Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_True_dataset_MNIST_epochs_100_lr_0.001_model_type_vgg11_weight_decay_0.0003.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.08946067094802856
Inter Cos: 0.11311887204647064
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: 0.1048562079668045
Inter Cos: 0.11975552886724472
Norm Quadratic Average: 85.62128448486328
Nearest Class Center Accuracy: 0.835375

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14370547235012054
Inter Cos: 0.1348772794008255
Norm Quadratic Average: 59.975486755371094
Nearest Class Center Accuracy: 0.847875

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14222870767116547
Inter Cos: 0.11976329982280731
Norm Quadratic Average: 56.26145553588867
Nearest Class Center Accuracy: 0.86925

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1590687334537506
Inter Cos: 0.09446103870868683
Norm Quadratic Average: 34.268898010253906
Nearest Class Center Accuracy: 0.907

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16477757692337036
Inter Cos: 0.08811463415622711
Norm Quadratic Average: 35.42938232421875
Nearest Class Center Accuracy: 0.929875

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18551886081695557
Inter Cos: 0.06856489181518555
Norm Quadratic Average: 24.041597366333008
Nearest Class Center Accuracy: 0.97175

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2680920958518982
Inter Cos: 0.08778756856918335
Norm Quadratic Average: 18.796945571899414
Nearest Class Center Accuracy: 0.99625

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 99.92269134521484
Linear Weight Rank: 4031
Intra Cos: 0.4742496609687805
Inter Cos: 0.11744643747806549
Norm Quadratic Average: 118.0914535522461
Nearest Class Center Accuracy: 1.0

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 40.38663101196289
Linear Weight Rank: 3671
Intra Cos: 0.6172261834144592
Inter Cos: 0.14863213896751404
Norm Quadratic Average: 64.34420013427734
Nearest Class Center Accuracy: 1.0

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.272766590118408
Linear Weight Rank: 10
Intra Cos: 0.7460322976112366
Inter Cos: 0.18114350736141205
Norm Quadratic Average: 41.286502838134766
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9066554307937622
Inter Cos: 0.2649412155151367
Norm Quadratic Average: 22.531396865844727
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.09559715485572814
Accuracy: 0.9725
NC1 Within Class Collapse: 1.705988883972168
NC2 Equinorm: Features: 0.07885652780532837, Weights: 0.015331864356994629
NC2 Equiangle: Features: 0.2023495144314236, Weights: 0.08813679483201768
NC3 Self-Duality: 0.6424577236175537
NC4 NCC Mismatch: 0.009000000000000008

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10133806616067886
Inter Cos: 0.11957792192697525
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: 0.1201634481549263
Inter Cos: 0.12973448634147644
Norm Quadratic Average: 84.2415771484375
Nearest Class Center Accuracy: 0.8265

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14186492562294006
Inter Cos: 0.15340019762516022
Norm Quadratic Average: 59.027244567871094
Nearest Class Center Accuracy: 0.845

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13753701746463776
Inter Cos: 0.13759702444076538
Norm Quadratic Average: 55.51740264892578
Nearest Class Center Accuracy: 0.859

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1492842584848404
Inter Cos: 0.114708311855793
Norm Quadratic Average: 34.012901306152344
Nearest Class Center Accuracy: 0.9

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15247854590415955
Inter Cos: 0.10975931584835052
Norm Quadratic Average: 35.24116516113281
Nearest Class Center Accuracy: 0.917

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1839110106229782
Inter Cos: 0.08558298647403717
Norm Quadratic Average: 23.864524841308594
Nearest Class Center Accuracy: 0.9495

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2463233917951584
Inter Cos: 0.10217303782701492
Norm Quadratic Average: 18.54843521118164
Nearest Class Center Accuracy: 0.9675

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 99.92269134521484
Linear Weight Rank: 4031
Intra Cos: 0.3952535092830658
Inter Cos: 0.12071681022644043
Norm Quadratic Average: 115.29986572265625
Nearest Class Center Accuracy: 0.976

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 40.38663101196289
Linear Weight Rank: 3671
Intra Cos: 0.5226693153381348
Inter Cos: 0.15414206683635712
Norm Quadratic Average: 62.50566482543945
Nearest Class Center Accuracy: 0.9745

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.272766590118408
Linear Weight Rank: 10
Intra Cos: 0.6300150156021118
Inter Cos: 0.1924775391817093
Norm Quadratic Average: 39.9388542175293
Nearest Class Center Accuracy: 0.973

Output Layer:
Intra Cos: 0.7903435826301575
Inter Cos: 0.2782096862792969
Norm Quadratic Average: 21.63956069946289
Nearest Class Center Accuracy: 0.972

