test_dataug Namespace Reference

Variables
list	tf_names
	device = torch.device('cuda')
string	device_name = 'CPU'
dictionary	tasks
int	n_inner_iter = 1
int	epochs = 200
int	dataug_epoch_start = 0
dictionary	optim_param
	model = LeNet(3,10)
	t0 = time.process_time()
	Classic ####. More...
	log = train_classic(model=model, opt_param=optim_param, epochs=epochs, print_freq=1)
	exec_time = time.process_time() - t0
list	times = [x["time"] for x in log]
dictionary	out = {"Accuracy": max([x["acc"] for x in log]), "Time": (np.mean(times),np.std(times), exec_time), 'Optimizer': optim_param['Inner'], "Device": device_name, "Log": log}
string	filename = "{}-{} epochs".format(str(model),epochs)
	f
	indent
	fig_name
dictionary	tf_dict = {k: TF.TF_dict[k] for k in tf_names}
	aug_model = Augmented_model(Data_augV7(TF_dict=tf_dict, N_TF=3, mix_dist=0.8, fixed_prob=False, fixed_mag=False, shared_mag=False), model).to(device)
	param_names

Detailed Description

Script to run experiment on smart augmentation.

Variable Documentation

optim_param

dictionary test_dataug.optim_param

Initial value:

= {
        'Meta':{
            'optim':'Adam',
            'lr':1e-2, #1e-2
        },
        'Inner':{
            'optim': 'SGD',
            'lr':1e-2, #1e-2
            'momentum':0.9, #0.9
        }
    }

t0

test_dataug.t0 = time.process_time()

Classic ####.

Augmented Dataset ####.

Augmented Model ####

tasks

dictionary test_dataug.tasks

Initial value:

= {
        #'classic',
        #'aug_dataset', #Moved to old code
        'aug_model'
    }