Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_314159_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: 0.09116753935813904
Inter Cos: 0.10967151075601578
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: 0.10039538890123367
Inter Cos: 0.12309455126523972
Norm Quadratic Average: 62.504356384277344
Nearest Class Center Accuracy: 0.8067333333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12505972385406494
Inter Cos: 0.15635108947753906
Norm Quadratic Average: 83.28399658203125
Nearest Class Center Accuracy: 0.8112166666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13034062087535858
Inter Cos: 0.16727298498153687
Norm Quadratic Average: 137.3266143798828
Nearest Class Center Accuracy: 0.81515

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18940599262714386
Inter Cos: 0.18274715542793274
Norm Quadratic Average: 107.63370513916016
Nearest Class Center Accuracy: 0.8607333333333334

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21683594584465027
Inter Cos: 0.18799585103988647
Norm Quadratic Average: 102.18080139160156
Nearest Class Center Accuracy: 0.8872666666666666

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23489904403686523
Inter Cos: 0.19486087560653687
Norm Quadratic Average: 97.45195770263672
Nearest Class Center Accuracy: 0.9044

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26642024517059326
Inter Cos: 0.1940399557352066
Norm Quadratic Average: 75.51991271972656
Nearest Class Center Accuracy: 0.9301166666666667

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.27259451150894165
Inter Cos: 0.19209469854831696
Norm Quadratic Average: 29.127281188964844
Nearest Class Center Accuracy: 0.9563166666666667

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4557289779186249
Inter Cos: 0.22983506321907043
Norm Quadratic Average: 15.38437557220459
Nearest Class Center Accuracy: 0.973

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5427005887031555
Inter Cos: 0.28028616309165955
Norm Quadratic Average: 13.86331844329834
Nearest Class Center Accuracy: 0.9756

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5924296975135803
Inter Cos: 0.31495529413223267
Norm Quadratic Average: 14.965276718139648
Nearest Class Center Accuracy: 0.9821166666666666

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.594821572303772
Inter Cos: 0.35993075370788574
Norm Quadratic Average: 9.970813751220703
Nearest Class Center Accuracy: 0.97295

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7831448316574097
Inter Cos: 0.3749896287918091
Norm Quadratic Average: 8.79164981842041
Nearest Class Center Accuracy: 0.9818833333333333

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8457461595535278
Inter Cos: 0.29675647616386414
Norm Quadratic Average: 9.73295783996582
Nearest Class Center Accuracy: 0.9887333333333334

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8770169615745544
Inter Cos: 0.338887482881546
Norm Quadratic Average: 10.245104789733887
Nearest Class Center Accuracy: 0.9914166666666666

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.881502628326416
Linear Weight Rank: 4029
Intra Cos: 0.8909627199172974
Inter Cos: 0.2536754012107849
Norm Quadratic Average: 48.85093307495117
Nearest Class Center Accuracy: 0.9938166666666667

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.2518186569213867
Linear Weight Rank: 3641
Intra Cos: 0.9195532202720642
Inter Cos: 0.24770866334438324
Norm Quadratic Average: 38.68667984008789
Nearest Class Center Accuracy: 0.99685

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.0202221870422363
Linear Weight Rank: 9
Intra Cos: 0.9240483045578003
Inter Cos: 0.222500741481781
Norm Quadratic Average: 29.181167602539062
Nearest Class Center Accuracy: 0.9981833333333333

Output Layer:
Intra Cos: 0.9171522855758667
Inter Cos: 0.258345365524292
Norm Quadratic Average: 24.318578720092773
Nearest Class Center Accuracy: 0.9995666666666667

Test Set:
Average Loss: 0.02931599906870979
Accuracy: 0.9926
NC1 Within Class Collapse: 1.132171869277954
NC2 Equinorm: Features: 0.1483636051416397, Weights: 0.040351249277591705
NC2 Equiangle: Features: 0.2725196838378906, Weights: 0.2024760988023546
NC3 Self-Duality: 0.1189647763967514
NC4 NCC Mismatch: 0.006399999999999961

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
Inter Cos: 0.12048851698637009
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: 0.11198822408914566
Inter Cos: 0.13545091450214386
Norm Quadratic Average: 62.71924591064453
Nearest Class Center Accuracy: 0.8198

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13978159427642822
Inter Cos: 0.17154011130332947
Norm Quadratic Average: 83.37003326416016
Nearest Class Center Accuracy: 0.8266

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1451158970594406
Inter Cos: 0.18380194902420044
Norm Quadratic Average: 137.52584838867188
Nearest Class Center Accuracy: 0.8295

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2052290141582489
Inter Cos: 0.2002655565738678
Norm Quadratic Average: 107.63075256347656
Nearest Class Center Accuracy: 0.8744

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23405177891254425
Inter Cos: 0.2057052105665207
Norm Quadratic Average: 102.1497802734375
Nearest Class Center Accuracy: 0.9005

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25000056624412537
Inter Cos: 0.21232359111309052
Norm Quadratic Average: 97.45379638671875
Nearest Class Center Accuracy: 0.9158

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2783944010734558
Inter Cos: 0.21111902594566345
Norm Quadratic Average: 75.68624877929688
Nearest Class Center Accuracy: 0.938

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28306594491004944
Inter Cos: 0.2073671966791153
Norm Quadratic Average: 29.217679977416992
Nearest Class Center Accuracy: 0.9587

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.47319650650024414
Inter Cos: 0.23830148577690125
Norm Quadratic Average: 15.448565483093262
Nearest Class Center Accuracy: 0.9726

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5463337302207947
Inter Cos: 0.3041118085384369
Norm Quadratic Average: 13.941873550415039
Nearest Class Center Accuracy: 0.9735

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5957443714141846
Inter Cos: 0.33417144417762756
Norm Quadratic Average: 15.071751594543457
Nearest Class Center Accuracy: 0.9785

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5990645289421082
Inter Cos: 0.36687278747558594
Norm Quadratic Average: 10.048562049865723
Nearest Class Center Accuracy: 0.9697

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7944390177726746
Inter Cos: 0.376116544008255
Norm Quadratic Average: 8.886938095092773
Nearest Class Center Accuracy: 0.9745

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8494812846183777
Inter Cos: 0.2965376675128937
Norm Quadratic Average: 9.852231979370117
Nearest Class Center Accuracy: 0.9804

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8773227334022522
Inter Cos: 0.3380841612815857
Norm Quadratic Average: 10.370550155639648
Nearest Class Center Accuracy: 0.9836

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.881502628326416
Linear Weight Rank: 4029
Intra Cos: 0.8901938796043396
Inter Cos: 0.2573454976081848
Norm Quadratic Average: 49.39011764526367
Nearest Class Center Accuracy: 0.9853

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.2518186569213867
Linear Weight Rank: 3641
Intra Cos: 0.9168670773506165
Inter Cos: 0.2594107687473297
Norm Quadratic Average: 39.117027282714844
Nearest Class Center Accuracy: 0.9884

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.0202221870422363
Linear Weight Rank: 9
Intra Cos: 0.9248819947242737
Inter Cos: 0.23636803030967712
Norm Quadratic Average: 29.508867263793945
Nearest Class Center Accuracy: 0.9892

Output Layer:
Intra Cos: 0.9179010391235352
Inter Cos: 0.25554096698760986
Norm Quadratic Average: 24.584354400634766
Nearest Class Center Accuracy: 0.9911

