Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.0003.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09116754680871964
Inter Cos: 0.10967149585485458
Norm Quadratic Average: 23.567672729492188
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09236925840377808
Inter Cos: 0.11156546324491501
Norm Quadratic Average: 60.30040740966797
Nearest Class Center Accuracy: 0.8125666666666667

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12383198738098145
Inter Cos: 0.14196562767028809
Norm Quadratic Average: 69.45153045654297
Nearest Class Center Accuracy: 0.8334333333333334

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13382911682128906
Inter Cos: 0.15242405235767365
Norm Quadratic Average: 108.41341400146484
Nearest Class Center Accuracy: 0.8423166666666667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2108486294746399
Inter Cos: 0.17420585453510284
Norm Quadratic Average: 85.29327392578125
Nearest Class Center Accuracy: 0.8974166666666666

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24079281091690063
Inter Cos: 0.17626847326755524
Norm Quadratic Average: 81.0411148071289
Nearest Class Center Accuracy: 0.9266

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26830101013183594
Inter Cos: 0.17549705505371094
Norm Quadratic Average: 68.4616470336914
Nearest Class Center Accuracy: 0.9457

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.30043017864227295
Inter Cos: 0.182584747672081
Norm Quadratic Average: 52.71028137207031
Nearest Class Center Accuracy: 0.9559833333333333

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3442196547985077
Inter Cos: 0.17645122110843658
Norm Quadratic Average: 22.457551956176758
Nearest Class Center Accuracy: 0.9760666666666666

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.48290419578552246
Inter Cos: 0.23604939877986908
Norm Quadratic Average: 15.506503105163574
Nearest Class Center Accuracy: 0.9876333333333334

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5693138241767883
Inter Cos: 0.2873917818069458
Norm Quadratic Average: 13.992860794067383
Nearest Class Center Accuracy: 0.99185

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6190354228019714
Inter Cos: 0.30849963426589966
Norm Quadratic Average: 13.679794311523438
Nearest Class Center Accuracy: 0.9948333333333333

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6623851656913757
Inter Cos: 0.29093804955482483
Norm Quadratic Average: 8.699071884155273
Nearest Class Center Accuracy: 0.9954833333333334

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8552830219268799
Inter Cos: 0.3193856477737427
Norm Quadratic Average: 7.2889556884765625
Nearest Class Center Accuracy: 0.9969166666666667

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9055655002593994
Inter Cos: 0.29942581057548523
Norm Quadratic Average: 6.872213840484619
Nearest Class Center Accuracy: 0.9972

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9216735363006592
Inter Cos: 0.28052759170532227
Norm Quadratic Average: 6.40712308883667
Nearest Class Center Accuracy: 0.9975333333333334

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.694623947143555
Linear Weight Rank: 4031
Intra Cos: 0.9296979308128357
Inter Cos: 0.2673334777355194
Norm Quadratic Average: 36.30080795288086
Nearest Class Center Accuracy: 0.9978833333333333

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 13.013982772827148
Linear Weight Rank: 3669
Intra Cos: 0.935504138469696
Inter Cos: 0.29793983697891235
Norm Quadratic Average: 31.000154495239258
Nearest Class Center Accuracy: 0.9983166666666666

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.5119707584381104
Linear Weight Rank: 10
Intra Cos: 0.9334027767181396
Inter Cos: 0.2907593846321106
Norm Quadratic Average: 27.019014358520508
Nearest Class Center Accuracy: 0.9988333333333334

Output Layer:
Intra Cos: 0.957693338394165
Inter Cos: 0.322426974773407
Norm Quadratic Average: 25.64775848388672
Nearest Class Center Accuracy: 0.9998666666666667

Test Set:
Average Loss: 0.021988058009107043
Accuracy: 0.9943
NC1 Within Class Collapse: 0.6696275472640991
NC2 Equinorm: Features: 0.11560782045125961, Weights: 0.04786539822816849
NC2 Equiangle: Features: 0.23901871575249567, Weights: 0.14494301478068033
NC3 Self-Duality: 0.18144887685775757
NC4 NCC Mismatch: 0.0040000000000000036

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219583660364151
Inter Cos: 0.12048852443695068
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: 0.1035764291882515
Inter Cos: 0.12304376065731049
Norm Quadratic Average: 60.40471267700195
Nearest Class Center Accuracy: 0.8247

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1377338021993637
Inter Cos: 0.15554562211036682
Norm Quadratic Average: 69.35362243652344
Nearest Class Center Accuracy: 0.8465

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14781659841537476
Inter Cos: 0.16683299839496613
Norm Quadratic Average: 108.3843002319336
Nearest Class Center Accuracy: 0.856

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2286633402109146
Inter Cos: 0.18913303315639496
Norm Quadratic Average: 85.24669647216797
Nearest Class Center Accuracy: 0.9076

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25929906964302063
Inter Cos: 0.19135719537734985
Norm Quadratic Average: 81.05033111572266
Nearest Class Center Accuracy: 0.9358

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28516316413879395
Inter Cos: 0.191043883562088
Norm Quadratic Average: 68.52665710449219
Nearest Class Center Accuracy: 0.9543

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3161639869213104
Inter Cos: 0.19901061058044434
Norm Quadratic Average: 52.80830383300781
Nearest Class Center Accuracy: 0.9611

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.35729944705963135
Inter Cos: 0.19271306693553925
Norm Quadratic Average: 22.52289390563965
Nearest Class Center Accuracy: 0.9769

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.490494042634964
Inter Cos: 0.25330546498298645
Norm Quadratic Average: 15.57105827331543
Nearest Class Center Accuracy: 0.9859

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5714967250823975
Inter Cos: 0.3069327473640442
Norm Quadratic Average: 14.073250770568848
Nearest Class Center Accuracy: 0.9876

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6191413998603821
Inter Cos: 0.3269560635089874
Norm Quadratic Average: 13.774152755737305
Nearest Class Center Accuracy: 0.99

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6571345925331116
Inter Cos: 0.30404436588287354
Norm Quadratic Average: 8.764561653137207
Nearest Class Center Accuracy: 0.9895

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8515686988830566
Inter Cos: 0.33353161811828613
Norm Quadratic Average: 7.351663112640381
Nearest Class Center Accuracy: 0.9905

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9020317196846008
Inter Cos: 0.3139157295227051
Norm Quadratic Average: 6.932401657104492
Nearest Class Center Accuracy: 0.9905

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9179300665855408
Inter Cos: 0.29067131876945496
Norm Quadratic Average: 6.462474346160889
Nearest Class Center Accuracy: 0.9907

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.694623947143555
Linear Weight Rank: 4031
Intra Cos: 0.9254236221313477
Inter Cos: 0.27103468775749207
Norm Quadratic Average: 36.616844177246094
Nearest Class Center Accuracy: 0.9915

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 13.013982772827148
Linear Weight Rank: 3669
Intra Cos: 0.9360214471817017
Inter Cos: 0.3005170226097107
Norm Quadratic Average: 31.2702579498291
Nearest Class Center Accuracy: 0.9921

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.5119707584381104
Linear Weight Rank: 10
Intra Cos: 0.9351971745491028
Inter Cos: 0.2932262420654297
Norm Quadratic Average: 27.251041412353516
Nearest Class Center Accuracy: 0.9924

Output Layer:
Intra Cos: 0.9539048671722412
Inter Cos: 0.3261803686618805
Norm Quadratic Average: 25.859357833862305
Nearest Class Center Accuracy: 0.9934

