Model save path: ./New_Models/bn_False_dataset_CIFAR10_epochs_200_lr_0.01_model_type_vgg19_rand_seed_314159_test_samples_None_train_samples_None_weight_decay_0.003.pth.tar
Training Set:
Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.01989092119038105
Inter Cos: 0.10477277636528015
Norm Quadratic Average: 27.597183227539062
Nearest Class Center Accuracy: 0.26968

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.020316479727625847
Inter Cos: 0.09506303071975708
Norm Quadratic Average: 31.219385147094727
Nearest Class Center Accuracy: 0.33214

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.024264391511678696
Inter Cos: 0.10525447875261307
Norm Quadratic Average: 27.002309799194336
Nearest Class Center Accuracy: 0.42164

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02612178400158882
Inter Cos: 0.0976787582039833
Norm Quadratic Average: 32.70258331298828
Nearest Class Center Accuracy: 0.48598

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02755390852689743
Inter Cos: 0.08719345927238464
Norm Quadratic Average: 19.713247299194336
Nearest Class Center Accuracy: 0.56086

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.03651999682188034
Inter Cos: 0.08777124434709549
Norm Quadratic Average: 12.874835014343262
Nearest Class Center Accuracy: 0.60622

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.04642273858189583
Inter Cos: 0.08683359622955322
Norm Quadratic Average: 6.64660120010376
Nearest Class Center Accuracy: 0.63752

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.05154580995440483
Inter Cos: 0.07665206491947174
Norm Quadratic Average: 2.900986671447754
Nearest Class Center Accuracy: 0.69388

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1060003787279129
Inter Cos: 0.1653696745634079
Norm Quadratic Average: 0.759097695350647
Nearest Class Center Accuracy: 0.79178

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.38911888003349304
Inter Cos: 0.39992767572402954
Norm Quadratic Average: 0.360819935798645
Nearest Class Center Accuracy: 0.8896

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5289332270622253
Inter Cos: 0.47739967703819275
Norm Quadratic Average: 0.3526966869831085
Nearest Class Center Accuracy: 0.94542

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5160530805587769
Inter Cos: 0.4645353853702545
Norm Quadratic Average: 0.49826332926750183
Nearest Class Center Accuracy: 0.96222

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5799194574356079
Inter Cos: 0.5406748056411743
Norm Quadratic Average: 0.5294605493545532
Nearest Class Center Accuracy: 0.97622

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7483391165733337
Inter Cos: 0.644134521484375
Norm Quadratic Average: 0.8169977068901062
Nearest Class Center Accuracy: 0.98434

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.783661425113678
Inter Cos: 0.6077480316162109
Norm Quadratic Average: 1.416674256324768
Nearest Class Center Accuracy: 0.9891

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7973386645317078
Inter Cos: 0.5512559413909912
Norm Quadratic Average: 2.440516948699951
Nearest Class Center Accuracy: 0.99378

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.6146352291107178
Linear Weight Rank: 3073
Intra Cos: 0.7963012456893921
Inter Cos: 0.46952980756759644
Norm Quadratic Average: 16.314178466796875
Nearest Class Center Accuracy: 0.99768

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.6881425380706787
Linear Weight Rank: 2927
Intra Cos: 0.8274461030960083
Inter Cos: 0.4711662828922272
Norm Quadratic Average: 17.77897834777832
Nearest Class Center Accuracy: 0.99988

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.6325695514678955
Linear Weight Rank: 9
Intra Cos: 0.8407819867134094
Inter Cos: 0.3715306520462036
Norm Quadratic Average: 17.33032989501953
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.8633661270141602
Inter Cos: 0.42594900727272034
Norm Quadratic Average: 18.51833724975586
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.8462774480819703
Accuracy: 0.8243
NC1 Within Class Collapse: 5.437938690185547
NC2 Equinorm: Features: 0.20929375290870667, Weights: 0.07230810075998306
NC2 Equiangle: Features: 0.33715137905544706, Weights: 0.18746566772460938
NC3 Self-Duality: 0.21315540373325348
NC4 NCC Mismatch: 0.03400000000000003

Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.018526075407862663
Inter Cos: 0.1042705774307251
Norm Quadratic Average: 27.550134658813477
Nearest Class Center Accuracy: 0.284

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0180768221616745
Inter Cos: 0.09536486864089966
Norm Quadratic Average: 31.176733016967773
Nearest Class Center Accuracy: 0.3453

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.023464487865567207
Inter Cos: 0.10545292496681213
Norm Quadratic Average: 26.98969268798828
Nearest Class Center Accuracy: 0.4345

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.024380723014473915
Inter Cos: 0.09851539134979248
Norm Quadratic Average: 32.70708465576172
Nearest Class Center Accuracy: 0.4985

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.025459492579102516
Inter Cos: 0.08846418559551239
Norm Quadratic Average: 19.736160278320312
Nearest Class Center Accuracy: 0.5704

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.03426480293273926
Inter Cos: 0.08809079974889755
Norm Quadratic Average: 12.903484344482422
Nearest Class Center Accuracy: 0.6064

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.04371912032365799
Inter Cos: 0.08671293407678604
Norm Quadratic Average: 6.663342475891113
Nearest Class Center Accuracy: 0.6312

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0476493164896965
Inter Cos: 0.07852556556463242
Norm Quadratic Average: 2.905376672744751
Nearest Class Center Accuracy: 0.6782

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09913163632154465
Inter Cos: 0.16781821846961975
Norm Quadratic Average: 0.758083164691925
Nearest Class Center Accuracy: 0.7498

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.33302170038223267
Inter Cos: 0.3970945179462433
Norm Quadratic Average: 0.35941311717033386
Nearest Class Center Accuracy: 0.7996

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4264259338378906
Inter Cos: 0.4411519467830658
Norm Quadratic Average: 0.35024744272232056
Nearest Class Center Accuracy: 0.8118

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4347655773162842
Inter Cos: 0.440716028213501
Norm Quadratic Average: 0.4941324293613434
Nearest Class Center Accuracy: 0.8076

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4568890929222107
Inter Cos: 0.4864979684352875
Norm Quadratic Average: 0.5234590768814087
Nearest Class Center Accuracy: 0.8

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5588778853416443
Inter Cos: 0.5605126023292542
Norm Quadratic Average: 0.8051021099090576
Nearest Class Center Accuracy: 0.8049

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5489705204963684
Inter Cos: 0.5521777272224426
Norm Quadratic Average: 1.3932955265045166
Nearest Class Center Accuracy: 0.8094

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5295928716659546
Inter Cos: 0.5294700860977173
Norm Quadratic Average: 2.3956305980682373
Nearest Class Center Accuracy: 0.8154

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.6146352291107178
Linear Weight Rank: 3073
Intra Cos: 0.5197381377220154
Inter Cos: 0.45693328976631165
Norm Quadratic Average: 16.004892349243164
Nearest Class Center Accuracy: 0.8175

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.6881425380706787
Linear Weight Rank: 2927
Intra Cos: 0.5088316202163696
Inter Cos: 0.47482284903526306
Norm Quadratic Average: 17.409496307373047
Nearest Class Center Accuracy: 0.8215

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.6325695514678955
Linear Weight Rank: 9
Intra Cos: 0.502848744392395
Inter Cos: 0.43161144852638245
Norm Quadratic Average: 16.944124221801758
Nearest Class Center Accuracy: 0.8237

Output Layer:
Intra Cos: 0.4897891879081726
Inter Cos: 0.46241772174835205
Norm Quadratic Average: 18.071269989013672
Nearest Class Center Accuracy: 0.8218

