Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_True_dataset_MNIST_epochs_100_lr_0.001_model_type_vgg11_weight_decay_0.001.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: 0.11311884969472885
Norm Quadratic Average: 23.53293800354004
Nearest Class Center Accuracy: 0.81175

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09970792382955551
Inter Cos: 0.12252239137887955
Norm Quadratic Average: 90.9190444946289
Nearest Class Center Accuracy: 0.836375

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1461758315563202
Inter Cos: 0.13672395050525665
Norm Quadratic Average: 55.25235366821289
Nearest Class Center Accuracy: 0.858375

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.144615039229393
Inter Cos: 0.12542493641376495
Norm Quadratic Average: 54.020225524902344
Nearest Class Center Accuracy: 0.875375

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.177565336227417
Inter Cos: 0.1112281084060669
Norm Quadratic Average: 33.487457275390625
Nearest Class Center Accuracy: 0.9115

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18097880482673645
Inter Cos: 0.09199230372905731
Norm Quadratic Average: 34.691226959228516
Nearest Class Center Accuracy: 0.932125

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20245453715324402
Inter Cos: 0.0955752283334732
Norm Quadratic Average: 23.88506507873535
Nearest Class Center Accuracy: 0.9735

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28733041882514954
Inter Cos: 0.09144478291273117
Norm Quadratic Average: 18.163780212402344
Nearest Class Center Accuracy: 0.996

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 96.62922668457031
Linear Weight Rank: 4031
Intra Cos: 0.49726560711860657
Inter Cos: 0.10822051763534546
Norm Quadratic Average: 115.48072814941406
Nearest Class Center Accuracy: 1.0

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 39.0543098449707
Linear Weight Rank: 3670
Intra Cos: 0.6396199464797974
Inter Cos: 0.13333669304847717
Norm Quadratic Average: 61.313209533691406
Nearest Class Center Accuracy: 1.0

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2087528705596924
Linear Weight Rank: 10
Intra Cos: 0.7537149786949158
Inter Cos: 0.14966043829917908
Norm Quadratic Average: 38.307125091552734
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9064016342163086
Inter Cos: 0.24684089422225952
Norm Quadratic Average: 20.519268035888672
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.09361316701769828
Accuracy: 0.976
NC1 Within Class Collapse: 1.6625964641571045
NC2 Equinorm: Features: 0.06460371613502502, Weights: 0.008857861161231995
NC2 Equiangle: Features: 0.19514045715332032, Weights: 0.08892012702094185
NC3 Self-Duality: 0.6223556995391846
NC4 NCC Mismatch: 0.005499999999999949

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10133809596300125
Inter Cos: 0.11957789212465286
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: 0.12545590102672577
Inter Cos: 0.12784823775291443
Norm Quadratic Average: 89.65880584716797
Nearest Class Center Accuracy: 0.8305

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1568700224161148
Inter Cos: 0.1457100510597229
Norm Quadratic Average: 54.7816047668457
Nearest Class Center Accuracy: 0.849

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15398627519607544
Inter Cos: 0.1262494921684265
Norm Quadratic Average: 53.54804611206055
Nearest Class Center Accuracy: 0.866

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17639322578907013
Inter Cos: 0.11244736611843109
Norm Quadratic Average: 33.432952880859375
Nearest Class Center Accuracy: 0.898

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1808990240097046
Inter Cos: 0.09724951535463333
Norm Quadratic Average: 34.67844772338867
Nearest Class Center Accuracy: 0.92

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2027306705713272
Inter Cos: 0.10894979536533356
Norm Quadratic Average: 23.911901473999023
Nearest Class Center Accuracy: 0.945

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26846563816070557
Inter Cos: 0.09487666934728622
Norm Quadratic Average: 18.081310272216797
Nearest Class Center Accuracy: 0.972

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 96.62922668457031
Linear Weight Rank: 4031
Intra Cos: 0.4166716933250427
Inter Cos: 0.11076029390096664
Norm Quadratic Average: 113.1827392578125
Nearest Class Center Accuracy: 0.9775

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 39.0543098449707
Linear Weight Rank: 3670
Intra Cos: 0.5430488586425781
Inter Cos: 0.14092180132865906
Norm Quadratic Average: 59.71023941040039
Nearest Class Center Accuracy: 0.977

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2087528705596924
Linear Weight Rank: 10
Intra Cos: 0.6505575180053711
Inter Cos: 0.15876707434654236
Norm Quadratic Average: 37.12163543701172
Nearest Class Center Accuracy: 0.9765

Output Layer:
Intra Cos: 0.8063901662826538
Inter Cos: 0.2524460256099701
Norm Quadratic Average: 19.747663497924805
Nearest Class Center Accuracy: 0.9755

