Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_338327_test_samples_None_train_samples_None_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: 0.09116753190755844
Inter Cos: 0.10967152565717697
Norm Quadratic Average: 23.567678451538086
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10922864824533463
Inter Cos: 0.13215529918670654
Norm Quadratic Average: 66.23278045654297
Nearest Class Center Accuracy: 0.8024333333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13139335811138153
Inter Cos: 0.1685827225446701
Norm Quadratic Average: 101.56816101074219
Nearest Class Center Accuracy: 0.79405

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14071892201900482
Inter Cos: 0.1828174889087677
Norm Quadratic Average: 173.53152465820312
Nearest Class Center Accuracy: 0.80125

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18010930716991425
Inter Cos: 0.1914321333169937
Norm Quadratic Average: 124.25312805175781
Nearest Class Center Accuracy: 0.8388333333333333

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20256999135017395
Inter Cos: 0.2105754017829895
Norm Quadratic Average: 97.22234344482422
Nearest Class Center Accuracy: 0.8676166666666667

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23095589876174927
Inter Cos: 0.2324502319097519
Norm Quadratic Average: 82.46298217773438
Nearest Class Center Accuracy: 0.8971

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2924386262893677
Inter Cos: 0.2608731985092163
Norm Quadratic Average: 64.19511413574219
Nearest Class Center Accuracy: 0.9315

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.29144567251205444
Inter Cos: 0.22319024801254272
Norm Quadratic Average: 24.89192008972168
Nearest Class Center Accuracy: 0.9350333333333334

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.35441383719444275
Inter Cos: 0.30230212211608887
Norm Quadratic Average: 15.720986366271973
Nearest Class Center Accuracy: 0.8914666666666666

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4098628759384155
Inter Cos: 0.34800100326538086
Norm Quadratic Average: 17.063920974731445
Nearest Class Center Accuracy: 0.9012666666666667

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4954737424850464
Inter Cos: 0.36931657791137695
Norm Quadratic Average: 20.83573341369629
Nearest Class Center Accuracy: 0.9437666666666666

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.552384614944458
Inter Cos: 0.3699920177459717
Norm Quadratic Average: 16.209318161010742
Nearest Class Center Accuracy: 0.9519666666666666

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.716210663318634
Inter Cos: 0.5347601175308228
Norm Quadratic Average: 14.156932830810547
Nearest Class Center Accuracy: 0.96925

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7796061635017395
Inter Cos: 0.5215772390365601
Norm Quadratic Average: 14.69073486328125
Nearest Class Center Accuracy: 0.9861666666666666

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8167240619659424
Inter Cos: 0.48230284452438354
Norm Quadratic Average: 15.562799453735352
Nearest Class Center Accuracy: 0.9926666666666667

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.7336969375610352
Linear Weight Rank: 601
Intra Cos: 0.8321547508239746
Inter Cos: 0.3900351822376251
Norm Quadratic Average: 64.11465454101562
Nearest Class Center Accuracy: 0.9964166666666666

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.760201096534729
Linear Weight Rank: 2728
Intra Cos: 0.8908874988555908
Inter Cos: 0.3247539699077606
Norm Quadratic Average: 43.59648132324219
Nearest Class Center Accuracy: 0.9986666666666667

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.7404552698135376
Linear Weight Rank: 9
Intra Cos: 0.9039148688316345
Inter Cos: 0.32352927327156067
Norm Quadratic Average: 28.963329315185547
Nearest Class Center Accuracy: 0.9988666666666667

Output Layer:
Intra Cos: 0.9493817090988159
Inter Cos: 0.4174196422100067
Norm Quadratic Average: 21.739948272705078
Nearest Class Center Accuracy: 0.9993166666666666

Test Set:
Average Loss: 0.03344421294360654
Accuracy: 0.9906
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.11136452108621597, Weights: 0.04993623495101929
NC2 Equiangle: Features: 0.2557171715630425, Weights: 0.23090616861979166
NC3 Self-Duality: 0.05987033247947693
NC4 NCC Mismatch: 0.006099999999999994

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582170248032
Inter Cos: 0.12048853188753128
Norm Quadratic Average: 23.595197677612305
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1207546517252922
Inter Cos: 0.14499786496162415
Norm Quadratic Average: 66.57112884521484
Nearest Class Center Accuracy: 0.8189

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14571025967597961
Inter Cos: 0.1844882071018219
Norm Quadratic Average: 101.9161605834961
Nearest Class Center Accuracy: 0.8111

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15552793443202972
Inter Cos: 0.19999562203884125
Norm Quadratic Average: 174.123291015625
Nearest Class Center Accuracy: 0.8173

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19235588610172272
Inter Cos: 0.20962132513523102
Norm Quadratic Average: 124.39967346191406
Nearest Class Center Accuracy: 0.8527

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2173188030719757
Inter Cos: 0.22915992140769958
Norm Quadratic Average: 97.39389038085938
Nearest Class Center Accuracy: 0.8806

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24579553306102753
Inter Cos: 0.2515949606895447
Norm Quadratic Average: 82.69163513183594
Nearest Class Center Accuracy: 0.9095

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.30748218297958374
Inter Cos: 0.2797574996948242
Norm Quadratic Average: 64.5802001953125
Nearest Class Center Accuracy: 0.9381

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.30757132172584534
Inter Cos: 0.24092239141464233
Norm Quadratic Average: 25.06165313720703
Nearest Class Center Accuracy: 0.9404

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3723493814468384
Inter Cos: 0.30136770009994507
Norm Quadratic Average: 15.826003074645996
Nearest Class Center Accuracy: 0.8986

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4292092025279999
Inter Cos: 0.375786691904068
Norm Quadratic Average: 17.186817169189453
Nearest Class Center Accuracy: 0.9073

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5166193842887878
Inter Cos: 0.3989311754703522
Norm Quadratic Average: 21.01582908630371
Nearest Class Center Accuracy: 0.9404

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5652608275413513
Inter Cos: 0.38881319761276245
Norm Quadratic Average: 16.326196670532227
Nearest Class Center Accuracy: 0.946

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7247405052185059
Inter Cos: 0.5332889556884766
Norm Quadratic Average: 14.281363487243652
Nearest Class Center Accuracy: 0.9622

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7923188209533691
Inter Cos: 0.5166833996772766
Norm Quadratic Average: 14.844084739685059
Nearest Class Center Accuracy: 0.9764

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8264747262001038
Inter Cos: 0.48541635274887085
Norm Quadratic Average: 15.741415977478027
Nearest Class Center Accuracy: 0.9831

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.7336969375610352
Linear Weight Rank: 601
Intra Cos: 0.837541937828064
Inter Cos: 0.39465999603271484
Norm Quadratic Average: 64.9155044555664
Nearest Class Center Accuracy: 0.9866

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.760201096534729
Linear Weight Rank: 2728
Intra Cos: 0.8904438018798828
Inter Cos: 0.34323886036872864
Norm Quadratic Average: 44.180049896240234
Nearest Class Center Accuracy: 0.9882

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.7404552698135376
Linear Weight Rank: 9
Intra Cos: 0.9007417559623718
Inter Cos: 0.34149476885795593
Norm Quadratic Average: 29.360774993896484
Nearest Class Center Accuracy: 0.9883

Output Layer:
Intra Cos: 0.9402135014533997
Inter Cos: 0.43450358510017395
Norm Quadratic Average: 22.04482078552246
Nearest Class Center Accuracy: 0.9899

