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_0.003.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: 0.0597204864025116
Inter Cos: 0.0771913081407547
Norm Quadratic Average: 2.466273307800293
Nearest Class Center Accuracy: 0.8103833333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10879139602184296
Inter Cos: 0.10309554636478424
Norm Quadratic Average: 1.5317354202270508
Nearest Class Center Accuracy: 0.8762166666666666

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10499406605958939
Inter Cos: 0.09988801926374435
Norm Quadratic Average: 1.1649715900421143
Nearest Class Center Accuracy: 0.88475

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1802387684583664
Inter Cos: 0.1256610006093979
Norm Quadratic Average: 0.8336682319641113
Nearest Class Center Accuracy: 0.9375

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2268417626619339
Inter Cos: 0.14126156270503998
Norm Quadratic Average: 0.6503618359565735
Nearest Class Center Accuracy: 0.9623

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2896839678287506
Inter Cos: 0.14395135641098022
Norm Quadratic Average: 0.5373929738998413
Nearest Class Center Accuracy: 0.9766

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3389120101928711
Inter Cos: 0.1608222872018814
Norm Quadratic Average: 0.487438827753067
Nearest Class Center Accuracy: 0.9817166666666667

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4147242605686188
Inter Cos: 0.1488916128873825
Norm Quadratic Average: 0.35611802339553833
Nearest Class Center Accuracy: 0.9950166666666667

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6418172121047974
Inter Cos: 0.20915336906909943
Norm Quadratic Average: 0.25901854038238525
Nearest Class Center Accuracy: 0.9987333333333334

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8029016852378845
Inter Cos: 0.20098528265953064
Norm Quadratic Average: 0.22297634184360504
Nearest Class Center Accuracy: 0.9994

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.851817786693573
Inter Cos: 0.1339477151632309
Norm Quadratic Average: 0.2340785562992096
Nearest Class Center Accuracy: 0.9999166666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8947968482971191
Inter Cos: 0.13020414113998413
Norm Quadratic Average: 0.27179253101348877
Nearest Class Center Accuracy: 0.99995

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9834207892417908
Inter Cos: -0.008218156173825264
Norm Quadratic Average: 0.3141302466392517
Nearest Class Center Accuracy: 0.9999833333333333

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9966921806335449
Inter Cos: -0.010105449706315994
Norm Quadratic Average: 0.5577118992805481
Nearest Class Center Accuracy: 0.9999833333333333

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9985644221305847
Inter Cos: -0.05570821836590767
Norm Quadratic Average: 1.0903570652008057
Nearest Class Center Accuracy: 0.9999833333333333

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.197477102279663
Linear Weight Rank: 137
Intra Cos: 0.9990887641906738
Inter Cos: -0.04974906146526337
Norm Quadratic Average: 25.73447036743164
Nearest Class Center Accuracy: 0.9999833333333333

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.2041797637939453
Linear Weight Rank: 1375
Intra Cos: 0.9990807175636292
Inter Cos: 0.0013656176161020994
Norm Quadratic Average: 17.55960464477539
Nearest Class Center Accuracy: 0.9999833333333333

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2007839679718018
Linear Weight Rank: 9
Intra Cos: 0.9990963935852051
Inter Cos: 0.031044507399201393
Norm Quadratic Average: 12.302860260009766
Nearest Class Center Accuracy: 0.9999833333333333

Output Layer:
Intra Cos: 0.9992126226425171
Inter Cos: 0.060326751321554184
Norm Quadratic Average: 9.100996017456055
Nearest Class Center Accuracy: 0.9999833333333333

Test Set:
Average Loss: 0.01731609467891976
Accuracy: 0.9963
NC1 Within Class Collapse: 0.07918249070644379
NC2 Equinorm: Features: 0.01641440950334072, Weights: 0.0061377426609396935
NC2 Equiangle: Features: 0.05950178040398492, Weights: 0.02572689586215549
NC3 Self-Duality: 0.007608410902321339
NC4 NCC Mismatch: 0.00019999999999997797

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: 0.06775227189064026
Inter Cos: 0.0800115168094635
Norm Quadratic Average: 2.4561920166015625
Nearest Class Center Accuracy: 0.8199

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11920520663261414
Inter Cos: 0.10463084280490875
Norm Quadratic Average: 1.5201389789581299
Nearest Class Center Accuracy: 0.8892

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11477845162153244
Inter Cos: 0.10083195567131042
Norm Quadratic Average: 1.159696340560913
Nearest Class Center Accuracy: 0.8948

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19340811669826508
Inter Cos: 0.13708330690860748
Norm Quadratic Average: 0.829716682434082
Nearest Class Center Accuracy: 0.9429

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2411314845085144
Inter Cos: 0.1545296013355255
Norm Quadratic Average: 0.6481747627258301
Nearest Class Center Accuracy: 0.964

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3046542704105377
Inter Cos: 0.15742622315883636
Norm Quadratic Average: 0.5360972881317139
Nearest Class Center Accuracy: 0.9755

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.35487693548202515
Inter Cos: 0.17390364408493042
Norm Quadratic Average: 0.4866434931755066
Nearest Class Center Accuracy: 0.9803

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4266224205493927
Inter Cos: 0.16291333734989166
Norm Quadratic Average: 0.3555446267127991
Nearest Class Center Accuracy: 0.9911

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6519560217857361
Inter Cos: 0.22293612360954285
Norm Quadratic Average: 0.2592295706272125
Nearest Class Center Accuracy: 0.9945

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8104349374771118
Inter Cos: 0.21407407522201538
Norm Quadratic Average: 0.22332964837551117
Nearest Class Center Accuracy: 0.9952

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8547449111938477
Inter Cos: 0.14603963494300842
Norm Quadratic Average: 0.23409584164619446
Nearest Class Center Accuracy: 0.9953

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8935636878013611
Inter Cos: 0.12952448427677155
Norm Quadratic Average: 0.2714597284793854
Nearest Class Center Accuracy: 0.9957

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9735280871391296
Inter Cos: -0.014714525081217289
Norm Quadratic Average: 0.3133224546909332
Nearest Class Center Accuracy: 0.9958

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9789119958877563
Inter Cos: 0.002614883240312338
Norm Quadratic Average: 0.5561575293540955
Nearest Class Center Accuracy: 0.996

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9809525609016418
Inter Cos: -0.04883831366896629
Norm Quadratic Average: 1.087187647819519
Nearest Class Center Accuracy: 0.996

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.197477102279663
Linear Weight Rank: 137
Intra Cos: 0.9831961393356323
Inter Cos: -0.04027248173952103
Norm Quadratic Average: 25.65818977355957
Nearest Class Center Accuracy: 0.9961

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.2041797637939453
Linear Weight Rank: 1375
Intra Cos: 0.9839281439781189
Inter Cos: 0.01009540818631649
Norm Quadratic Average: 17.50797462463379
Nearest Class Center Accuracy: 0.9963

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2007839679718018
Linear Weight Rank: 9
Intra Cos: 0.9843900203704834
Inter Cos: 0.0394979864358902
Norm Quadratic Average: 12.267452239990234
Nearest Class Center Accuracy: 0.9963

Output Layer:
Intra Cos: 0.9852580428123474
Inter Cos: 0.06822207570075989
Norm Quadratic Average: 9.074878692626953
Nearest Class Center Accuracy: 0.9963

