"]].to(
self.device)
#########################################
############ Network Init ###############
#########################################
self.base_size = self.image_size[0]
self.base_ratio = int(np.log2(self.base_size))
self.define_nets()
if self.num_gpus > 1:
self.encoder = nn.DataParallel(self.encoder, device_ids=self.gpus)
self.decoder_L = nn.DataParallel(self.decoder_L,
device_ids=self.gpus)
self.decoder_F = nn.DataParallel(self.decoder_F,
device_ids=self.gpus)
f.write('\n')
cfg = dict()
cfg['batch_size'] = 64
cfg['scale'] = 0.5
if cfg['scale'] == 0.5:
mnet_v2_mean = [0.4679]
mnet_v2_std = [0.2699]
else:
mnet_v2_mean = [0.4679]
mnet_v2_std = [0.2699]
train_set = OpenEDS(root_path=root_path + 'train',
transform=transforms.Compose(
[Rescale(cfg['scale']), ToTensor(), Normalize(mnet_v2_mean, mnet_v2_std)])) #
val_set = OpenEDS(root_path=root_path + 'validation',
transform=transforms.Compose(
[Rescale(cfg['scale']), ToTensor(), Normalize(mnet_v2_mean, mnet_v2_std)])) #
test_set = OpenEDS(root_path=root_path + 'test',
transform=transforms.Compose(
[Rescale(cfg['scale']), ToTensor(), Normalize(mnet_v2_mean, mnet_v2_std)])) #
loaders = {'train': torch.utils.data.DataLoader(train_set, batch_size=args.batch_size, shuffle=True,
num_workers=args.num_workers),
'val': torch.utils.data.DataLoader(val_set, batch_size=args.batch_size, shuffle=False,
num_workers=args.num_workers, pin_memory=False),
'test': torch.utils.data.DataLoader(test_set, batch_size=args.batch_size, shuffle=False,
num_workers=args.num_workers, pin_memory=False)
}
def __init__(self):
self.cfg_json = "config/MIMIC_StackGANtest.json"
self.cfg = self.pare_cfg(self.cfg_json)
self.exp_name = self.cfg["EXPER_NAME"]
self.encoder_resume = self.cfg["RESUME_ENCODER"]
self.decoder_resume_F = self.cfg["RESUME_DECODER_F"]
self.decoder_resume_L = self.cfg["RESUME_DECODER_L"]
self.word_dict = self.cfg["DICTIONARY"]
self.text_csv = self.cfg["TEXT_CSV"]
self.img_csv = self.cfg["IMG_CSV"]
self.data_root = self.cfg["DATA_ROOT"]
self.image_size = tuple(self.cfg["IMAGE_SIZE"])
self.name = self.cfg["EXPER_NAME"]
self.test_csv = self.cfg["TEST_CSV"]
self.save_img_dir = './save_image/MIMIC_Stack256'
if os.path.exists(self.save_img_dir)==False:
os.mkdir(self.save_img_dir)
self.ENCODERS = {
"baseENCODER": baseEncoder,
"baseENCODERv2": baseEncoderv2,
"harchyENCODER": harchyEncoder
}
self.dataset = {
"OPENI": OpeniDataset2,
"MIMIC-CXR": MIMICDataset2
}
##################################################
################# Dataset Setup ##################
##################################################
self.t2i_dataset = self.dataset[self.cfg["DATASET"]](csv_txt=self.text_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose([
Rescale(self.image_size),
Equalize(),
ToTensor()
]))
self.testset = self.dataset[self.cfg["DATASET"]](csv_txt=self.test_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose([
Rescale(self.image_size),
Equalize(),
ToTensor()
]))
self.testset = self.t2i_dataset
self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
s_gpus = self.cfg["GPU_ID"].split(',')
self.gpus = [int(ix) for ix in s_gpus]
self.num_gpus = len(self.gpus)
self.test_dataloader = DataLoader(self.testset,
batch_size=12,
shuffle=False,
num_workers=8,
drop_last=True)
self.base_size = self.image_size[0]
self.base_ratio = int(np.log2(self.base_size))
#########################################
############ Network Init ###############
#########################################
self.define_nets()
self.decoder_L= self.define_nets()
self.decoder_F= self.define_nets()
self.encoder = self.ENCODERS[self.cfg["ENCODER"]](vocab_size=self.t2i_dataset.vocab_size,
embed_size=self.cfg["E_EMBED_SIZE"],
hidden_size=self.cfg["E_HIDEN_SIZE"],
max_len=[self.t2i_dataset.max_len_finding,
self.t2i_dataset.max_len_impression],
unit=self.cfg["RNN_CELL"],
feature_base_dim=self.cfg["D_CHANNEL_SIZE"]
).to(self.device)
self.encoder = nn.DataParallel(self.encoder, device_ids=self.gpus)
def setup(self):
pkl_dir = self.config.split_dir
with open(os.path.join(pkl_dir, "splits.pkl"), 'rb') as f:
splits = pickle.load(f)
tr_keys = splits[self.config.fold]['train']
val_keys = splits[self.config.fold]['val']
test_keys = splits[self.config.fold]['test']
print("pkl_dir: ", pkl_dir)
print("tr_keys: ", tr_keys)
print("val_keys: ", val_keys)
print("test_keys: ", test_keys)
self.device = torch.device(
self.config.device if torch.cuda.is_available() else "cpu")
task = self.config.dataset_name
self.train_data_loader = torch.utils.data.DataLoader(
NucleusDataset(self.config.data_root_dir,
train=True,
transform=transforms.Compose([
Normalize(),
Rescale(self.config.patch_size),
ToTensor()
]),
target_transform=transforms.Compose([
Normalize(),
Rescale(self.config.patch_size),
ToTensor()
]),
mode="train",
keys=tr_keys,
taskname=task),
batch_size=self.config.batch_size,
shuffle=True)
self.val_data_loader = torch.utils.data.DataLoader(
NucleusDataset(self.config.data_root_dir,
train=True,
transform=transforms.Compose([
Normalize(),
Rescale(self.config.patch_size),
ToTensor()
]),
target_transform=transforms.Compose([
Normalize(),
Rescale(self.config.patch_size),
ToTensor()
]),
mode="val",
keys=val_keys,
taskname=self.config.dataset_name),
batch_size=self.config.batch_size,
shuffle=True)
self.test_data_loader = torch.utils.data.DataLoader(
NucleusDataset(self.config.data_root_dir,
train=True,
transform=transforms.Compose([
Normalize(),
Rescale(self.config.patch_size),
ToTensor()
]),
target_transform=transforms.Compose([
Normalize(),
Rescale(self.config.patch_size),
ToTensor()
]),
mode="test",
keys=test_keys,
taskname=self.config.dataset_name),
batch_size=self.config.batch_size,
shuffle=True)
self.model = UNet(num_classes=self.config.num_classes,
in_channels=self.config.in_channels)
#self.model = UNet()
self.model.to(self.device)
self.bce_weight = 0.5
self.optimizer = optim.Adam(self.model.parameters(),
lr=self.config.learning_rate)
self.scheduler = ReduceLROnPlateau(self.optimizer, 'min')
# If directory for checkpoint is provided, we load it.
if self.config.do_load_checkpoint:
if self.config.checkpoint_dir == '':
print(
'checkpoint_dir is empty, please provide directory to load checkpoint.'
)
else:
self.load_checkpoint(name=self.config.checkpoint_dir,
save_types=("model"))
self.save_checkpoint(name="checkpoint_start")
self.elog.print('Experiment set up.')
For now focus on wether two signatures are matches or not.
Afterwards add forgery detection to network.
"""
import torch
import torch.nn as nn
from torchvision import transforms
from torch.utils.data import DataLoader, Dataset
from utils import SignatureDataset, Rescale, ToTensor
from model import SigNet
import argparse
DATADIR = './data/sign_data/'
rescale = Rescale((100, 100))
# TODO Speed up data loader & add cuda support
train_set = SignatureDataset(root='./data/sign_data/',
split='train',
transforms=transforms.Compose(
[rescale, ToTensor()]))
test_set = SignatureDataset(root='./data/sign_Data/',
split='test',
transforms=transforms.Compose(
[rescale, ToTensor()]))
if __name__ == '__main__':
# TODO: Add argparser
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
# encoding: utf-8 ''' @author: andy @contact: [email protected] @github: https://github.com/AndyandViky @csdn: https://blog.csdn.net/AndyViky @file: compose-transforms.py @time: 2019/5/25 下午8:31 @desc: 使用compose转换图片格式，使用utils中的工具类 ''' import torchvision.transforms as transforms from utils import Rescale, RandomCrop, ToTensor from dataclass import MdataSample data = MdataSample("", "", transform=transforms.Compose( [Rescale(256), RandomCrop(224), ToTensor()]))
######################################
############### Main #################
######################################
# default settings
learning_rate = 0.001
batch_size = 4
num_classes = 2
log_schedule = 10
epochCount = 5
num_workers = 1
print_every = batch_size
transform = T.Compose([
Rescale(32),
T.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
train_data = MammogramDataset("Mini_DDSM_Upload", "train", transform=transform)
test_data = MammogramDataset("Mini_DDSM_Upload", "test")
VAL_RATIO = 0.2
NUM_VAL = int(len(train_data) * VAL_RATIO)
NUM_TRAIN = len(train_data) - NUM_VAL
NUM_TEST = len(test_data)
BATCH_SIZE = batch_size
loader_train = DataLoader(train_data,
batch_size=BATCH_SIZE,
def train(epochs, batch_size, learning_rate):
torch.manual_seed(1234)
train_loader = torch.utils.data.DataLoader(SegThorDataset(
"/home/WIN-UNI-DUE/smnemada/Master_Thesis/SegThor/data/train",
phase='train',
transform=transforms.Compose([Rescale(1.0),
Normalize(),
ToTensor()])),
batch_size=batch_size,
shuffle=True)
'''
# Loading validation data
val_set = SegThorDataset("/home/WIN-UNI-DUE/smnemada/Master_Thesis/SegThor/data_val", phase='val',
transform=transforms.Compose([
Rescale(0.5),
Normalize(),
ToTensor2()
]))
val_loader = torch.utils.data.DataLoader(dataset=val_set,
batch_size=1,
shuffle=False)
'''
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = UNet().to(device)
model.apply(weight_init)
#optimizer = optim.Adam(model.parameters(), lr=learning_rate) #learning rate to 0.001 for initial stage
optimizer = optim.SGD(model.parameters(),
lr=0.01,
momentum=0.9,
weight_decay=0.00001)
#optimizer = adabound.AdaBound(params = model.parameters(), lr = 0.001, final_lr = 0.1)
for epoch in range(epochs):
f = open('train_output.log', 'a')
f.write('Epoch {}/{}\n'.format(epoch + 1, epochs))
f.write('-' * 10)
running_loss = 0.0
running_loss_label = np.zeros(5)
for batch_idx, sample in enumerate(train_loader):
train_data, labels = sample['image'].to(
device,
dtype=torch.float), sample['label'].to(device,
dtype=torch.uint8)
optimizer.zero_grad()
output = model(train_data)
loss_label, loss = dice_loss2(output, labels)
loss.backward()
optimizer.step()
running_loss += loss.item()
for i in range(5):
running_loss_label[i] += loss_label[i]
epoch_loss = running_loss / len(train_loader)
writer.add_scalar('Train/Loss', epoch_loss, epoch)
f.write("\n Total Dice Loss: {:.4f}\n".format(epoch_loss))
epoch_loss_class = np.true_divide(running_loss_label,
len(train_loader))
f.write(
"Dice per class: Background = {:.4f} Eusophagus = {:.4f} Heart = {:.4f} Trachea = {:.4f} Aorta = {:.4f}\n"
.format(epoch_loss_class[0], epoch_loss_class[1],
epoch_loss_class[2], epoch_loss_class[3],
epoch_loss_class[4]))
#f.write("Dice per class: Background = {:.4f} Eusophagus = {:.4f}\n".format(epoch_loss_class[0], epoch_loss_class[1]))
f.close()
if epoch % 4 == 0:
os.makedirs("models", exist_ok=True)
torch.save(model, "models/model.pt")
# export scalar data to JSON for external processing
writer.export_scalars_to_json("./all_scalars.json")
writer.close()
os.makedirs("models", exist_ok=True)
torch.save(model, "models/model.pt")
def test():
test_path = '/home/WIN-UNI-DUE/smnemada/Master_Thesis/SegThor/data_sub/test'
for patient in tqdm(os.listdir(test_path)):
count = 0
area = 0
file = patient
x = file.split(".")
filename = x[0] + '.' + x[1]
print("patient = ", patient)
test_set = SegThorDataset(test_path,
patient=patient,
phase='test',
transform=transforms.Compose([
Rescale(1.0, labeled=False),
Normalize(labeled=False),
ToTensor(labeled=False)
]))
test_loader = torch.utils.data.DataLoader(dataset=test_set,
batch_size=1,
shuffle=False)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = torch.load("models/model.pt")
model.eval()
'''
with torch.no_grad():
for batch_idx, sample in enumerate(test_loader):
images = sample['image'].to(device, dtype=torch.float)
outputs = model(images)
print("tensor: {} and {} ".format(images.size(), outputs.size()))
images = tensor_to_numpy(images)
max_idx = torch.argmax(outputs, 1, keepdim=True)
max_idx = tensor_to_numpy(max_idx)
print("numpy: {} and {} ".format(images.shape, max_idx.shape))
fig=plt.figure()
fig.add_subplot(1,2,1)
plt.imshow(max_idx)
fig.add_subplot(1,2,2)
plt.imshow(images)
plt.show()
# fig.close()
count = count + 1
if count==150:
break
'''
# '''
seg_vol_2d = zeros([len(test_set), 512, 512])
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = torch.load("models/model.pt")
model.eval()
model.to(device)
with torch.no_grad():
for batch_idx, sample in enumerate(test_loader):
images = sample['image'].to(device, dtype=torch.float)
outputs = model(images)
images = tensor_to_numpy(images)
max_idx = torch.argmax(outputs, 1, keepdim=True)
max_idx = tensor_to_numpy(max_idx)
# for k in range(outputs.size(0)):
# print(max_idx.shape)
slice_v = max_idx[:,:]
slice_v = slice_v.astype(float32)
slice_v = ndimage.interpolation.zoom(slice_v, zoom=1, order=0, mode='nearest', prefilter=True)
seg_vol_2d[count,:,:] = slice_v
count = count + 1
segmentation = sitk.GetImageFromArray(seg_vol_2d, isVector=False)
print(segmentation.GetSize())
sitk.WriteImage(sitk.Cast( segmentation, sitk.sitkUInt8 ), filename, True)
cfg = dict()
cfg['batch_size'] = 64
cfg['scale'] = 0.5
# Original, mean 0.4679, std 0.2699
# Gamma correction: mean 0.3977, std 0.2307
if cfg['scale'] == 0.5:
mnet_v2_mean = [0.4679]
mnet_v2_std = [0.2699]
else:
mnet_v2_mean = [0.4679]
mnet_v2_std = [0.2699]
train_set = OpenEDS(root_path=root_path + 'train',
transform=transforms.Compose([
Rescale(cfg['scale']),
Brightness(brightness=(0.5, 2.75)),
ToTensor(),
Normalize(mnet_v2_mean, mnet_v2_std)
]))
val_set = OpenEDS(root_path=root_path + 'validation',
transform=transforms.Compose([
Rescale(cfg['scale']),
ToTensor(),
Normalize(mnet_v2_mean, mnet_v2_std)
])) #
test_set = OpenEDS(root_path=root_path + 'test',
transform=transforms.Compose([
Rescale(cfg['scale']),
def __init__(self):
self.cfg_json = "config/MIMIC_StackGAN.json"
self.cfg = self.pare_cfg(self.cfg_json)
self.exp_name = self.cfg["EXPER_NAME"]
self.max_epoch = self.cfg["MAX_EPOCH"]
self.encoder_checkpoint = self.cfg["CHECKPOINT_ENCODER"]
self.decoder_checkpoint = self.cfg["CHECKPOINT_DECODER"]
self.D_checkpoint = self.cfg["CHECKPOINT_D"]
self.check_create_checkpoint()
self.encoder_resume = self.cfg["RESUME_ENCODER"]
self.decoder_resume_F = self.cfg["RESUME_DECODER_F"]
self.decoder_resume_L = self.cfg["RESUME_DECODER_L"]
self.D_resume_F, self.D_resume_L = self.cfg["RESUME_D"]
self.train_csv = self.cfg["TRAIN_CSV"]
self.val_csv = self.cfg["VAL_CSV"]
self.test_csv = self.cfg["TEST_CSV"]
self.text_csv = self.cfg["TEXT_CSV"]
self.img_csv = self.cfg["IMG_CSV"]
self.data_root = self.cfg["DATA_ROOT"]
self.batch_size = self.cfg["BATCH_SIZE"]
self.image_size = tuple(self.cfg["IMAGE_SIZE"])
self.name = self.cfg["EXPER_NAME"]
self.pix_loss_ratio = self.cfg["PIXEL_LOSS_RATIO"]
self.adv_loss_ratio = self.cfg["ADV_LOSS_RATIO"]
self.checkpoint_epoch = self.cfg["CHECKPOINT_EPOCH"]
self.beta1 = self.cfg["beta1"]
self.word_dict = self.cfg["DICTIONARY"]
self.writer = SummaryWriter(os.path.join("runs", self.exp_name))
self.dataset = {"OPENI": OpeniDataset2, "MIMIC-CXR": MIMICDataset2}
##################################################
################# Dataset Setup ##################
##################################################
self.t2i_dataset = self.dataset[self.cfg["DATASET"]](
csv_txt=self.text_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose(
[Rescale(self.image_size),
Equalize(), ToTensor()]))
self.trainset = self.dataset[self.cfg["DATASET"]][0](
csv_txt=self.train_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose(
[Rescale(self.image_size),
Equalize(), ToTensor()]))
self.valset = self.dataset[self.cfg["DATASET"]][0](
csv_txt=self.val_csv,
csv_img=self.img_csv,
root=self.data_root,
word_dict=self.word_dict,
transform=transforms.Compose(
[Rescale(self.image_size),
Equalize(), ToTensor()]))
self.device = torch.device(
"cuda" if torch.cuda.is_available() else "cpu")
s_gpus = self.cfg["GPU_ID"].split(',')
self.gpus = [int(ix) for ix in s_gpus]
self.num_gpus = len(self.gpus)
#########################################
############ Loss Function ##############
#########################################
content_losses = {"L2": nn.MSELoss(), "L1": nn.L1Loss()}
self.G_criterion = content_losses[self.cfg["CONTENT_LOSS"]].to(
self.device)
#########################################
############ Network Init ###############
#########################################
self.decoder_L, self.D_L = self.define_nets()
self.decoder_F, self.D_F = self.define_nets()
self.encoder = harchyEncoder(
vocab_size=self.t2i_dataset.vocab_size,
embed_size=self.cfg["E_EMBED_SIZE"],
hidden_size=self.cfg["E_HIDEN_SIZE"],
max_len=[
self.t2i_dataset.max_len_finding,
self.t2i_dataset.max_len_impression
],
unit=self.cfg["RNN_CELL"],
feature_base_dim=self.cfg["D_CHANNEL_SIZE"]).to(self.device)
self.encoder = nn.DataParallel(self.encoder, device_ids=self.gpus)
for i in range(len(segthor_dataset)):
sample = segthor_dataset[i]
# print(i, sample['image'].size())
plt.imshow(sample['image'])
plt.show()
if i == 50:
break
'''
# '''
## Loading data for training phase
segthor_dataset = SegThorDataset(
datapath=
"/home/WIN-UNI-DUE/smnemada/Master_Thesis/SegThor/data_sub/train",
phase='train',
transform=transforms.Compose([
Rescale(1.0, labeled=True),
Normalize(labeled=True),
JointTransform2D(crop=(288, 288), p_flip=0.5),
ToTensor(labeled=True)
]))
for i in range(len(segthor_dataset)):
sample = segthor_dataset[i]
print(i, sample['image'].size(), sample['label'].size())
if i == 5:
break
# '''