def demo_test(args):
if args.doc:
args = config_loader(args.doc, args)
# config
# model_config(args, save=False) # print model configuration of evaluation
# set cuda
torch.cuda.set_device(args.gpu_id)
# model
model = model_builder(args.model_name, args.scale, **args.model_args).cuda()
# criteriohn
criterion = criterion_builder(args.criterion)
# dataset
test_set = AxisDataSet(args.test_path, args.target_path)
test_loader = DataLoader(test_set,
batch_size=args.batch_size,
shuffle=False,
num_workers=args.num_workers,
# pin_memory=True,
pin_memory=False,
)
# test
test(model, test_loader, criterion, args)
def model_env(args):
"""building model environment avoiding to instantiate model.
Args:
args : model arguments which is control by demo_utils.argument_setting
Returns:
model (torch.nn): build model in cuda device
criterion(torch.nn): build criterion. Default to mse loss
extractor(torch.nn): build vgg content loss in cuda device
"""
if args.doc:
args = config_loader(args.doc, args)
# set cuda device
torch.cuda.set_device(args.gpu_id)
# model version control
version = args.load if type(args.load) is int else 0
# model path and parameter
model_path = os.path.join(
args.log_path, args.model_name, f'version_{version}',f'{args.model_name}_{args.scale}x.pt')
checkpoint = torch.load(model_path, map_location=f'cuda:{args.gpu_id}')
# loading model
model = model_builder(args.model_name, args.scale, **args.model_args).cuda()
model.load_state_dict(checkpoint['state_dict'])
# build criterion
criterion = criterion_builder(args.criterion)
# loading feature extractor
extractor = FeatureExtractor().cuda() if args.content_loss else None
return model, criterion, extractor
import uuid
import json
import time
import traceback
"""
Note: Adding eventlet and monkey_patching it to ensure that emit messages are not
accumulated and sent out in bursts
"""
import eventlet
eventlet.monkey_patch()
"""
Load config
"""
config = config_loader()
app = Flask(__name__)
socketio = SocketIO(app)
@socketio.on('authenticate')
def handle_authenticate(args):
"""
Handles authentication for a particular
API_KEY and challenge_id pair
Request Params:
API_KEY : String
Holds the API Key of the participant
challenge_id : String
'state_dict': model.state_dict(),
'epoch': epoch,
'train_iter': checkpoint['train_iter'],
'valid_iter': checkpoint['valid_iter'],
}, model_path)
writer.close()
if __name__ == '__main__':
# argument setting
train_args = train_argument()
# replace args by document file
if train_args.doc:
train_args = config_loader(train_args.doc, train_args)
# set cuda
torch.cuda.set_device(train_args.gpu_id)
# model
model = model_builder(train_args.model_name, train_args.scale,
**train_args.model_args).cuda()
# optimizer and critera
optimizer = optimizer_builder(train_args.optim) # optimizer class
optimizer = optimizer( # optmizer instance
model.parameters(),
lr=train_args.lr,
weight_decay=train_args.weight_decay)
criterion = criterion_builder(train_args.criterion)
content_loss = args.beta * criterion(gen_feature, real_feature)
# for compatible
loss = content_loss + mse_loss
err += loss.sum().item() * inputs.size(0)
err /= len(test_loader.dataset)
print(f'test error:{err:.4f}')
if __name__ == '__main__':
# argument setting
test_args = test_argument()
if test_args.doc:
test_args = config_loader(test_args.doc, test_args)
# config
model_config(test_args,
save=False) # print model configuration of evaluation
# set cuda
torch.cuda.set_device(test_args.gpu_id)
# model
model = model_builder(test_args.model_name, test_args.scale,
**test_args.model_args).cuda()
# criteriohn
criterion = criterion_builder(test_args.criterion)
# optimizer = None # don't need optimizer in test
def __init__(self):
self.clean = Clean()
self.despesas_getter = Despesas()
self.config = config_loader('configuration.yaml')
self.data = self.get_data()
def __init__(self):
self.config = config_loader('configuration.yaml')
def __init__(self, data: Optional[datetime] = None):
self.data = data if data else datetime.today()
self.config = config_loader('configuration.yaml')
