Functions — github.com/a1600012888/YOPO-You-Only-Propagate-Once

hub / github.com/a1600012888/YOPO-You-Only-Propagate-Once / functions

Functions230 in github.com/a1600012888/YOPO-You-Only-Propagate-Once

⨍ Functions 230◇ Types & classes 65

↓ 56 callersMethodto

(self, device)

lib/attack/pgd.py:105

↓ 17 callersMethodcreate_evaluation_attack_method

evaluating the robustness of model against xxx adversary Return None means only measuring clean accuracy

lib/training/config.py:76

↓ 17 callersFunctioneval_one_epoch

(net, batch_generator, DEVICE=torch.device('cuda:0'), AttackMethod = None)

lib/training/train.py:72

↓ 16 callersFunctionload_checkpoint

(file_name, net = None, optimizer = None, lr_scheduler = None)

lib/utils/misc.py:85

↓ 16 callersFunctiontorch_accuracy

param output, target: should be torch Variable

lib/utils/misc.py:6

↓ 9 callersMethodcreate_lr_scheduler

(self, optimizer:torch.optim.Optimizer)

lib/training/config.py:59

↓ 9 callersMethodcreate_optimizer

params (iterable): iterable of parameters to optimize or dicts defining parameter groups

lib/training/config.py:51

↓ 9 callersFunctionsave_checkpoint

(now_epoch, net, optimizer, lr_scheduler, file_name)

lib/utils/misc.py:73

↓ 8 callersMethodcreate_loss_function

(self)

lib/training/config.py:63

↓ 7 callersMethodstep

Perform Iterative Sign Gradient on eta ret: inp + eta

experiments/CIFAR10/wide34.yopo-5-3/training_function.py:23

↓ 7 callersMethodstep

Perform Iterative Sign Gradient on eta ret: inp + eta

experiments/MNIST/YOPO-5-10/training_function.py:23

↓ 6 callersMethodstep

Perform Iterative Sign Gradient on eta ret: inp + eta

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/training_function.py:24

↓ 6 callersMethodstep

Perform Iterative Sign Gradient on eta ret: inp + eta

experiments/CIFAR10/pre-res18.yopo-5-3/training_function.py:23

↓ 5 callersMethodstep

(self, inp, p, eta)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/training_function.py:24

↓ 4 callersMethod_make_layer

(self, block, planes, num_blocks, stride)

lib/base_model/network.py:58

↓ 4 callersMethod_make_layer

(self, block, planes, num_blocks, stride)

lib/base_model/cifar_resnet18.py:82

↓ 4 callersMethod_make_layer

(self, block, planes, num_blocks, stride)

lib/base_model/preact_resnet.py:58

↓ 4 callersMethod_make_layer

(self, block, planes, num_blocks, stride)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/network.py:58

↓ 4 callersMethod_make_layer

(self, block, planes, num_blocks, stride)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/network.py:58

↓ 4 callersMethod_make_layer

(self, block, planes, num_blocks, stride)

experiments/CIFAR10/pre-res18.pgd10/network.py:58

↓ 4 callersMethod_make_layer

(self, block, planes, num_blocks, stride)

experiments/CIFAR10/pre-res18.yopo-5-3/network.py:58

↓ 4 callersMethodcreate_attack_method

Perform adversarial training against xxx adversary Return None means natural training

lib/training/config.py:67

↓ 4 callersMethodupdate

(self, mean_var, count=1)

lib/utils/misc.py:45

↓ 3 callersMethod__init__

(self, in_planes, planes, stride=1)

lib/base_model/network.py:16

↓ 3 callersMethod__init__

(self, in_planes, planes, stride=1)

lib/base_model/preact_resnet.py:15

↓ 3 callersMethod__init__

(self, in_planes, planes, stride=1)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/network.py:16

↓ 3 callersMethod__init__

(self, in_planes, planes, stride=1)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/network.py:16

↓ 3 callersMethod__init__

(self, in_planes, planes, stride=1)

experiments/CIFAR10/pre-res18.pgd10/network.py:16

↓ 3 callersMethod__init__

(self, in_planes, planes, stride=1)

experiments/CIFAR10/pre-res18.yopo-5-3/network.py:16

↓ 3 callersFunctioncreate_network

()

experiments/CIFAR10/wide34.natural/network.py:4

↓ 3 callersFunctioncreate_network

()

experiments/CIFAR10/wide34.pgd10/network.py:4

↓ 3 callersFunctioncreate_network

()

experiments/CIFAR10/wide34.yopo-5-3/network.py:4

↓ 3 callersFunctioncreate_network

()

experiments/MNIST/YOPO-5-10/network.py:63

↓ 3 callersFunctioncreate_network

()

experiments/MNIST/pgd40/network.py:5

↓ 3 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/wide34.natural/dataset.py:19

↓ 3 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/pre-res18.pgd10/dataset.py:19

↓ 3 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/pre-res18.yopo-5-3/dataset.py:19

↓ 3 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/wide34.pgd10/dataset.py:19

↓ 3 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/wide34.yopo-5-3/dataset.py:19

↓ 3 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/MNIST/YOPO-5-10/dataset.py:14

↓ 3 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/MNIST/pgd40/dataset.py:14

↓ 3 callersFunctionl2_norm

(x)

experiments/CIFAR10-TRADES/baseline.res-pre18.TRADES.10step/trades.py:14

↓ 3 callersMethodto

(self, device)

lib/attack/attack_base.py:17

↓ 3 callersFunctiontrain_one_epoch

:param attack_freq: Frequencies of training with adversarial examples. -1 indicates natural training :param AttackMethod: the attack method

experiments/CIFAR10/wide34.yopo-5-3/training_function.py:63

↓ 2 callersFunctionPreActResNet18

()

lib/base_model/network.py:107

↓ 2 callersFunctionPreActResNet18

()

lib/base_model/preact_resnet.py:104

↓ 2 callersFunctionPreActResNet18

()

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/network.py:96

↓ 2 callersFunctionPreActResNet18

()

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/network.py:96

↓ 2 callersFunctionPreActResNet18

()

experiments/CIFAR10/pre-res18.pgd10/network.py:107

↓ 2 callersFunctionPreActResNet18

()

experiments/CIFAR10/pre-res18.yopo-5-3/network.py:96

↓ 2 callersMethod__init__

(self, in_planes, out_planes, stride, dropRate=0.0)

lib/base_model/wide_resnet.py:18

↓ 2 callersMethod__init__

(self, block, num_blocks, num_classes=10)

lib/base_model/cifar_resnet18.py:70

↓ 2 callersMethodattack

(self, net, inp, label, target = None)

lib/attack/pgd.py:81

↓ 2 callersFunctioncal_l2_norm

(layer: torch.nn.Module)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/loss.py:37

↓ 2 callersFunctioncal_l2_norm

(layer: torch.nn.Module)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/loss.py:40

↓ 2 callersFunctionclip_eta

helper functions to project eta into epsilon norm ball :param eta: Perturbation tensor (should be of size(N, C, H, W)) :param norm: which

lib/attack/attack_base.py:22

↓ 2 callersFunctioncreate_network

()

experiments/CIFAR10-TRADES/baseline.res-pre18.TRADES.10step/network.py:5

↓ 2 callersFunctioncreate_network

()

experiments/CIFAR10/pre-res18.pgd10/network.py:142

↓ 2 callersFunctioncreate_network

()

experiments/CIFAR10/pre-res18.yopo-5-3/network.py:131

↓ 2 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/dataset.py:17

↓ 2 callersFunctioncreate_test_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/dataset.py:17

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/dataset.py:5

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/dataset.py:5

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/wide34.natural/dataset.py:5

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/pre-res18.pgd10/dataset.py:5

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/pre-res18.yopo-5-3/dataset.py:5

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/wide34.pgd10/dataset.py:5

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/CIFAR10/wide34.yopo-5-3/dataset.py:5

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/MNIST/YOPO-5-10/dataset.py:5

↓ 2 callersFunctioncreate_train_dataset

(batch_size = 128, root = '../data')

experiments/MNIST/pgd40/dataset.py:5

↓ 2 callersFunctiontrain_one_epoch

:param attack_freq: Frequencies of training with adversarial examples. -1 indicates natural training :param AttackMethod: the attack method

experiments/CIFAR10/pre-res18.yopo-5-3/training_function.py:63

↓ 2 callersFunctiontrain_one_epoch

:param attack_freq: Frequencies of training with adversarial examples. -1 indicates natural training :param AttackMethod: the attack method

experiments/MNIST/YOPO-5-10/training_function.py:63

↓ 1 callersMethod__init__

(self, layer, reg_cof = 1e-4)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/loss.py:9

↓ 1 callersMethod__init__

(self, layer, reg_cof = 1e-4)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/loss.py:9

↓ 1 callersMethod__init__

(self, layer, reg_cof = 1e-4)

experiments/CIFAR10/pre-res18.yopo-5-3/loss.py:9

↓ 1 callersMethod__init__

(self, layer, reg_cof = 1e-4)

experiments/CIFAR10/wide34.yopo-5-3/loss.py:9

↓ 1 callersMethod__init__

(self, layer, reg_cof = 1e-4)

experiments/MNIST/YOPO-5-10/loss.py:9

↓ 1 callersMethod_make_layer

(self, block, in_planes, out_planes, nb_layers, stride, dropRate)

lib/base_model/wide_resnet.py:50

↓ 1 callersFunctionadd_path

(path)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/config.py:8

↓ 1 callersFunctionadd_path

(path)

experiments/CIFAR10-TRADES/baseline.res-pre18.TRADES.10step/config.py:8

↓ 1 callersFunctionadd_path

(path)

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/config.py:9

↓ 1 callersFunctionadd_path

(path)

experiments/CIFAR10/wide34.natural/config.py:8

↓ 1 callersFunctionadd_path

(path)

experiments/CIFAR10/pre-res18.pgd10/config.py:8

↓ 1 callersFunctionadd_path

(path)

experiments/CIFAR10/pre-res18.yopo-5-3/config.py:9

↓ 1 callersFunctionadd_path

(path)

experiments/CIFAR10/wide34.pgd10/config.py:8

↓ 1 callersFunctionadd_path

(path)

experiments/CIFAR10/wide34.yopo-5-3/config.py:9

↓ 1 callersFunctionadd_path

(path)

experiments/MNIST/YOPO-5-10/config.py:8

↓ 1 callersFunctionadd_path

(path)

experiments/MNIST/pgd40/config.py:7

↓ 1 callersFunctionadjust_learning_rate

decrease the learning rate

experiments/CIFAR10-TRADES/baseline.res-pre18.TRADES.10step/train_trades_cifar10.py:83

↓ 1 callersFunctioncal_l2_norm

(layer: torch.nn.Module)

experiments/CIFAR10/pre-res18.yopo-5-3/loss.py:49

↓ 1 callersFunctioncal_l2_norm

(layer: torch.nn.Module)

experiments/CIFAR10/wide34.yopo-5-3/loss.py:49

↓ 1 callersFunctioncal_l2_norm

(layer: torch.nn.Module)

experiments/MNIST/YOPO-5-10/loss.py:38

↓ 1 callersFunctioncreate_network

()

lib/base_model/wide_resnet.py:121

↓ 1 callersFunctioncreate_network

()

lib/base_model/small_cnn.py:71

↓ 1 callersFunctioncreate_network

()

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-2-5/network.py:131

↓ 1 callersFunctioncreate_network

()

experiments/CIFAR10-TRADES/pre-res18.TRADES-YOPO-3-4/network.py:131

↓ 1 callersFunctionmain

()

experiments/CIFAR10-TRADES/baseline.res-pre18.TRADES.10step/train_trades_cifar10.py:96

↓ 1 callersFunctionmake_symlink

Note: overwriting enabled!

lib/utils/misc.py:104

↓ 1 callersMethodreset

(self)

lib/utils/misc.py:39

↓ 1 callersMethodsingle_attack

Given the original image and the perturbation computed so far, computes a new perturbation. :param net: :param inp: o

lib/attack/pgd.py:43

next →1–100 of 230, ranked by callers