Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_314159_test_samples_None_train_samples_None_weight_decay_0.0005.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: 0.060646310448646545
Inter Cos: 0.0749090313911438
Norm Quadratic Average: 18.382465362548828
Nearest Class Center Accuracy: 0.8293833333333334

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0977388322353363
Inter Cos: 0.0925145074725151
Norm Quadratic Average: 12.222477912902832
Nearest Class Center Accuracy: 0.8753833333333333

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10020658373832703
Inter Cos: 0.09376952797174454
Norm Quadratic Average: 12.439138412475586
Nearest Class Center Accuracy: 0.8869666666666667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17476561665534973
Inter Cos: 0.11713840067386627
Norm Quadratic Average: 8.725056648254395
Nearest Class Center Accuracy: 0.9371333333333334

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2023244947195053
Inter Cos: 0.12156568467617035
Norm Quadratic Average: 8.6663818359375
Nearest Class Center Accuracy: 0.9597

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23676170408725739
Inter Cos: 0.11765789240598679
Norm Quadratic Average: 8.912595748901367
Nearest Class Center Accuracy: 0.9707833333333333

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.27034279704093933
Inter Cos: 0.1096792221069336
Norm Quadratic Average: 9.285369873046875
Nearest Class Center Accuracy: 0.9775666666666667

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3563923239707947
Inter Cos: 0.12766024470329285
Norm Quadratic Average: 6.629382133483887
Nearest Class Center Accuracy: 0.9935333333333334

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5314146876335144
Inter Cos: 0.16900648176670074
Norm Quadratic Average: 7.222292423248291
Nearest Class Center Accuracy: 0.9978833333333333

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6618276238441467
Inter Cos: 0.1301613599061966
Norm Quadratic Average: 7.673666000366211
Nearest Class Center Accuracy: 0.9991666666666666

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7620351910591125
Inter Cos: 0.07264988124370575
Norm Quadratic Average: 7.829037189483643
Nearest Class Center Accuracy: 0.9997666666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8589066863059998
Inter Cos: 0.10602627694606781
Norm Quadratic Average: 6.7479472160339355
Nearest Class Center Accuracy: 0.9999

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9534117579460144
Inter Cos: 0.057546038180589676
Norm Quadratic Average: 4.176762104034424
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9718782901763916
Inter Cos: 0.04137028008699417
Norm Quadratic Average: 4.226950168609619
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9869495034217834
Inter Cos: -0.004834545776247978
Norm Quadratic Average: 4.336410045623779
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.675838470458984
Linear Weight Rank: 4031
Intra Cos: 0.9975496530532837
Inter Cos: -0.032947733998298645
Norm Quadratic Average: 43.573726654052734
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.267246246337891
Linear Weight Rank: 3667
Intra Cos: 0.9978397488594055
Inter Cos: 0.032474029809236526
Norm Quadratic Average: 27.165315628051758
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.0642848014831543
Linear Weight Rank: 10
Intra Cos: 0.9973781108856201
Inter Cos: 0.05212811753153801
Norm Quadratic Average: 17.671344757080078
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.999548077583313
Inter Cos: 0.11899814009666443
Norm Quadratic Average: 12.3702974319458
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.02056192752950592
Accuracy: 0.9966
NC1 Within Class Collapse: 0.0837421715259552
NC2 Equinorm: Features: 0.031869638711214066, Weights: 0.014697803184390068
NC2 Equiangle: Features: 0.08582366307576497, Weights: 0.062394269307454425
NC3 Self-Duality: 0.048253633081912994
NC4 NCC Mismatch: 0.0

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: 0.06832509487867355
Inter Cos: 0.07626118510961533
Norm Quadratic Average: 18.301424026489258
Nearest Class Center Accuracy: 0.8417

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10767447203397751
Inter Cos: 0.09363705664873123
Norm Quadratic Average: 12.132667541503906
Nearest Class Center Accuracy: 0.8865

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10970985889434814
Inter Cos: 0.09492873400449753
Norm Quadratic Average: 12.360898971557617
Nearest Class Center Accuracy: 0.8946

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1858607828617096
Inter Cos: 0.12039592117071152
Norm Quadratic Average: 8.656877517700195
Nearest Class Center Accuracy: 0.9446

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21513374149799347
Inter Cos: 0.12076627463102341
Norm Quadratic Average: 8.608030319213867
Nearest Class Center Accuracy: 0.9638

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2496914118528366
Inter Cos: 0.11628728359937668
Norm Quadratic Average: 8.853886604309082
Nearest Class Center Accuracy: 0.973

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28251445293426514
Inter Cos: 0.11321946978569031
Norm Quadratic Average: 9.231668472290039
Nearest Class Center Accuracy: 0.9769

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3667464852333069
Inter Cos: 0.13895486295223236
Norm Quadratic Average: 6.598006248474121
Nearest Class Center Accuracy: 0.9899

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5383507013320923
Inter Cos: 0.16747628152370453
Norm Quadratic Average: 7.193724632263184
Nearest Class Center Accuracy: 0.9932

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6653305888175964
Inter Cos: 0.12599804997444153
Norm Quadratic Average: 7.649575233459473
Nearest Class Center Accuracy: 0.9942

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.762604832649231
Inter Cos: 0.06705024093389511
Norm Quadratic Average: 7.810068607330322
Nearest Class Center Accuracy: 0.995

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8543389439582825
Inter Cos: 0.1065521240234375
Norm Quadratic Average: 6.734713077545166
Nearest Class Center Accuracy: 0.9949

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9439780712127686
Inter Cos: 0.05071819946169853
Norm Quadratic Average: 4.169539451599121
Nearest Class Center Accuracy: 0.9957

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9606810212135315
Inter Cos: 0.03600262105464935
Norm Quadratic Average: 4.219327926635742
Nearest Class Center Accuracy: 0.9961

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9735674858093262
Inter Cos: -0.008469551801681519
Norm Quadratic Average: 4.327723979949951
Nearest Class Center Accuracy: 0.9966

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.675838470458984
Linear Weight Rank: 4031
Intra Cos: 0.9817348122596741
Inter Cos: -0.03184264898300171
Norm Quadratic Average: 43.47510528564453
Nearest Class Center Accuracy: 0.9966

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.267246246337891
Linear Weight Rank: 3667
Intra Cos: 0.9825398325920105
Inter Cos: 0.0320533886551857
Norm Quadratic Average: 27.10488510131836
Nearest Class Center Accuracy: 0.9966

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.0642848014831543
Linear Weight Rank: 10
Intra Cos: 0.9823569655418396
Inter Cos: 0.05096707493066788
Norm Quadratic Average: 17.634851455688477
Nearest Class Center Accuracy: 0.9966

Output Layer:
Intra Cos: 0.9869064092636108
Inter Cos: 0.12771649658679962
Norm Quadratic Average: 12.341221809387207
Nearest Class Center Accuracy: 0.9966

