Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_846264_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: 0.10967151820659637
Norm Quadratic Average: 23.567686080932617
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0873868390917778
Inter Cos: 0.10512777417898178
Norm Quadratic Average: 55.73658752441406
Nearest Class Center Accuracy: 0.8147833333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12471285462379456
Inter Cos: 0.13181845843791962
Norm Quadratic Average: 57.37749099731445
Nearest Class Center Accuracy: 0.8448333333333333

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13818903267383575
Inter Cos: 0.1423947960138321
Norm Quadratic Average: 78.65567016601562
Nearest Class Center Accuracy: 0.8554166666666667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2170206755399704
Inter Cos: 0.15942364931106567
Norm Quadratic Average: 57.07106018066406
Nearest Class Center Accuracy: 0.9088166666666667

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23701946437358856
Inter Cos: 0.17665253579616547
Norm Quadratic Average: 50.027015686035156
Nearest Class Center Accuracy: 0.9352166666666667

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2729034721851349
Inter Cos: 0.17522968351840973
Norm Quadratic Average: 38.11199951171875
Nearest Class Center Accuracy: 0.9551833333333334

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.32082197070121765
Inter Cos: 0.17671048641204834
Norm Quadratic Average: 27.523387908935547
Nearest Class Center Accuracy: 0.96625

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4040054976940155
Inter Cos: 0.18018805980682373
Norm Quadratic Average: 12.237442970275879
Nearest Class Center Accuracy: 0.9857333333333334

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5940161943435669
Inter Cos: 0.28546765446662903
Norm Quadratic Average: 9.167288780212402
Nearest Class Center Accuracy: 0.9932166666666666

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6908858418464661
Inter Cos: 0.31785959005355835
Norm Quadratic Average: 8.960087776184082
Nearest Class Center Accuracy: 0.9960166666666667

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7257667779922485
Inter Cos: 0.3120465576648712
Norm Quadratic Average: 9.316102981567383
Nearest Class Center Accuracy: 0.9970333333333333

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7320654392242432
Inter Cos: 0.2673679292201996
Norm Quadratic Average: 6.183906078338623
Nearest Class Center Accuracy: 0.99445

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8786477446556091
Inter Cos: 0.38443267345428467
Norm Quadratic Average: 5.430243492126465
Nearest Class Center Accuracy: 0.99555

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9238241910934448
Inter Cos: 0.4227214753627777
Norm Quadratic Average: 5.327502727508545
Nearest Class Center Accuracy: 0.99645

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9399347305297852
Inter Cos: 0.4320909082889557
Norm Quadratic Average: 5.263396263122559
Nearest Class Center Accuracy: 0.9971

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.703895568847656
Linear Weight Rank: 4031
Intra Cos: 0.9487080574035645
Inter Cos: 0.4174528419971466
Norm Quadratic Average: 32.4323616027832
Nearest Class Center Accuracy: 0.998

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 13.027377128601074
Linear Weight Rank: 3670
Intra Cos: 0.9576588273048401
Inter Cos: 0.3816935420036316
Norm Quadratic Average: 28.717805862426758
Nearest Class Center Accuracy: 0.99895

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.5633864402770996
Linear Weight Rank: 10
Intra Cos: 0.9621933102607727
Inter Cos: 0.3195911943912506
Norm Quadratic Average: 25.162765502929688
Nearest Class Center Accuracy: 0.99935

Output Layer:
Intra Cos: 0.9774139523506165
Inter Cos: 0.2968858778476715
Norm Quadratic Average: 23.97992706298828
Nearest Class Center Accuracy: 0.9999666666666667

Test Set:
Average Loss: 0.023677841870329577
Accuracy: 0.9939
NC1 Within Class Collapse: 0.46905332803726196
NC2 Equinorm: Features: 0.10306178778409958, Weights: 0.03425430878996849
NC2 Equiangle: Features: 0.26939078436957464, Weights: 0.1422854847378201
NC3 Self-Duality: 0.1989700347185135
NC4 NCC Mismatch: 0.0027000000000000357

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582170248032
Inter Cos: 0.12048851698637009
Norm Quadratic Average: 23.59519386291504
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09756570309400558
Inter Cos: 0.11358928680419922
Norm Quadratic Average: 55.83282470703125
Nearest Class Center Accuracy: 0.8268

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1375548094511032
Inter Cos: 0.14228327572345734
Norm Quadratic Average: 57.31743621826172
Nearest Class Center Accuracy: 0.8588

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15121975541114807
Inter Cos: 0.1502572000026703
Norm Quadratic Average: 78.68573760986328
Nearest Class Center Accuracy: 0.8667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23575279116630554
Inter Cos: 0.1728530377149582
Norm Quadratic Average: 57.061485290527344
Nearest Class Center Accuracy: 0.9167

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2576245665550232
Inter Cos: 0.1714305430650711
Norm Quadratic Average: 50.05206298828125
Nearest Class Center Accuracy: 0.9422

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2929129898548126
Inter Cos: 0.17931042611598969
Norm Quadratic Average: 38.14738845825195
Nearest Class Center Accuracy: 0.9597

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.33947667479515076
Inter Cos: 0.19196529686450958
Norm Quadratic Average: 27.578092575073242
Nearest Class Center Accuracy: 0.9687

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.42176058888435364
Inter Cos: 0.1950024962425232
Norm Quadratic Average: 12.284123420715332
Nearest Class Center Accuracy: 0.9849

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.604423999786377
Inter Cos: 0.30407339334487915
Norm Quadratic Average: 9.218461990356445
Nearest Class Center Accuracy: 0.9895

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6981610655784607
Inter Cos: 0.3385547697544098
Norm Quadratic Average: 9.019253730773926
Nearest Class Center Accuracy: 0.9904

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7309901118278503
Inter Cos: 0.33329886198043823
Norm Quadratic Average: 9.379694938659668
Nearest Class Center Accuracy: 0.9906

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7341047525405884
Inter Cos: 0.28044357895851135
Norm Quadratic Average: 6.2281575202941895
Nearest Class Center Accuracy: 0.9876

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8756045699119568
Inter Cos: 0.37969475984573364
Norm Quadratic Average: 5.471807479858398
Nearest Class Center Accuracy: 0.9887

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9188005328178406
Inter Cos: 0.41531556844711304
Norm Quadratic Average: 5.367478847503662
Nearest Class Center Accuracy: 0.9899

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9343477487564087
Inter Cos: 0.42455217242240906
Norm Quadratic Average: 5.301273822784424
Nearest Class Center Accuracy: 0.9911

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.703895568847656
Linear Weight Rank: 4031
Intra Cos: 0.9421107172966003
Inter Cos: 0.4105182886123657
Norm Quadratic Average: 32.63736343383789
Nearest Class Center Accuracy: 0.992

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 13.027377128601074
Linear Weight Rank: 3670
Intra Cos: 0.9476817846298218
Inter Cos: 0.37524497509002686
Norm Quadratic Average: 28.896678924560547
Nearest Class Center Accuracy: 0.9931

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.5633864402770996
Linear Weight Rank: 10
Intra Cos: 0.9506561756134033
Inter Cos: 0.31343555450439453
Norm Quadratic Average: 25.318729400634766
Nearest Class Center Accuracy: 0.9934

Output Layer:
Intra Cos: 0.9656806588172913
Inter Cos: 0.30186793208122253
Norm Quadratic Average: 24.134016036987305
Nearest Class Center Accuracy: 0.9935

