upload inits

.gitignore

@@ -160,8 +160,8 @@ cython_debug/
 #.idea/
 
 
-__*
-__*/*
+___*
+___*/*
 /data/*
 /logo/*
 /demos/_*

mair/__init__.py

@@ -0,0 +1,6 @@
+from .nn.robmodel import RobModel
+from .utils import load_model
+
+# from .utils.datasets import Datasets
+
+__version__ = "1.0.0"

mair/attacks/__init__.py

@@ -0,0 +1,99 @@
+# None attacks
+from .attacks.vanila import VANILA
+from .attacks.gn import GN
+
+# Linf attacks
+from .attacks.fgsm import FGSM
+from .attacks.bim import BIM
+from .attacks.rfgsm import RFGSM
+from .attacks.pgd import PGD
+from .attacks.eotpgd import EOTPGD
+from .attacks.ffgsm import FFGSM
+from .attacks.tpgd import TPGD
+from .attacks.mifgsm import MIFGSM
+from .attacks.upgd import UPGD
+from .attacks.apgd import APGD
+from .attacks.apgdt import APGDT
+from .attacks.difgsm import DIFGSM
+from .attacks.tifgsm import TIFGSM
+from .attacks.jitter import Jitter
+from .attacks.nifgsm import NIFGSM
+from .attacks.pgdrs import PGDRS
+from .attacks.sinifgsm import SINIFGSM
+from .attacks.vmifgsm import VMIFGSM
+from .attacks.vnifgsm import VNIFGSM
+from .attacks.spsa import SPSA
+from .attacks.pifgsm import PIFGSM
+from .attacks.pifgsmpp import PIFGSMPP
+
+# L2 attacks
+from .attacks.cw import CW
+from .attacks.pgdl2 import PGDL2
+from .attacks.pgdrsl2 import PGDRSL2
+from .attacks.deepfool import DeepFool
+from .attacks.eaden import EADEN
+
+# L1 attacks
+from .attacks.eadl1 import EADL1
+
+# L0 attacks
+from .attacks.sparsefool import SparseFool
+from .attacks.onepixel import OnePixel
+from .attacks.pixle import Pixle
+from .attacks.jsma import JSMA
+
+# Linf, L2 attacks
+from .attacks.fab import FAB
+from .attacks.autoattack import AutoAttack
+from .attacks.square import Square
+
+# Wrapper Class
+from .wrappers.multiattack import MultiAttack
+from .wrappers.lgv import LGV
+
+__version__ = "3.5.1"
+__all__ = [
+    "VANILA",
+    "GN",
+    "FGSM",
+    "BIM",
+    "RFGSM",
+    "PGD",
+    "EOTPGD",
+    "FFGSM",
+    "TPGD",
+    "MIFGSM",
+    "UPGD",
+    "APGD",
+    "APGDT",
+    "DIFGSM",
+    "TIFGSM",
+    "Jitter",
+    "NIFGSM",
+    "PGDRS",
+    "SINIFGSM",
+    "VMIFGSM",
+    "VNIFGSM",
+    "SPSA",
+    "JSMA",
+    "EADL1",
+    "EADEN",
+    "PIFGSM",
+    "PIFGSMPP",
+    "CW",
+    "PGDL2",
+    "DeepFool",
+    "PGDRSL2",
+    "SparseFool",
+    "OnePixel",
+    "Pixle",
+    "FAB",
+    "AutoAttack",
+    "Square",
+    "MultiAttack",
+    "LGV",
+]
+__wrapper__ = [
+    "LGV",
+    "MultiAttack",
+]

mair/defenses/__init__.py

@@ -0,0 +1,4 @@
+from .advtraining.standard import Standard
+from .advtraining.at import AT
+from .advtraining.trades import TRADES
+from .advtraining.mart import MART

mair/optim/__init__.py

@@ -0,0 +1 @@
+from .minimizer import *

mair/utils/__init__.py

@@ -0,0 +1,5 @@
+# from .datasets import Datasets
+from .models import load_model
+from .cuda import manual_seed
+from .data import get_subloader
+from .eval import get_accuracy

mair/utils/models/__init__.py

@@ -0,0 +1,90 @@
+from .lenet import LeNet, LeNetPP
+from .mnist_ates import MNIST_ATES
+from .mnist_dat import MNIST_DAT
+from .mnist_fast import MNIST_Fast
+from .preactresnet import PreActBlock, PreActResNet
+from .resnet import ResBasicBlock, ResNet
+from .densenet import DenseNet, Bottleneck
+from .vgg import VGG
+from .wideresnet import WideResNet
+
+
+def load_model(model_name, n_classes):
+    if model_name == "LeNet":
+        return LeNet(n_classes)
+
+    if model_name == "LeNetPP":
+        return LeNetPP(n_classes)
+
+    elif model_name == "MNIST_ATES":
+        return MNIST_ATES(n_classes)
+
+    elif model_name == "MNIST_DAT":
+        return MNIST_DAT(n_classes)
+
+    elif model_name == "MNIST_Fast":
+        return MNIST_Fast(n_classes)
+
+    elif model_name == "WRN28-10":
+        model = WideResNet(
+            depth=28, num_classes=n_classes, widen_factor=10, dropRate=0.0
+        )
+
+    elif model_name == "WRN34-10":
+        model = WideResNet(
+            depth=34, num_classes=n_classes, widen_factor=10, dropRate=0.0
+        )
+
+    elif model_name == "PRN18":
+        model = PreActResNet(
+            PreActBlock, num_blocks=[2, 2, 2, 2], num_classes=n_classes
+        )
+
+    elif model_name == "ResNet10":
+        model = ResNet(ResBasicBlock, [1, 1, 1, 1], n_classes, in_channels=1)
+
+    elif model_name == "ResNet18":
+        model = ResNet(ResBasicBlock, [2, 2, 2, 2], n_classes)
+
+    elif model_name == "ResNet34":
+        model = ResNet(ResBasicBlock, [3, 4, 6, 3], n_classes)
+
+    elif model_name == "ResNet50":
+        model = ResNet(ResBasicBlock, [3, 4, 6, 3], n_classes)
+
+    elif model_name == "ResNet101":
+        model = ResNet(ResBasicBlock, [3, 4, 23, 3], n_classes)
+
+    elif model_name == "ResNet152":
+        model = ResNet(ResBasicBlock, [3, 8, 36, 3], n_classes)
+
+    elif model_name == "DenseNet121":
+        model = DenseNet(Bottleneck, [6, 12, 24, 16], growth_rate=32)
+
+    elif model_name == "DenseNet169":
+        model = DenseNet(Bottleneck, [6, 12, 32, 32], growth_rate=32)
+
+    elif model_name == "DenseNet201":
+        model = DenseNet(Bottleneck, [6, 12, 48, 32], growth_rate=32)
+
+    elif model_name == "DenseNet161":
+        model = DenseNet(Bottleneck, [6, 12, 36, 24], growth_rate=48)
+
+    elif model_name == "VGG11":
+        model = VGG("VGG11", n_classes)
+
+    elif model_name == "VGG13":
+        model = VGG("VGG13", n_classes)
+
+    elif model_name == "VGG16":
+        model = VGG("VGG16", n_classes)
+
+    elif model_name == "VGG19":
+        model = VGG("VGG19", n_classes)
+
+    else:
+        raise ValueError("Invalid model name.")
+
+    print(model_name, "is loaded.")
+
+    return model