Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_265358_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: 0.09116753190755844
Inter Cos: 0.10967151820659637
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: 0.06095251813530922
Inter Cos: 0.07655159384012222
Norm Quadratic Average: 6.371209621429443
Nearest Class Center Accuracy: 0.8200166666666666

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09872247278690338
Inter Cos: 0.09559990465641022
Norm Quadratic Average: 4.518721580505371
Nearest Class Center Accuracy: 0.87325

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09684020280838013
Inter Cos: 0.08749004453420639
Norm Quadratic Average: 3.938356637954712
Nearest Class Center Accuracy: 0.8915666666666666

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17071866989135742
Inter Cos: 0.12038345634937286
Norm Quadratic Average: 3.166593074798584
Nearest Class Center Accuracy: 0.9404333333333333

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20578992366790771
Inter Cos: 0.12986581027507782
Norm Quadratic Average: 2.2415292263031006
Nearest Class Center Accuracy: 0.9610166666666666

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2469760775566101
Inter Cos: 0.13298575580120087
Norm Quadratic Average: 2.0728952884674072
Nearest Class Center Accuracy: 0.9732166666666666

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.29088953137397766
Inter Cos: 0.12652187049388885
Norm Quadratic Average: 1.9728820323944092
Nearest Class Center Accuracy: 0.9792

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3771296739578247
Inter Cos: 0.16866987943649292
Norm Quadratic Average: 1.5338140726089478
Nearest Class Center Accuracy: 0.9935833333333334

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5774287581443787
Inter Cos: 0.16015928983688354
Norm Quadratic Average: 1.1002863645553589
Nearest Class Center Accuracy: 0.9984666666666666

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7354514002799988
Inter Cos: 0.08745989203453064
Norm Quadratic Average: 0.982104480266571
Nearest Class Center Accuracy: 0.9997333333333334

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8298086524009705
Inter Cos: 0.016033239662647247
Norm Quadratic Average: 0.8123540282249451
Nearest Class Center Accuracy: 0.9999666666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9000406265258789
Inter Cos: 0.0457000732421875
Norm Quadratic Average: 0.686141848564148
Nearest Class Center Accuracy: 0.9999833333333333

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9777294397354126
Inter Cos: -0.019132189452648163
Norm Quadratic Average: 0.5800431370735168
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9951416850090027
Inter Cos: -0.03035322017967701
Norm Quadratic Average: 0.6432859301567078
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9989551901817322
Inter Cos: -0.04984346777200699
Norm Quadratic Average: 1.056762933731079
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.090203046798706
Linear Weight Rank: 4028
Intra Cos: 0.9996830821037292
Inter Cos: -0.05572430044412613
Norm Quadratic Average: 26.38365364074707
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.475113868713379
Linear Weight Rank: 3637
Intra Cos: 0.9996460676193237
Inter Cos: -0.011057316325604916
Norm Quadratic Average: 18.693519592285156
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.298121690750122
Linear Weight Rank: 9
Intra Cos: 0.9996144771575928
Inter Cos: 0.03126811608672142
Norm Quadratic Average: 13.62085247039795
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9998737573623657
Inter Cos: 0.1000320166349411
Norm Quadratic Average: 10.688549995422363
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.022112156802369283
Accuracy: 0.996
NC1 Within Class Collapse: 0.07285124063491821
NC2 Equinorm: Features: 0.01455835159868002, Weights: 0.008347817696630955
NC2 Equiangle: Features: 0.0770268334282769, Weights: 0.05182754728529188
NC3 Self-Duality: 0.013641643337905407
NC4 NCC Mismatch: 9.999999999998899e-05

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
Inter Cos: 0.12048853933811188
Norm Quadratic Average: 23.595195770263672
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0688241645693779
Inter Cos: 0.07899397611618042
Norm Quadratic Average: 6.349233150482178
Nearest Class Center Accuracy: 0.8289

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10820164531469345
Inter Cos: 0.09740615636110306
Norm Quadratic Average: 4.487148284912109
Nearest Class Center Accuracy: 0.8855

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10591737926006317
Inter Cos: 0.08882036060094833
Norm Quadratic Average: 3.9206817150115967
Nearest Class Center Accuracy: 0.9021

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18191136419773102
Inter Cos: 0.11962238699197769
Norm Quadratic Average: 3.150552988052368
Nearest Class Center Accuracy: 0.9459

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21908049285411835
Inter Cos: 0.12797656655311584
Norm Quadratic Average: 2.2316410541534424
Nearest Class Center Accuracy: 0.9636

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26164692640304565
Inter Cos: 0.13103093206882477
Norm Quadratic Average: 2.0654916763305664
Nearest Class Center Accuracy: 0.9746

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3069833517074585
Inter Cos: 0.12324932217597961
Norm Quadratic Average: 1.9655921459197998
Nearest Class Center Accuracy: 0.9789

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3913434147834778
Inter Cos: 0.17303743958473206
Norm Quadratic Average: 1.5281908512115479
Nearest Class Center Accuracy: 0.9899

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5873638391494751
Inter Cos: 0.1588543951511383
Norm Quadratic Average: 1.0970134735107422
Nearest Class Center Accuracy: 0.9933

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7420753836631775
Inter Cos: 0.089383065700531
Norm Quadratic Average: 0.9804404377937317
Nearest Class Center Accuracy: 0.9946

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8322257399559021
Inter Cos: 0.022280283272266388
Norm Quadratic Average: 0.8114046454429626
Nearest Class Center Accuracy: 0.9954

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8959599137306213
Inter Cos: 0.04920026659965515
Norm Quadratic Average: 0.6852928400039673
Nearest Class Center Accuracy: 0.9955

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9631065130233765
Inter Cos: -0.020378757268190384
Norm Quadratic Average: 0.5788400173187256
Nearest Class Center Accuracy: 0.9958

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9767096638679504
Inter Cos: -0.030379079282283783
Norm Quadratic Average: 0.6414486765861511
Nearest Class Center Accuracy: 0.996

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9810322523117065
Inter Cos: -0.04976701736450195
Norm Quadratic Average: 1.0536785125732422
Nearest Class Center Accuracy: 0.996

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.090203046798706
Linear Weight Rank: 4028
Intra Cos: 0.9831388592720032
Inter Cos: -0.046978726983070374
Norm Quadratic Average: 26.306060791015625
Nearest Class Center Accuracy: 0.996

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.475113868713379
Linear Weight Rank: 3637
Intra Cos: 0.9835562109947205
Inter Cos: -0.004380170255899429
Norm Quadratic Average: 18.639095306396484
Nearest Class Center Accuracy: 0.996

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.298121690750122
Linear Weight Rank: 9
Intra Cos: 0.9838894605636597
Inter Cos: 0.03976033627986908
Norm Quadratic Average: 13.582531929016113
Nearest Class Center Accuracy: 0.996

Output Layer:
Intra Cos: 0.9849923253059387
Inter Cos: 0.1084817498922348
Norm Quadratic Average: 10.658431053161621
Nearest Class Center Accuracy: 0.996

