Executing method neural_path_kmeans on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.72
Ratio, Val accuracy: 0.900 42.43
Ratio, Test accuracy: 0.900 43.80
Ratio, Val accuracy: 0.700 42.90
Ratio, Test accuracy: 0.700 44.39
Ratio, Val accuracy: 0.500 42.38
Ratio, Test accuracy: 0.500 43.44
Ratio, Val accuracy: 0.300 42.99
Ratio, Test accuracy: 0.300 43.89
Ratio, Val accuracy: 0.200 40.33
Ratio, Test accuracy: 0.200 41.53
Ratio, Val accuracy: 0.150 32.42
Ratio, Test accuracy: 0.150 33.83
Ratio, Val accuracy: 0.100 30.19
Ratio, Test accuracy: 0.100 30.63
Ratio, Val accuracy: 0.050 8.30
Ratio, Test accuracy: 0.050 8.38
Ratio, Val accuracy: 0.020 1.23
Ratio, Test accuracy: 0.020 1.25
Ratio, Val accuracy: 0.010 0.94
Ratio, Test accuracy: 0.010 1.01
Ratio, Val accuracy: 0.005 0.94
Ratio, Test accuracy: 0.005 1.00
Executing method random_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.72
Ratio, Val accuracy: 0.900 42.18
Ratio, Test accuracy: 0.900 43.84
Ratio, Val accuracy: 0.700 42.40
Ratio, Test accuracy: 0.700 43.86
Ratio, Val accuracy: 0.500 42.47
Ratio, Test accuracy: 0.500 43.48
Ratio, Val accuracy: 0.300 38.77
Ratio, Test accuracy: 0.300 40.26
Ratio, Val accuracy: 0.200 38.59
Ratio, Test accuracy: 0.200 39.02
Ratio, Val accuracy: 0.150 33.54
Ratio, Test accuracy: 0.150 34.56
Ratio, Val accuracy: 0.100 26.55
Ratio, Test accuracy: 0.100 27.75
Ratio, Val accuracy: 0.050 17.30
Ratio, Test accuracy: 0.050 17.95
Ratio, Val accuracy: 0.020 6.13
Ratio, Test accuracy: 0.020 6.35
Ratio, Val accuracy: 0.010 3.74
Ratio, Test accuracy: 0.010 3.91
Ratio, Val accuracy: 0.005 0.94
Ratio, Test accuracy: 0.005 1.00
Executing method l1_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.72
Ratio, Val accuracy: 0.900 42.05
Ratio, Test accuracy: 0.900 43.55
Ratio, Val accuracy: 0.700 43.29
Ratio, Test accuracy: 0.700 44.13
Ratio, Val accuracy: 0.500 40.92
Ratio, Test accuracy: 0.500 42.01
Ratio, Val accuracy: 0.300 42.03
Ratio, Test accuracy: 0.300 43.21
Ratio, Val accuracy: 0.200 37.69
Ratio, Test accuracy: 0.200 39.36
Ratio, Val accuracy: 0.150 32.43
Ratio, Test accuracy: 0.150 33.66
Ratio, Val accuracy: 0.100 23.70
Ratio, Test accuracy: 0.100 25.22
Ratio, Val accuracy: 0.050 13.62
Ratio, Test accuracy: 0.050 13.78
Ratio, Val accuracy: 0.020 7.46
Ratio, Test accuracy: 0.020 7.01
Executing method neural_path_kmeans on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.11
Ratio, Val accuracy: 0.900 41.52
Ratio, Test accuracy: 0.900 42.66
Ratio, Val accuracy: 0.700 41.27
Ratio, Test accuracy: 0.700 42.28
Ratio, Val accuracy: 0.500 40.69
Ratio, Test accuracy: 0.500 41.17
Ratio, Val accuracy: 0.300 40.18
Ratio, Test accuracy: 0.300 40.64
Ratio, Val accuracy: 0.200 36.34
Ratio, Test accuracy: 0.200 36.56
Ratio, Val accuracy: 0.150 36.72
Ratio, Test accuracy: 0.150 37.08
Ratio, Val accuracy: 0.100 30.85
Ratio, Test accuracy: 0.100 31.02
Ratio, Val accuracy: 0.050 13.16
Ratio, Test accuracy: 0.050 12.87
Ratio, Val accuracy: 0.020 2.15
Ratio, Test accuracy: 0.020 2.18
Ratio, Val accuracy: 0.010 1.02
Ratio, Test accuracy: 0.010 1.01
Ratio, Val accuracy: 0.005 1.02
Ratio, Test accuracy: 0.005 1.00
Executing method random_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.11
Ratio, Val accuracy: 0.900 40.34
Ratio, Test accuracy: 0.900 41.66
Ratio, Val accuracy: 0.700 42.10
Ratio, Test accuracy: 0.700 43.04
Ratio, Val accuracy: 0.500 41.23
Ratio, Test accuracy: 0.500 41.97
Ratio, Val accuracy: 0.300 37.18
Ratio, Test accuracy: 0.300 38.07
Ratio, Val accuracy: 0.200 35.30
Ratio, Test accuracy: 0.200 36.34
Ratio, Val accuracy: 0.150 30.78
Ratio, Test accuracy: 0.150 31.42
Ratio, Val accuracy: 0.100 31.99
Ratio, Test accuracy: 0.100 32.40
Ratio, Val accuracy: 0.050 17.05
Ratio, Test accuracy: 0.050 17.13
Ratio, Val accuracy: 0.020 6.39
Ratio, Test accuracy: 0.020 6.55
Ratio, Val accuracy: 0.010 2.56
Ratio, Test accuracy: 0.010 2.58
Ratio, Val accuracy: 0.005 1.52
Ratio, Test accuracy: 0.005 1.48
Executing method l1_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.11
Ratio, Val accuracy: 0.900 40.83
Ratio, Test accuracy: 0.900 41.78
Ratio, Val accuracy: 0.700 39.85
Ratio, Test accuracy: 0.700 41.06
Ratio, Val accuracy: 0.500 39.30
Ratio, Test accuracy: 0.500 40.64
Ratio, Val accuracy: 0.300 36.30
Ratio, Test accuracy: 0.300 37.20
Ratio, Val accuracy: 0.200 35.58
Ratio, Test accuracy: 0.200 36.51
Ratio, Val accuracy: 0.150 38.44
Ratio, Test accuracy: 0.150 38.98
Ratio, Val accuracy: 0.100 33.21
Ratio, Test accuracy: 0.100 33.80
Ratio, Val accuracy: 0.050 15.06
Ratio, Test accuracy: 0.050 15.80
Ratio, Val accuracy: 0.020 4.47
Ratio, Test accuracy: 0.020 4.57
Ratio, Val accuracy: 0.010 2.70
Ratio, Test accuracy: 0.010 2.71
Ratio, Val accuracy: 0.005 1.61
Ratio, Test accuracy: 0.005 1.85
VGGBnDrop(
  (features): Sequential(
    (0): ConvBNReLU(
      (conv): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (1): Dropout(p=0.3, inplace=False)
    (2): ConvBNReLU(
      (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (3): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
    (4): ConvBNReLU(
      (conv): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (5): Dropout(p=0.4, inplace=False)
    (6): ConvBNReLU(
      (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (7): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
    (8): ConvBNReLU(
      (conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (9): Dropout(p=0.4, inplace=False)
    (10): ConvBNReLU(
      (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (11): Dropout(p=0.4, inplace=False)
    (12): ConvBNReLU(
      (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (13): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
    (14): ConvBNReLU(
      (conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (15): Dropout(p=0.4, inplace=False)
    (16): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (17): Dropout(p=0.4, inplace=False)
    (18): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (19): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
    (20): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (21): Dropout(p=0.4, inplace=False)
    (22): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (23): Dropout(p=0.4, inplace=False)
    (24): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (25): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
  )
  (classifier): Sequential(
    (0): Dropout(p=0.5, inplace=False)
    (1): Linear(in_features=512, out_features=512, bias=True)
    (2): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (3): ReLU(inplace=True)
    (4): Dropout(p=0.5, inplace=False)
    (5): Linear(in_features=512, out_features=100, bias=True)
  )
)
Executing method neural_path_kmeans on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.61
Ratio, Val accuracy: 0.900 41.66
Ratio, Test accuracy: 0.900 42.57
Ratio, Val accuracy: 0.700 41.38
Ratio, Test accuracy: 0.700 42.67
Ratio, Val accuracy: 0.500 42.14
Ratio, Test accuracy: 0.500 43.37
Ratio, Val accuracy: 0.300 38.83
Ratio, Test accuracy: 0.300 39.77
Ratio, Val accuracy: 0.200 39.92
Ratio, Test accuracy: 0.200 41.24
Ratio, Val accuracy: 0.150 38.88
Ratio, Test accuracy: 0.150 39.82
Ratio, Val accuracy: 0.100 29.98
Ratio, Test accuracy: 0.100 30.98
Ratio, Val accuracy: 0.050 13.66
Ratio, Test accuracy: 0.050 13.11
Ratio, Val accuracy: 0.020 3.46
Ratio, Test accuracy: 0.020 3.34
Ratio, Val accuracy: 0.010 2.24
Ratio, Test accuracy: 0.010 2.14
Ratio, Val accuracy: 0.005 1.02
Ratio, Test accuracy: 0.005 1.00
Executing method random_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.61
Ratio, Val accuracy: 0.900 42.10
Ratio, Test accuracy: 0.900 43.21
Ratio, Val accuracy: 0.700 40.70
Ratio, Test accuracy: 0.700 42.26
Ratio, Val accuracy: 0.500 41.66
Ratio, Test accuracy: 0.500 42.69
Ratio, Val accuracy: 0.300 35.97
Ratio, Test accuracy: 0.300 37.70
Ratio, Val accuracy: 0.200 41.12
Ratio, Test accuracy: 0.200 42.21
Ratio, Val accuracy: 0.150 33.16
Ratio, Test accuracy: 0.150 33.46
Ratio, Val accuracy: 0.100 32.36
Ratio, Test accuracy: 0.100 33.45
Ratio, Val accuracy: 0.050 21.30
Ratio, Test accuracy: 0.050 21.74
Ratio, Val accuracy: 0.020 8.85
Ratio, Test accuracy: 0.020 8.89
Ratio, Val accuracy: 0.010 4.10
Ratio, Test accuracy: 0.010 3.76
Ratio, Val accuracy: 0.005 1.54
Ratio, Test accuracy: 0.005 1.54
Executing method l1_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 49.61
Ratio, Val accuracy: 0.900 41.46
Ratio, Test accuracy: 0.900 42.48
Ratio, Val accuracy: 0.700 41.89
Ratio, Test accuracy: 0.700 43.32
Ratio, Val accuracy: 0.500 41.22
Ratio, Test accuracy: 0.500 42.62
Ratio, Val accuracy: 0.300 39.45
Ratio, Test accuracy: 0.300 41.09
Ratio, Val accuracy: 0.200 34.18
Ratio, Test accuracy: 0.200 35.87
Ratio, Val accuracy: 0.150 32.92
Ratio, Test accuracy: 0.150 33.80
Ratio, Val accuracy: 0.100 29.83
Ratio, Test accuracy: 0.100 30.72
Ratio, Val accuracy: 0.050 21.38
Ratio, Test accuracy: 0.050 22.20
Ratio, Val accuracy: 0.020 4.22
Ratio, Test accuracy: 0.020 4.06
Ratio, Val accuracy: 0.010 3.45
Ratio, Test accuracy: 0.010 3.40
Ratio, Val accuracy: 0.005 2.13
Ratio, Test accuracy: 0.005 2.08
Ratio, Val accuracy: 0.010 2.17
Ratio, Test accuracy: 0.010 2.15
Ratio, Val accuracy: 0.005 1.35
Ratio, Test accuracy: 0.005 1.26
Executing method neural_path_kmeans on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 48.96
Ratio, Val accuracy: 0.900 43.09
Ratio, Test accuracy: 0.900 43.87
Ratio, Val accuracy: 0.700 43.34
Ratio, Test accuracy: 0.700 44.05
Ratio, Val accuracy: 0.500 42.98
Ratio, Test accuracy: 0.500 43.94
Ratio, Val accuracy: 0.300 40.33
Ratio, Test accuracy: 0.300 41.30
Ratio, Val accuracy: 0.200 38.04
Ratio, Test accuracy: 0.200 39.05
Ratio, Val accuracy: 0.150 35.12
Ratio, Test accuracy: 0.150 35.09
Ratio, Val accuracy: 0.100 26.54
Ratio, Test accuracy: 0.100 27.23
Ratio, Val accuracy: 0.050 7.63
Ratio, Test accuracy: 0.050 7.56
Ratio, Val accuracy: 0.020 2.78
Ratio, Test accuracy: 0.020 2.78
Ratio, Val accuracy: 0.010 1.12
Ratio, Test accuracy: 0.010 1.01
Ratio, Val accuracy: 0.005 1.10
Ratio, Test accuracy: 0.005 1.00
Executing method random_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 48.96
Ratio, Val accuracy: 0.900 42.26
Ratio, Test accuracy: 0.900 42.92
Ratio, Val accuracy: 0.700 40.90
Ratio, Test accuracy: 0.700 41.50
Ratio, Val accuracy: 0.500 42.13
Ratio, Test accuracy: 0.500 42.87
Ratio, Val accuracy: 0.300 42.08
Ratio, Test accuracy: 0.300 42.88
Ratio, Val accuracy: 0.200 35.42
Ratio, Test accuracy: 0.200 36.26
Ratio, Val accuracy: 0.150 25.10
Ratio, Test accuracy: 0.150 25.61
Ratio, Val accuracy: 0.100 25.78
Ratio, Test accuracy: 0.100 25.87
Ratio, Val accuracy: 0.050 21.44
Ratio, Test accuracy: 0.050 21.86
Ratio, Val accuracy: 0.020 6.31
Ratio, Test accuracy: 0.020 6.14
Ratio, Val accuracy: 0.010 1.93
Ratio, Test accuracy: 0.010 1.80
Ratio, Val accuracy: 0.005 1.93
Ratio, Test accuracy: 0.005 1.91
Executing method l1_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 48.96
Ratio, Val accuracy: 0.900 42.71
Ratio, Test accuracy: 0.900 43.52
Ratio, Val accuracy: 0.700 42.77
Ratio, Test accuracy: 0.700 43.23
Ratio, Val accuracy: 0.500 42.43
Ratio, Test accuracy: 0.500 42.73
Ratio, Val accuracy: 0.300 38.33
Ratio, Test accuracy: 0.300 39.06
Ratio, Val accuracy: 0.200 38.62
Ratio, Test accuracy: 0.200 39.55
Ratio, Val accuracy: 0.150 36.40
Ratio, Test accuracy: 0.150 37.59
Ratio, Val accuracy: 0.100 28.55
Ratio, Test accuracy: 0.100 29.18
Ratio, Val accuracy: 0.050 15.35
Ratio, Test accuracy: 0.050 15.39
Ratio, Val accuracy: 0.020 8.74
Ratio, Test accuracy: 0.020 8.70
Ratio, Val accuracy: 0.010 3.33
Ratio, Test accuracy: 0.010 3.12
Ratio, Val accuracy: 0.005 1.51
Ratio, Test accuracy: 0.005 1.57
VGGBnDrop(
  (features): Sequential(
    (0): ConvBNReLU(
      (conv): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (1): Dropout(p=0.3, inplace=False)
    (2): ConvBNReLU(
      (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (3): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
    (4): ConvBNReLU(
      (conv): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (5): Dropout(p=0.4, inplace=False)
    (6): ConvBNReLU(
      (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (7): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
    (8): ConvBNReLU(
      (conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (9): Dropout(p=0.4, inplace=False)
    (10): ConvBNReLU(
      (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (11): Dropout(p=0.4, inplace=False)
    (12): ConvBNReLU(
      (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (13): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
    (14): ConvBNReLU(
      (conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (15): Dropout(p=0.4, inplace=False)
    (16): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (17): Dropout(p=0.4, inplace=False)
    (18): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (19): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
    (20): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (21): Dropout(p=0.4, inplace=False)
    (22): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (23): Dropout(p=0.4, inplace=False)
    (24): ConvBNReLU(
      (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (bn): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (25): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)
  )
  (classifier): Sequential(
    (0): Dropout(p=0.5, inplace=False)
    (1): Linear(in_features=512, out_features=512, bias=True)
    (2): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (3): ReLU(inplace=True)
    (4): Dropout(p=0.5, inplace=False)
    (5): Linear(in_features=512, out_features=100, bias=True)
  )
)
Executing method neural_path_kmeans on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 48.98
Ratio, Val accuracy: 0.900 41.69
Ratio, Test accuracy: 0.900 42.76
Ratio, Val accuracy: 0.700 41.91
Ratio, Test accuracy: 0.700 42.91
Ratio, Val accuracy: 0.500 40.73
Ratio, Test accuracy: 0.500 41.80
Ratio, Val accuracy: 0.300 39.83
Ratio, Test accuracy: 0.300 40.96
Ratio, Val accuracy: 0.200 37.79
Ratio, Test accuracy: 0.200 38.48
Ratio, Val accuracy: 0.150 34.67
Ratio, Test accuracy: 0.150 35.06
Ratio, Val accuracy: 0.100 27.56
Ratio, Test accuracy: 0.100 28.20
Ratio, Val accuracy: 0.050 10.73
Ratio, Test accuracy: 0.050 10.81
Ratio, Val accuracy: 0.020 1.58
Ratio, Test accuracy: 0.020 1.66
Ratio, Val accuracy: 0.010 0.90
Ratio, Test accuracy: 0.010 1.00
Ratio, Val accuracy: 0.005 0.90
Ratio, Test accuracy: 0.005 1.00
Executing method random_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 48.98
Ratio, Val accuracy: 0.900 42.29
Ratio, Test accuracy: 0.900 43.10
Ratio, Val accuracy: 0.700 39.72
Ratio, Test accuracy: 0.700 40.69
Ratio, Val accuracy: 0.500 39.93
Ratio, Test accuracy: 0.500 40.70
Ratio, Val accuracy: 0.300 36.10
Ratio, Test accuracy: 0.300 37.41
Ratio, Val accuracy: 0.200 38.05
Ratio, Test accuracy: 0.200 38.80
Ratio, Val accuracy: 0.150 36.88
Ratio, Test accuracy: 0.150 37.65
Ratio, Val accuracy: 0.100 25.94
Ratio, Test accuracy: 0.100 26.98
Ratio, Val accuracy: 0.050 17.23
Ratio, Test accuracy: 0.050 17.88
Ratio, Val accuracy: 0.020 5.58
Ratio, Test accuracy: 0.020 5.93
Ratio, Val accuracy: 0.010 4.38
Ratio, Test accuracy: 0.010 4.14
Ratio, Val accuracy: 0.005 3.34
Ratio, Test accuracy: 0.005 3.65
Executing method l1_structured on model CIFAR-VGG and dataset CIFAR100
Ratio, Test accuracy: 1.000 48.98
Ratio, Val accuracy: 0.900 41.38
Ratio, Test accuracy: 0.900 42.01
Ratio, Val accuracy: 0.700 40.54
Ratio, Test accuracy: 0.700 41.21
Ratio, Val accuracy: 0.500 42.70
Ratio, Test accuracy: 0.500 43.18
Ratio, Val accuracy: 0.300 40.49
Ratio, Test accuracy: 0.300 40.74
Ratio, Val accuracy: 0.200 38.35
Ratio, Test accuracy: 0.200 38.91
Ratio, Val accuracy: 0.150 23.38
Ratio, Test accuracy: 0.150 23.91
Ratio, Val accuracy: 0.100 29.27
Ratio, Test accuracy: 0.100 29.79
Ratio, Val accuracy: 0.050 13.39
Ratio, Test accuracy: 0.050 13.97
Ratio, Val accuracy: 0.020 4.91
Ratio, Test accuracy: 0.020 4.96
Ratio, Val accuracy: 0.010 2.93
Ratio, Test accuracy: 0.010 3.37
Ratio, Val accuracy: 0.005 1.39
Ratio, Test accuracy: 0.005 1.56
