Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.01.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: 0.05803171172738075
Inter Cos: 0.07837893813848495
Norm Quadratic Average: 2.3331050872802734
Nearest Class Center Accuracy: 0.8068333333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10221390426158905
Inter Cos: 0.0997990146279335
Norm Quadratic Average: 1.3396003246307373
Nearest Class Center Accuracy: 0.87215

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09682179987430573
Inter Cos: 0.09651745110750198
Norm Quadratic Average: 1.059615969657898
Nearest Class Center Accuracy: 0.8764333333333333

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1762300282716751
Inter Cos: 0.12054862827062607
Norm Quadratic Average: 0.6546478867530823
Nearest Class Center Accuracy: 0.9369333333333333

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23330913484096527
Inter Cos: 0.13388481736183167
Norm Quadratic Average: 0.47032615542411804
Nearest Class Center Accuracy: 0.9622666666666667

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3251086473464966
Inter Cos: 0.15147480368614197
Norm Quadratic Average: 0.39364093542099
Nearest Class Center Accuracy: 0.9740833333333333

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.35706132650375366
Inter Cos: 0.15938739478588104
Norm Quadratic Average: 0.3434637784957886
Nearest Class Center Accuracy: 0.9772666666666666

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.42178937792778015
Inter Cos: 0.16357974708080292
Norm Quadratic Average: 0.200532928109169
Nearest Class Center Accuracy: 0.9920833333333333

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6542782187461853
Inter Cos: 0.24271154403686523
Norm Quadratic Average: 0.1368781328201294
Nearest Class Center Accuracy: 0.9976666666666667

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7949520945549011
Inter Cos: 0.28184452652931213
Norm Quadratic Average: 0.12911802530288696
Nearest Class Center Accuracy: 0.9994166666666666

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8494788408279419
Inter Cos: 0.2034587860107422
Norm Quadratic Average: 0.13780446350574493
Nearest Class Center Accuracy: 0.99995

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8802866339683533
Inter Cos: 0.18629054725170135
Norm Quadratic Average: 0.12374788522720337
Nearest Class Center Accuracy: 0.9999333333333333

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9890724420547485
Inter Cos: 0.1753227859735489
Norm Quadratic Average: 0.17187021672725677
Nearest Class Center Accuracy: 0.9999666666666667

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9971254467964172
Inter Cos: 0.17438368499279022
Norm Quadratic Average: 0.3920743465423584
Nearest Class Center Accuracy: 0.9999666666666667

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9986802935600281
Inter Cos: 0.21427461504936218
Norm Quadratic Average: 0.9827470183372498
Nearest Class Center Accuracy: 0.9999666666666667

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.0326988697052
Linear Weight Rank: 8
Intra Cos: 0.9992434978485107
Inter Cos: 0.3141254484653473
Norm Quadratic Average: 23.451812744140625
Nearest Class Center Accuracy: 0.9999666666666667

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.0339391231536865
Linear Weight Rank: 1421
Intra Cos: 0.9993047714233398
Inter Cos: 0.2740800678730011
Norm Quadratic Average: 16.640405654907227
Nearest Class Center Accuracy: 0.9999666666666667

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.0348875522613525
Linear Weight Rank: 8
Intra Cos: 0.9993272423744202
Inter Cos: 0.21707864105701447
Norm Quadratic Average: 12.090373992919922
Nearest Class Center Accuracy: 0.9999666666666667

Output Layer:
Intra Cos: 0.9994526505470276
Inter Cos: 0.26724010705947876
Norm Quadratic Average: 9.45997142791748
Nearest Class Center Accuracy: 0.9999666666666667

Test Set:
Average Loss: 0.0216666271045804
Accuracy: 0.9957
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.02875320427119732, Weights: 0.007801148574799299
NC2 Equiangle: Features: 0.16788479487101238, Weights: 0.1624747806125217
NC3 Self-Duality: 0.03729500621557236
NC4 NCC Mismatch: 0.00039999999999995595

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: 0.06612785160541534
Inter Cos: 0.08104605972766876
Norm Quadratic Average: 2.3246450424194336
Nearest Class Center Accuracy: 0.8169

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11168550699949265
Inter Cos: 0.10158475488424301
Norm Quadratic Average: 1.3316665887832642
Nearest Class Center Accuracy: 0.8827

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10653675347566605
Inter Cos: 0.0989380031824112
Norm Quadratic Average: 1.0573567152023315
Nearest Class Center Accuracy: 0.8864

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19015657901763916
Inter Cos: 0.11910125613212585
Norm Quadratic Average: 0.6526490449905396
Nearest Class Center Accuracy: 0.9426

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25059419870376587
Inter Cos: 0.1311304271221161
Norm Quadratic Average: 0.46993935108184814
Nearest Class Center Accuracy: 0.9628

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3420044481754303
Inter Cos: 0.16205716133117676
Norm Quadratic Average: 0.3936409652233124
Nearest Class Center Accuracy: 0.9739

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.37295690178871155
Inter Cos: 0.17239262163639069
Norm Quadratic Average: 0.34313181042671204
Nearest Class Center Accuracy: 0.9756

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.43604618310928345
Inter Cos: 0.17309637367725372
Norm Quadratic Average: 0.20041179656982422
Nearest Class Center Accuracy: 0.9882

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6651023030281067
Inter Cos: 0.2523651719093323
Norm Quadratic Average: 0.13697032630443573
Nearest Class Center Accuracy: 0.9913

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7973200082778931
Inter Cos: 0.29010602831840515
Norm Quadratic Average: 0.1292274296283722
Nearest Class Center Accuracy: 0.994

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8501442670822144
Inter Cos: 0.21095915138721466
Norm Quadratic Average: 0.13778997957706451
Nearest Class Center Accuracy: 0.9949

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8773967623710632
Inter Cos: 0.17613252997398376
Norm Quadratic Average: 0.12346910685300827
Nearest Class Center Accuracy: 0.9945

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9710678458213806
Inter Cos: 0.17956231534481049
Norm Quadratic Average: 0.1714652180671692
Nearest Class Center Accuracy: 0.9956

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9756441712379456
Inter Cos: 0.1782296895980835
Norm Quadratic Average: 0.39106258749961853
Nearest Class Center Accuracy: 0.9956

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9768374562263489
Inter Cos: 0.21800701320171356
Norm Quadratic Average: 0.9799144864082336
Nearest Class Center Accuracy: 0.9957

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.0326988697052
Linear Weight Rank: 8
Intra Cos: 0.9793042540550232
Inter Cos: 0.3157728612422943
Norm Quadratic Average: 23.381160736083984
Nearest Class Center Accuracy: 0.9957

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.0339391231536865
Linear Weight Rank: 1421
Intra Cos: 0.9801713228225708
Inter Cos: 0.27625447511672974
Norm Quadratic Average: 16.588586807250977
Nearest Class Center Accuracy: 0.9957

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.0348875522613525
Linear Weight Rank: 8
Intra Cos: 0.9807480573654175
Inter Cos: 0.22012972831726074
Norm Quadratic Average: 12.052136421203613
Nearest Class Center Accuracy: 0.9956

Output Layer:
Intra Cos: 0.9820606708526611
Inter Cos: 0.264775812625885
Norm Quadratic Average: 9.429844856262207
Nearest Class Center Accuracy: 0.9955

