Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_265358_test_samples_None_train_samples_None_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: 0.09116753190755844
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567670822143555
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1194334328174591
Inter Cos: 0.14274385571479797
Norm Quadratic Average: 39.59005355834961
Nearest Class Center Accuracy: 0.8173666666666667

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1877107471227646
Inter Cos: 0.17217953503131866
Norm Quadratic Average: 39.489723205566406
Nearest Class Center Accuracy: 0.85455

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2189904898405075
Inter Cos: 0.1920868456363678
Norm Quadratic Average: 37.11538314819336
Nearest Class Center Accuracy: 0.8945666666666666

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.240243062376976
Inter Cos: 0.1812596619129181
Norm Quadratic Average: 16.52130889892578
Nearest Class Center Accuracy: 0.94395

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.39302879571914673
Inter Cos: 0.27752262353897095
Norm Quadratic Average: 9.525032043457031
Nearest Class Center Accuracy: 0.9725333333333334

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5595489740371704
Inter Cos: 0.3371855914592743
Norm Quadratic Average: 5.64387845993042
Nearest Class Center Accuracy: 0.9908166666666667

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7867444753646851
Inter Cos: 0.3168565034866333
Norm Quadratic Average: 5.3143768310546875
Nearest Class Center Accuracy: 0.9972166666666666

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.163398504257202
Linear Weight Rank: 4028
Intra Cos: 0.8674465417861938
Inter Cos: 0.26219624280929565
Norm Quadratic Average: 28.210060119628906
Nearest Class Center Accuracy: 0.99845

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.5638904571533203
Linear Weight Rank: 3639
Intra Cos: 0.9099074006080627
Inter Cos: 0.2658122479915619
Norm Quadratic Average: 24.88187026977539
Nearest Class Center Accuracy: 0.9994166666666666

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.368995189666748
Linear Weight Rank: 9
Intra Cos: 0.9213138222694397
Inter Cos: 0.24743784964084625
Norm Quadratic Average: 21.1077823638916
Nearest Class Center Accuracy: 0.99975

Output Layer:
Intra Cos: 0.9453779458999634
Inter Cos: 0.30334120988845825
Norm Quadratic Average: 20.260107040405273
Nearest Class Center Accuracy: 0.9999166666666667

Test Set:
Average Loss: 0.021273355832509697
Accuracy: 0.993
NC1 Within Class Collapse: 1.0773671865463257
NC2 Equinorm: Features: 0.09653820097446442, Weights: 0.04199757054448128
NC2 Equiangle: Features: 0.2446154064602322, Weights: 0.178814697265625
NC3 Self-Duality: 0.0826428085565567
NC4 NCC Mismatch: 0.0023999999999999577

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219583660364151
Inter Cos: 0.12048852443695068
Norm Quadratic Average: 23.595197677612305
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.132874995470047
Inter Cos: 0.1565825343132019
Norm Quadratic Average: 39.57965087890625
Nearest Class Center Accuracy: 0.8313

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2011205404996872
Inter Cos: 0.18389756977558136
Norm Quadratic Average: 39.37623977661133
Nearest Class Center Accuracy: 0.8718

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23278415203094482
Inter Cos: 0.2036007195711136
Norm Quadratic Average: 37.0411376953125
Nearest Class Center Accuracy: 0.9051

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.251076877117157
Inter Cos: 0.1967940330505371
Norm Quadratic Average: 16.49163246154785
Nearest Class Center Accuracy: 0.9516

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.408783495426178
Inter Cos: 0.2971561849117279
Norm Quadratic Average: 9.531431198120117
Nearest Class Center Accuracy: 0.9752

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5640385150909424
Inter Cos: 0.3582487106323242
Norm Quadratic Average: 5.673101425170898
Nearest Class Center Accuracy: 0.9866

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7905828356742859
Inter Cos: 0.32818859815597534
Norm Quadratic Average: 5.357755661010742
Nearest Class Center Accuracy: 0.99

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.163398504257202
Linear Weight Rank: 4028
Intra Cos: 0.8684979677200317
Inter Cos: 0.276994526386261
Norm Quadratic Average: 28.438276290893555
Nearest Class Center Accuracy: 0.9908

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.5638904571533203
Linear Weight Rank: 3639
Intra Cos: 0.9090423583984375
Inter Cos: 0.28668758273124695
Norm Quadratic Average: 25.072467803955078
Nearest Class Center Accuracy: 0.9916

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.368995189666748
Linear Weight Rank: 9
Intra Cos: 0.9190472960472107
Inter Cos: 0.2672567367553711
Norm Quadratic Average: 21.266618728637695
Nearest Class Center Accuracy: 0.9924

Output Layer:
Intra Cos: 0.9387738108634949
Inter Cos: 0.3230433762073517
Norm Quadratic Average: 20.410409927368164
Nearest Class Center Accuracy: 0.993

