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_0.0001.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: 0.07402542233467102
Inter Cos: 0.09163127839565277
Norm Quadratic Average: 100.90279388427734
Nearest Class Center Accuracy: 0.82275

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11497285962104797
Inter Cos: 0.11558156460523605
Norm Quadratic Average: 54.915592193603516
Nearest Class Center Accuracy: 0.8640666666666666

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11589857935905457
Inter Cos: 0.11438395082950592
Norm Quadratic Average: 62.584598541259766
Nearest Class Center Accuracy: 0.8758

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1875656247138977
Inter Cos: 0.13047264516353607
Norm Quadratic Average: 37.67844009399414
Nearest Class Center Accuracy: 0.92665

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21151511371135712
Inter Cos: 0.1336275339126587
Norm Quadratic Average: 41.075531005859375
Nearest Class Center Accuracy: 0.9483666666666667

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2341974377632141
Inter Cos: 0.13340429961681366
Norm Quadratic Average: 41.962650299072266
Nearest Class Center Accuracy: 0.9615166666666667

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2615545988082886
Inter Cos: 0.13759352266788483
Norm Quadratic Average: 43.16518783569336
Nearest Class Center Accuracy: 0.9689166666666666

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.30814650654792786
Inter Cos: 0.1579035222530365
Norm Quadratic Average: 28.491037368774414
Nearest Class Center Accuracy: 0.9893333333333333

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4068129360675812
Inter Cos: 0.20559781789779663
Norm Quadratic Average: 29.584199905395508
Nearest Class Center Accuracy: 0.9956666666666667

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5203011631965637
Inter Cos: 0.2154218852519989
Norm Quadratic Average: 31.306957244873047
Nearest Class Center Accuracy: 0.9985666666666667

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6215071678161621
Inter Cos: 0.22982823848724365
Norm Quadratic Average: 32.013919830322266
Nearest Class Center Accuracy: 0.9994833333333333

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7458605170249939
Inter Cos: 0.31436970829963684
Norm Quadratic Average: 25.44106101989746
Nearest Class Center Accuracy: 0.9994

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8739487528800964
Inter Cos: 0.2690030634403229
Norm Quadratic Average: 16.73533821105957
Nearest Class Center Accuracy: 0.9998333333333334

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9175562262535095
Inter Cos: 0.22829113900661469
Norm Quadratic Average: 17.671350479125977
Nearest Class Center Accuracy: 0.9999833333333333

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9377588629722595
Inter Cos: 0.13488174974918365
Norm Quadratic Average: 18.520231246948242
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.81584930419922
Linear Weight Rank: 4031
Intra Cos: 0.9442368149757385
Inter Cos: 0.019912950694561005
Norm Quadratic Average: 123.86241149902344
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.953075408935547
Linear Weight Rank: 3671
Intra Cos: 0.9826430082321167
Inter Cos: 0.013954786583781242
Norm Quadratic Average: 65.61104583740234
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.443403482437134
Linear Weight Rank: 10
Intra Cos: 0.9826053380966187
Inter Cos: 0.030470117926597595
Norm Quadratic Average: 32.15059280395508
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9982618689537048
Inter Cos: 0.25018924474716187
Norm Quadratic Average: 19.49738121032715
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.01975919206755766
Accuracy: 0.9964
NC1 Within Class Collapse: 0.22527927160263062
NC2 Equinorm: Features: 0.057588256895542145, Weights: 0.016659477725625038
NC2 Equiangle: Features: 0.07163393232557509, Weights: 0.08223918279012045
NC3 Self-Duality: 0.5855610966682434
NC4 NCC Mismatch: 0.0006000000000000449

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: 0.08314280211925507
Inter Cos: 0.09419243037700653
Norm Quadratic Average: 100.7691650390625
Nearest Class Center Accuracy: 0.8354

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12666955590248108
Inter Cos: 0.12024274468421936
Norm Quadratic Average: 54.63124465942383
Nearest Class Center Accuracy: 0.8761

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12695400416851044
Inter Cos: 0.11848776787519455
Norm Quadratic Average: 62.24258041381836
Nearest Class Center Accuracy: 0.8863

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20242096483707428
Inter Cos: 0.14274336397647858
Norm Quadratic Average: 37.3878059387207
Nearest Class Center Accuracy: 0.9368

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2257312834262848
Inter Cos: 0.1463782787322998
Norm Quadratic Average: 40.764671325683594
Nearest Class Center Accuracy: 0.9528

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24862748384475708
Inter Cos: 0.14678294956684113
Norm Quadratic Average: 41.67967224121094
Nearest Class Center Accuracy: 0.9624

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.27591902017593384
Inter Cos: 0.151390939950943
Norm Quadratic Average: 42.91809844970703
Nearest Class Center Accuracy: 0.9704

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3195520341396332
Inter Cos: 0.16148141026496887
Norm Quadratic Average: 28.386812210083008
Nearest Class Center Accuracy: 0.9875

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.41489794850349426
Inter Cos: 0.2074425220489502
Norm Quadratic Average: 29.51511573791504
Nearest Class Center Accuracy: 0.9917

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5258172154426575
Inter Cos: 0.21503683924674988
Norm Quadratic Average: 31.249284744262695
Nearest Class Center Accuracy: 0.9936

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6242238283157349
Inter Cos: 0.2273855060338974
Norm Quadratic Average: 31.96772575378418
Nearest Class Center Accuracy: 0.9943

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7410479187965393
Inter Cos: 0.3087078034877777
Norm Quadratic Average: 25.407777786254883
Nearest Class Center Accuracy: 0.9932

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8682296872138977
Inter Cos: 0.2626171112060547
Norm Quadratic Average: 16.710851669311523
Nearest Class Center Accuracy: 0.9938

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9088388085365295
Inter Cos: 0.22264167666435242
Norm Quadratic Average: 17.639081954956055
Nearest Class Center Accuracy: 0.9945

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9260257482528687
Inter Cos: 0.12930651009082794
Norm Quadratic Average: 18.483840942382812
Nearest Class Center Accuracy: 0.9952

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.81584930419922
Linear Weight Rank: 4031
Intra Cos: 0.9276288151741028
Inter Cos: 0.023601049557328224
Norm Quadratic Average: 123.61197662353516
Nearest Class Center Accuracy: 0.9955

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.953075408935547
Linear Weight Rank: 3671
Intra Cos: 0.9637086987495422
Inter Cos: 0.009466277435421944
Norm Quadratic Average: 65.42777252197266
Nearest Class Center Accuracy: 0.9959

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.443403482437134
Linear Weight Rank: 10
Intra Cos: 0.9623847007751465
Inter Cos: 0.0331527516245842
Norm Quadratic Average: 32.07123565673828
Nearest Class Center Accuracy: 0.9962

Output Layer:
Intra Cos: 0.9872577786445618
Inter Cos: 0.25828391313552856
Norm Quadratic Average: 19.4301815032959
Nearest Class Center Accuracy: 0.9962

