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_0.007.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: 0.05969248339533806
Inter Cos: 0.07784024626016617
Norm Quadratic Average: 2.5382134914398193
Nearest Class Center Accuracy: 0.8083833333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10557614266872406
Inter Cos: 0.10000919550657272
Norm Quadratic Average: 1.4709551334381104
Nearest Class Center Accuracy: 0.8753

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10535260289907455
Inter Cos: 0.09963023662567139
Norm Quadratic Average: 1.225574254989624
Nearest Class Center Accuracy: 0.8795333333333333

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17429165542125702
Inter Cos: 0.12258539348840714
Norm Quadratic Average: 0.798223078250885
Nearest Class Center Accuracy: 0.9383666666666667

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23406335711479187
Inter Cos: 0.13581153750419617
Norm Quadratic Average: 0.569563090801239
Nearest Class Center Accuracy: 0.9634833333333334

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.31737375259399414
Inter Cos: 0.15030710399150848
Norm Quadratic Average: 0.4898010194301605
Nearest Class Center Accuracy: 0.975

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3537663221359253
Inter Cos: 0.13563038408756256
Norm Quadratic Average: 0.434857577085495
Nearest Class Center Accuracy: 0.9778166666666667

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.414028137922287
Inter Cos: 0.15283967554569244
Norm Quadratic Average: 0.27430686354637146
Nearest Class Center Accuracy: 0.9929166666666667

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6467887759208679
Inter Cos: 0.2901870906352997
Norm Quadratic Average: 0.1771908551454544
Nearest Class Center Accuracy: 0.99745

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8056492209434509
Inter Cos: 0.2559802532196045
Norm Quadratic Average: 0.1634548306465149
Nearest Class Center Accuracy: 0.9997833333333334

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.831779956817627
Inter Cos: 0.12352223694324493
Norm Quadratic Average: 0.1732328236103058
Nearest Class Center Accuracy: 0.9999666666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8868563175201416
Inter Cos: 0.1226227805018425
Norm Quadratic Average: 0.1912897229194641
Nearest Class Center Accuracy: 0.99995

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9826175570487976
Inter Cos: 0.023408139124512672
Norm Quadratic Average: 0.22578607499599457
Nearest Class Center Accuracy: 0.9999833333333333

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9957829713821411
Inter Cos: 0.03836977481842041
Norm Quadratic Average: 0.48123329877853394
Nearest Class Center Accuracy: 0.9999833333333333

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9988220930099487
Inter Cos: 0.042451631277799606
Norm Quadratic Average: 1.0893021821975708
Nearest Class Center Accuracy: 0.9999833333333333

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.106685161590576
Linear Weight Rank: 10
Intra Cos: 0.9992436766624451
Inter Cos: 0.15433204174041748
Norm Quadratic Average: 24.86113166809082
Nearest Class Center Accuracy: 0.9999833333333333

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.1083757877349854
Linear Weight Rank: 1433
Intra Cos: 0.9994389414787292
Inter Cos: 0.19938962161540985
Norm Quadratic Average: 16.958864212036133
Nearest Class Center Accuracy: 0.9999833333333333

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.109081268310547
Linear Weight Rank: 9
Intra Cos: 0.9994685053825378
Inter Cos: 0.17311227321624756
Norm Quadratic Average: 11.714460372924805
Nearest Class Center Accuracy: 0.9999833333333333

Output Layer:
Intra Cos: 0.9995747208595276
Inter Cos: 0.1309899389743805
Norm Quadratic Average: 8.537718772888184
Nearest Class Center Accuracy: 0.9999833333333333

Test Set:
Average Loss: 0.020514494854211807
Accuracy: 0.9958
NC1 Within Class Collapse: 0.07820047438144684
NC2 Equinorm: Features: 0.01775149442255497, Weights: 0.005771825090050697
NC2 Equiangle: Features: 0.10627612007988824, Weights: 0.06774924066331652
NC3 Self-Duality: 0.02706296555697918
NC4 NCC Mismatch: 0.00019999999999997797

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: 0.06776159256696701
Inter Cos: 0.08044163137674332
Norm Quadratic Average: 2.530714273452759
Nearest Class Center Accuracy: 0.8196

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11564911156892776
Inter Cos: 0.10152051597833633
Norm Quadratic Average: 1.4611567258834839
Nearest Class Center Accuracy: 0.8861

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11564677953720093
Inter Cos: 0.10203654319047928
Norm Quadratic Average: 1.2223683595657349
Nearest Class Center Accuracy: 0.8892

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18675793707370758
Inter Cos: 0.1291552186012268
Norm Quadratic Average: 0.7953864932060242
Nearest Class Center Accuracy: 0.9433

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24924491345882416
Inter Cos: 0.14149273931980133
Norm Quadratic Average: 0.5683650970458984
Nearest Class Center Accuracy: 0.9652

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.33416157960891724
Inter Cos: 0.14949770271778107
Norm Quadratic Average: 0.4893817901611328
Nearest Class Center Accuracy: 0.9757

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.36934053897857666
Inter Cos: 0.1418674737215042
Norm Quadratic Average: 0.4342939853668213
Nearest Class Center Accuracy: 0.9777

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.42964881658554077
Inter Cos: 0.16661034524440765
Norm Quadratic Average: 0.2740737795829773
Nearest Class Center Accuracy: 0.9888

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6592996120452881
Inter Cos: 0.3003503382205963
Norm Quadratic Average: 0.17740625143051147
Nearest Class Center Accuracy: 0.992

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8122031688690186
Inter Cos: 0.26657798886299133
Norm Quadratic Average: 0.16378504037857056
Nearest Class Center Accuracy: 0.9942

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8358665704727173
Inter Cos: 0.13192495703697205
Norm Quadratic Average: 0.17319339513778687
Nearest Class Center Accuracy: 0.9953

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8842816948890686
Inter Cos: 0.1313597708940506
Norm Quadratic Average: 0.19104726612567902
Nearest Class Center Accuracy: 0.9953

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9733014106750488
Inter Cos: 0.0351567268371582
Norm Quadratic Average: 0.22504040598869324
Nearest Class Center Accuracy: 0.9956

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9776145219802856
Inter Cos: 0.046405479311943054
Norm Quadratic Average: 0.47964009642601013
Nearest Class Center Accuracy: 0.9958

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9774090647697449
Inter Cos: 0.04656140133738518
Norm Quadratic Average: 1.0852802991867065
Nearest Class Center Accuracy: 0.9959

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.106685161590576
Linear Weight Rank: 10
Intra Cos: 0.978771984577179
Inter Cos: 0.15590432286262512
Norm Quadratic Average: 24.7731990814209
Nearest Class Center Accuracy: 0.9957

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.1083757877349854
Linear Weight Rank: 1433
Intra Cos: 0.9797191023826599
Inter Cos: 0.19971278309822083
Norm Quadratic Average: 16.8962459564209
Nearest Class Center Accuracy: 0.9957

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.109081268310547
Linear Weight Rank: 9
Intra Cos: 0.9798387885093689
Inter Cos: 0.17403876781463623
Norm Quadratic Average: 11.670181274414062
Nearest Class Center Accuracy: 0.9956

Output Layer:
Intra Cos: 0.9804090261459351
Inter Cos: 0.14299005270004272
Norm Quadratic Average: 8.504404067993164
Nearest Class Center Accuracy: 0.9956

