Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_338327_test_samples_None_train_samples_None_weight_decay_0.0001.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: 0.0717189833521843
Inter Cos: 0.08826989680528641
Norm Quadratic Average: 103.04496002197266
Nearest Class Center Accuracy: 0.8236

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1107683852314949
Inter Cos: 0.11283821612596512
Norm Quadratic Average: 63.79889678955078
Nearest Class Center Accuracy: 0.8609

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11747219413518906
Inter Cos: 0.11140294373035431
Norm Quadratic Average: 60.581092834472656
Nearest Class Center Accuracy: 0.876

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18962237238883972
Inter Cos: 0.13127635419368744
Norm Quadratic Average: 38.92336654663086
Nearest Class Center Accuracy: 0.9238833333333333

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21913504600524902
Inter Cos: 0.1342165321111679
Norm Quadratic Average: 40.909629821777344
Nearest Class Center Accuracy: 0.9436

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2464582920074463
Inter Cos: 0.13623696565628052
Norm Quadratic Average: 42.4877815246582
Nearest Class Center Accuracy: 0.9590333333333333

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26623278856277466
Inter Cos: 0.1332368701696396
Norm Quadratic Average: 43.354766845703125
Nearest Class Center Accuracy: 0.9679

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.31289762258529663
Inter Cos: 0.15782994031906128
Norm Quadratic Average: 28.477296829223633
Nearest Class Center Accuracy: 0.98895

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3964047431945801
Inter Cos: 0.18926331400871277
Norm Quadratic Average: 28.971385955810547
Nearest Class Center Accuracy: 0.995

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.49707284569740295
Inter Cos: 0.19421781599521637
Norm Quadratic Average: 30.53562355041504
Nearest Class Center Accuracy: 0.9978

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5905336737632751
Inter Cos: 0.18293675780296326
Norm Quadratic Average: 31.109304428100586
Nearest Class Center Accuracy: 0.9994833333333333

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7185980081558228
Inter Cos: 0.265691876411438
Norm Quadratic Average: 24.657989501953125
Nearest Class Center Accuracy: 0.9996666666666667

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8520903587341309
Inter Cos: 0.21998806297779083
Norm Quadratic Average: 16.427465438842773
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9121935963630676
Inter Cos: 0.18039540946483612
Norm Quadratic Average: 17.302082061767578
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9379255175590515
Inter Cos: 0.1165795624256134
Norm Quadratic Average: 18.288576126098633
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.80482482910156
Linear Weight Rank: 4031
Intra Cos: 0.9568378925323486
Inter Cos: 0.05706975236535072
Norm Quadratic Average: 121.90642547607422
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.938922882080078
Linear Weight Rank: 3671
Intra Cos: 0.985988199710846
Inter Cos: 0.04703018441796303
Norm Quadratic Average: 64.86445617675781
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.3968887329101562
Linear Weight Rank: 10
Intra Cos: 0.9853875637054443
Inter Cos: 0.03954574093222618
Norm Quadratic Average: 31.180992126464844
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9985280632972717
Inter Cos: 0.20580974221229553
Norm Quadratic Average: 18.635034561157227
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.024368457284436592
Accuracy: 0.9952
NC1 Within Class Collapse: 0.20858293771743774
NC2 Equinorm: Features: 0.045438941568136215, Weights: 0.011165321804583073
NC2 Equiangle: Features: 0.07876411543952094, Weights: 0.07748265796237522
NC3 Self-Duality: 0.5791512727737427
NC4 NCC Mismatch: 9.999999999998899e-05

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: 0.08082606643438339
Inter Cos: 0.09193116426467896
Norm Quadratic Average: 102.87355041503906
Nearest Class Center Accuracy: 0.8374

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12308818846940994
Inter Cos: 0.1145550012588501
Norm Quadratic Average: 63.38533020019531
Nearest Class Center Accuracy: 0.8753

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.129710391163826
Inter Cos: 0.12313899397850037
Norm Quadratic Average: 60.178001403808594
Nearest Class Center Accuracy: 0.8869

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20547698438167572
Inter Cos: 0.14363524317741394
Norm Quadratic Average: 38.61871337890625
Nearest Class Center Accuracy: 0.9327

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23558290302753448
Inter Cos: 0.14159835875034332
Norm Quadratic Average: 40.59425735473633
Nearest Class Center Accuracy: 0.9488

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26346883177757263
Inter Cos: 0.13339707255363464
Norm Quadratic Average: 42.201690673828125
Nearest Class Center Accuracy: 0.9607

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2832728922367096
Inter Cos: 0.1302131563425064
Norm Quadratic Average: 43.1248893737793
Nearest Class Center Accuracy: 0.9686

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3240808844566345
Inter Cos: 0.15973390638828278
Norm Quadratic Average: 28.381444931030273
Nearest Class Center Accuracy: 0.9872

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.40508583188056946
Inter Cos: 0.18950583040714264
Norm Quadratic Average: 28.891530990600586
Nearest Class Center Accuracy: 0.991

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5034311413764954
Inter Cos: 0.19271326065063477
Norm Quadratic Average: 30.47703742980957
Nearest Class Center Accuracy: 0.9924

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5936220288276672
Inter Cos: 0.17921294271945953
Norm Quadratic Average: 31.064197540283203
Nearest Class Center Accuracy: 0.9937

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7160645127296448
Inter Cos: 0.25988128781318665
Norm Quadratic Average: 24.62963104248047
Nearest Class Center Accuracy: 0.9928

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8436226844787598
Inter Cos: 0.2125406414270401
Norm Quadratic Average: 16.40995979309082
Nearest Class Center Accuracy: 0.9938

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.900449812412262
Inter Cos: 0.1728663444519043
Norm Quadratic Average: 17.277841567993164
Nearest Class Center Accuracy: 0.9946

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9217532873153687
Inter Cos: 0.10982771962881088
Norm Quadratic Average: 18.259666442871094
Nearest Class Center Accuracy: 0.995

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.80482482910156
Linear Weight Rank: 4031
Intra Cos: 0.93401700258255
Inter Cos: 0.059918250888586044
Norm Quadratic Average: 121.7016830444336
Nearest Class Center Accuracy: 0.995

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.938922882080078
Linear Weight Rank: 3671
Intra Cos: 0.9623690247535706
Inter Cos: 0.04886457696557045
Norm Quadratic Average: 64.7057876586914
Nearest Class Center Accuracy: 0.9953

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.3968887329101562
Linear Weight Rank: 10
Intra Cos: 0.9601860046386719
Inter Cos: 0.04092084616422653
Norm Quadratic Average: 31.110309600830078
Nearest Class Center Accuracy: 0.9951

Output Layer:
Intra Cos: 0.985609233379364
Inter Cos: 0.21052111685276031
Norm Quadratic Average: 18.568668365478516
Nearest Class Center Accuracy: 0.9951

