Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_False_dataset_MNIST_epochs_50_lr_0.001_model_type_vgg11_weight_decay_0.003.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.08946064114570618
Inter Cos: 0.11311884224414825
Norm Quadratic Average: 23.532936096191406
Nearest Class Center Accuracy: 0.81175

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10988636314868927
Inter Cos: 0.1341933012008667
Norm Quadratic Average: 47.25871276855469
Nearest Class Center Accuracy: 0.824375

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15459436178207397
Inter Cos: 0.17775017023086548
Norm Quadratic Average: 45.040061950683594
Nearest Class Center Accuracy: 0.807125

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17026260495185852
Inter Cos: 0.19223006069660187
Norm Quadratic Average: 56.62126541137695
Nearest Class Center Accuracy: 0.815125

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19092188775539398
Inter Cos: 0.18197491765022278
Norm Quadratic Average: 34.41887283325195
Nearest Class Center Accuracy: 0.854125

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2174687683582306
Inter Cos: 0.18912756443023682
Norm Quadratic Average: 31.733421325683594
Nearest Class Center Accuracy: 0.8935

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2861764430999756
Inter Cos: 0.17553068697452545
Norm Quadratic Average: 18.095869064331055
Nearest Class Center Accuracy: 0.936375

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4156356453895569
Inter Cos: 0.19258669018745422
Norm Quadratic Average: 13.693105697631836
Nearest Class Center Accuracy: 0.976125

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 92.40340423583984
Linear Weight Rank: 4031
Intra Cos: 0.639240562915802
Inter Cos: 0.24135807156562805
Norm Quadratic Average: 61.04727554321289
Nearest Class Center Accuracy: 0.99725

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 37.35926055908203
Linear Weight Rank: 3670
Intra Cos: 0.7356324195861816
Inter Cos: 0.24681925773620605
Norm Quadratic Average: 39.09096908569336
Nearest Class Center Accuracy: 0.99925

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.366819381713867
Linear Weight Rank: 10
Intra Cos: 0.7808847427368164
Inter Cos: 0.26580187678337097
Norm Quadratic Average: 29.908721923828125
Nearest Class Center Accuracy: 0.999375

Output Layer:
Intra Cos: 0.8193708658218384
Inter Cos: 0.36407721042633057
Norm Quadratic Average: 21.017290115356445
Nearest Class Center Accuracy: 0.9985

Test Set:
Average Loss: 0.072949194252491
Accuracy: 0.978
NC1 Within Class Collapse: 1.8423991203308105
NC2 Equinorm: Features: 0.11271364986896515, Weights: 0.011452188715338707
NC2 Equiangle: Features: 0.2493549558851454, Weights: 0.09530474344889323
NC3 Self-Duality: 0.5265071392059326
NC4 NCC Mismatch: 0.01100000000000001

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10133811086416245
Inter Cos: 0.11957789957523346
Norm Quadratic Average: 22.82430076599121
Nearest Class Center Accuracy: 0.809

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13285794854164124
Inter Cos: 0.144385427236557
Norm Quadratic Average: 45.890594482421875
Nearest Class Center Accuracy: 0.8205

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17261914908885956
Inter Cos: 0.19916518032550812
Norm Quadratic Average: 43.77405548095703
Nearest Class Center Accuracy: 0.8065

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1820266842842102
Inter Cos: 0.2175491452217102
Norm Quadratic Average: 54.964393615722656
Nearest Class Center Accuracy: 0.815

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17467361688613892
Inter Cos: 0.2037414163351059
Norm Quadratic Average: 33.62477493286133
Nearest Class Center Accuracy: 0.8485

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1987585425376892
Inter Cos: 0.21088042855262756
Norm Quadratic Average: 31.073450088500977
Nearest Class Center Accuracy: 0.885

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2598382830619812
Inter Cos: 0.2037915289402008
Norm Quadratic Average: 17.701702117919922
Nearest Class Center Accuracy: 0.935

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3747764229774475
Inter Cos: 0.224814772605896
Norm Quadratic Average: 13.316239356994629
Nearest Class Center Accuracy: 0.958

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 92.40340423583984
Linear Weight Rank: 4031
Intra Cos: 0.5809274911880493
Inter Cos: 0.23425084352493286
Norm Quadratic Average: 58.79435348510742
Nearest Class Center Accuracy: 0.9675

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 37.35926055908203
Linear Weight Rank: 3670
Intra Cos: 0.6735191345214844
Inter Cos: 0.25150877237319946
Norm Quadratic Average: 37.596580505371094
Nearest Class Center Accuracy: 0.974

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.366819381713867
Linear Weight Rank: 10
Intra Cos: 0.7147351503372192
Inter Cos: 0.2853109836578369
Norm Quadratic Average: 28.808181762695312
Nearest Class Center Accuracy: 0.976

Output Layer:
Intra Cos: 0.7456074357032776
Inter Cos: 0.3906099796295166
Norm Quadratic Average: 20.22468376159668
Nearest Class Center Accuracy: 0.9715

