Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_265358_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: 0.10967151820659637
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: 0.0860685333609581
Inter Cos: 0.1049153283238411
Norm Quadratic Average: 63.315223693847656
Nearest Class Center Accuracy: 0.8143833333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12806133925914764
Inter Cos: 0.1333368569612503
Norm Quadratic Average: 57.16053009033203
Nearest Class Center Accuracy: 0.8565166666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1345076858997345
Inter Cos: 0.13487602770328522
Norm Quadratic Average: 73.09934997558594
Nearest Class Center Accuracy: 0.8678666666666667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21028491854667664
Inter Cos: 0.15022234618663788
Norm Quadratic Average: 48.691368103027344
Nearest Class Center Accuracy: 0.9246166666666666

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23392924666404724
Inter Cos: 0.15367628633975983
Norm Quadratic Average: 45.28816223144531
Nearest Class Center Accuracy: 0.94575

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26627615094184875
Inter Cos: 0.161407470703125
Norm Quadratic Average: 38.590675354003906
Nearest Class Center Accuracy: 0.9600166666666666

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3037489652633667
Inter Cos: 0.16297629475593567
Norm Quadratic Average: 32.4320068359375
Nearest Class Center Accuracy: 0.9685

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3751722276210785
Inter Cos: 0.20398648083209991
Norm Quadratic Average: 16.049129486083984
Nearest Class Center Accuracy: 0.9848833333333333

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.49746426939964294
Inter Cos: 0.20007798075675964
Norm Quadratic Average: 13.369058609008789
Nearest Class Center Accuracy: 0.99205

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.611974835395813
Inter Cos: 0.2303495705127716
Norm Quadratic Average: 12.561997413635254
Nearest Class Center Accuracy: 0.99505

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6938474774360657
Inter Cos: 0.24035657942295074
Norm Quadratic Average: 12.192959785461426
Nearest Class Center Accuracy: 0.99675

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7621628642082214
Inter Cos: 0.17620955407619476
Norm Quadratic Average: 8.236108779907227
Nearest Class Center Accuracy: 0.9947166666666667

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8919677734375
Inter Cos: 0.3240397572517395
Norm Quadratic Average: 7.654083251953125
Nearest Class Center Accuracy: 0.9953666666666666

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9274991750717163
Inter Cos: 0.4114223122596741
Norm Quadratic Average: 7.265028476715088
Nearest Class Center Accuracy: 0.99595

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9388539791107178
Inter Cos: 0.4437851011753082
Norm Quadratic Average: 6.791738510131836
Nearest Class Center Accuracy: 0.9965833333333334

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.79903411865234
Linear Weight Rank: 4031
Intra Cos: 0.9525099396705627
Inter Cos: 0.4143075942993164
Norm Quadratic Average: 38.38568878173828
Nearest Class Center Accuracy: 0.9976333333333334

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.919307708740234
Linear Weight Rank: 3670
Intra Cos: 0.9565091133117676
Inter Cos: 0.39786091446876526
Norm Quadratic Average: 31.582256317138672
Nearest Class Center Accuracy: 0.9986166666666667

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 3.024183511734009
Linear Weight Rank: 10
Intra Cos: 0.9564375877380371
Inter Cos: 0.3211313486099243
Norm Quadratic Average: 27.803199768066406
Nearest Class Center Accuracy: 0.9992333333333333

Output Layer:
Intra Cos: 0.9865781664848328
Inter Cos: 0.34410417079925537
Norm Quadratic Average: 27.927038192749023
Nearest Class Center Accuracy: 0.9999833333333333

Test Set:
Average Loss: 0.025731167977734413
Accuracy: 0.9944
NC1 Within Class Collapse: 0.4857279658317566
NC2 Equinorm: Features: 0.13140800595283508, Weights: 0.07489628344774246
NC2 Equiangle: Features: 0.22793087429470485, Weights: 0.13229830000135634
NC3 Self-Duality: 0.32507553696632385
NC4 NCC Mismatch: 0.0050000000000000044

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: 0.09634671360254288
Inter Cos: 0.11248159408569336
Norm Quadratic Average: 63.383056640625
Nearest Class Center Accuracy: 0.8259

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14076165854930878
Inter Cos: 0.13772094249725342
Norm Quadratic Average: 57.057220458984375
Nearest Class Center Accuracy: 0.8683

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14685076475143433
Inter Cos: 0.13927718997001648
Norm Quadratic Average: 73.02281951904297
Nearest Class Center Accuracy: 0.8796

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22542411088943481
Inter Cos: 0.1628919243812561
Norm Quadratic Average: 48.615814208984375
Nearest Class Center Accuracy: 0.9348

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24873457849025726
Inter Cos: 0.16251543164253235
Norm Quadratic Average: 45.21353530883789
Nearest Class Center Accuracy: 0.9526

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28068605065345764
Inter Cos: 0.16437599062919617
Norm Quadratic Average: 38.54354476928711
Nearest Class Center Accuracy: 0.9636

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3190934956073761
Inter Cos: 0.1659316122531891
Norm Quadratic Average: 32.40958786010742
Nearest Class Center Accuracy: 0.9709

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3876960575580597
Inter Cos: 0.20714707672595978
Norm Quadratic Average: 16.056514739990234
Nearest Class Center Accuracy: 0.9835

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5074412822723389
Inter Cos: 0.21484187245368958
Norm Quadratic Average: 13.407668113708496
Nearest Class Center Accuracy: 0.9884

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6197136044502258
Inter Cos: 0.2487809807062149
Norm Quadratic Average: 12.621695518493652
Nearest Class Center Accuracy: 0.9891

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6991980075836182
Inter Cos: 0.25914129614830017
Norm Quadratic Average: 12.264527320861816
Nearest Class Center Accuracy: 0.9898

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7637017369270325
Inter Cos: 0.18407246470451355
Norm Quadratic Average: 8.29440975189209
Nearest Class Center Accuracy: 0.9886

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.888242244720459
Inter Cos: 0.32330524921417236
Norm Quadratic Average: 7.7091264724731445
Nearest Class Center Accuracy: 0.9879

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9268257021903992
Inter Cos: 0.40957462787628174
Norm Quadratic Average: 7.314931869506836
Nearest Class Center Accuracy: 0.9882

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.937130868434906
Inter Cos: 0.4412935972213745
Norm Quadratic Average: 6.836069583892822
Nearest Class Center Accuracy: 0.9884

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.79903411865234
Linear Weight Rank: 4031
Intra Cos: 0.9485483169555664
Inter Cos: 0.41151925921440125
Norm Quadratic Average: 38.6239128112793
Nearest Class Center Accuracy: 0.9889

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.919307708740234
Linear Weight Rank: 3670
Intra Cos: 0.95135897397995
Inter Cos: 0.3954879641532898
Norm Quadratic Average: 31.780118942260742
Nearest Class Center Accuracy: 0.9907

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 3.024183511734009
Linear Weight Rank: 10
Intra Cos: 0.9502744674682617
Inter Cos: 0.32001054286956787
Norm Quadratic Average: 27.977096557617188
Nearest Class Center Accuracy: 0.9917

Output Layer:
Intra Cos: 0.9758912920951843
Inter Cos: 0.3493872880935669
Norm Quadratic Average: 28.08202362060547
Nearest Class Center Accuracy: 0.9937

