Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_True_dataset_CIFAR10_epochs_100_lr_0.001_model_type_vgg11_weight_decay_0.001.pth.tar
Training Set:
Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.021450402215123177
Inter Cos: 0.11371058225631714
Norm Quadratic Average: 27.691621780395508
Nearest Class Center Accuracy: 0.279125

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.022717682644724846
Inter Cos: 0.08386020362377167
Norm Quadratic Average: 83.46713256835938
Nearest Class Center Accuracy: 0.33325

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02424004301428795
Inter Cos: 0.08712425827980042
Norm Quadratic Average: 62.4416618347168
Nearest Class Center Accuracy: 0.373125

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.022015128284692764
Inter Cos: 0.07528096437454224
Norm Quadratic Average: 65.78900909423828
Nearest Class Center Accuracy: 0.3975

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.028819842264056206
Inter Cos: 0.08070674538612366
Norm Quadratic Average: 42.38957595825195
Nearest Class Center Accuracy: 0.424875

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.028915785253047943
Inter Cos: 0.06707606464624405
Norm Quadratic Average: 43.21894836425781
Nearest Class Center Accuracy: 0.4665

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.03963622823357582
Inter Cos: 0.07406767457723618
Norm Quadratic Average: 27.764789581298828
Nearest Class Center Accuracy: 0.54425

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.059534016996622086
Inter Cos: 0.07212966680526733
Norm Quadratic Average: 19.764907836914062
Nearest Class Center Accuracy: 0.8375

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 96.63853454589844
Linear Weight Rank: 4031
Intra Cos: 0.1738060563802719
Inter Cos: 0.10221543163061142
Norm Quadratic Average: 104.8436050415039
Nearest Class Center Accuracy: 1.0

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 39.06978225708008
Linear Weight Rank: 3670
Intra Cos: 0.40590471029281616
Inter Cos: 0.19945724308490753
Norm Quadratic Average: 53.879981994628906
Nearest Class Center Accuracy: 1.0

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.458373546600342
Linear Weight Rank: 10
Intra Cos: 0.643965482711792
Inter Cos: 0.3110678493976593
Norm Quadratic Average: 37.24169921875
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.8713770508766174
Inter Cos: 0.5369093418121338
Norm Quadratic Average: 25.68733787536621
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 2.3698979721069335
Accuracy: 0.5815
NC1 Within Class Collapse: 37.711936950683594
NC2 Equinorm: Features: 0.20395906269550323, Weights: 0.017713231965899467
NC2 Equiangle: Features: 0.4437687767876519, Weights: 0.0921054310268826
NC3 Self-Duality: 0.6297578811645508
NC4 NCC Mismatch: 0.14300000000000002

Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.018480461090803146
Inter Cos: 0.09352367371320724
Norm Quadratic Average: 27.530664443969727
Nearest Class Center Accuracy: 0.2895

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.022190598770976067
Inter Cos: 0.07497008889913559
Norm Quadratic Average: 83.22297668457031
Nearest Class Center Accuracy: 0.355

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.024790091440081596
Inter Cos: 0.076621875166893
Norm Quadratic Average: 62.170379638671875
Nearest Class Center Accuracy: 0.396

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02173113450407982
Inter Cos: 0.06697235256433487
Norm Quadratic Average: 65.58692169189453
Nearest Class Center Accuracy: 0.4285

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.025561312213540077
Inter Cos: 0.07193225622177124
Norm Quadratic Average: 42.25083541870117
Nearest Class Center Accuracy: 0.4535

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.024430768564343452
Inter Cos: 0.059058740735054016
Norm Quadratic Average: 43.11703872680664
Nearest Class Center Accuracy: 0.489

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02787676639854908
Inter Cos: 0.06841367483139038
Norm Quadratic Average: 27.658546447753906
Nearest Class Center Accuracy: 0.488

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.03527715057134628
Inter Cos: 0.0675763264298439
Norm Quadratic Average: 19.58998680114746
Nearest Class Center Accuracy: 0.571

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 96.63853454589844
Linear Weight Rank: 4031
Intra Cos: 0.06474731117486954
Inter Cos: 0.10509145259857178
Norm Quadratic Average: 101.10198211669922
Nearest Class Center Accuracy: 0.607

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 39.06978225708008
Linear Weight Rank: 3670
Intra Cos: 0.13470737636089325
Inter Cos: 0.2114780694246292
Norm Quadratic Average: 49.68943786621094
Nearest Class Center Accuracy: 0.591

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.458373546600342
Linear Weight Rank: 10
Intra Cos: 0.2137308120727539
Inter Cos: 0.33366045355796814
Norm Quadratic Average: 33.06206512451172
Nearest Class Center Accuracy: 0.5775

Output Layer:
Intra Cos: 0.31384262442588806
Inter Cos: 0.5239772796630859
Norm Quadratic Average: 22.202186584472656
Nearest Class Center Accuracy: 0.5615

