Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_846264_test_samples_None_train_samples_None_weight_decay_0.0005.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: 0.06110326945781708
Inter Cos: 0.07167075574398041
Norm Quadratic Average: 19.89371109008789
Nearest Class Center Accuracy: 0.8284

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09865382313728333
Inter Cos: 0.09223314374685287
Norm Quadratic Average: 12.850698471069336
Nearest Class Center Accuracy: 0.8735666666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09882279485464096
Inter Cos: 0.09322996437549591
Norm Quadratic Average: 12.861248016357422
Nearest Class Center Accuracy: 0.8821

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17268772423267365
Inter Cos: 0.12084238976240158
Norm Quadratic Average: 8.223920822143555
Nearest Class Center Accuracy: 0.9321666666666667

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2051556557416916
Inter Cos: 0.1297987997531891
Norm Quadratic Average: 8.583121299743652
Nearest Class Center Accuracy: 0.95485

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23818351328372955
Inter Cos: 0.12550793588161469
Norm Quadratic Average: 8.865586280822754
Nearest Class Center Accuracy: 0.9694333333333334

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2683124840259552
Inter Cos: 0.1212194561958313
Norm Quadratic Average: 9.210695266723633
Nearest Class Center Accuracy: 0.97785

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3430718779563904
Inter Cos: 0.11300507932901382
Norm Quadratic Average: 6.556027889251709
Nearest Class Center Accuracy: 0.9943166666666666

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5174081325531006
Inter Cos: 0.14646314084529877
Norm Quadratic Average: 7.080638885498047
Nearest Class Center Accuracy: 0.9986

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.642260730266571
Inter Cos: 0.11065704375505447
Norm Quadratic Average: 7.561836242675781
Nearest Class Center Accuracy: 0.9996166666666667

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7384564280509949
Inter Cos: 0.08169500529766083
Norm Quadratic Average: 7.804131984710693
Nearest Class Center Accuracy: 0.99995

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8447044491767883
Inter Cos: 0.14075753092765808
Norm Quadratic Average: 6.745450019836426
Nearest Class Center Accuracy: 0.9998833333333333

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9560755491256714
Inter Cos: 0.029615351930260658
Norm Quadratic Average: 4.161795616149902
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9729635119438171
Inter Cos: 0.020941752940416336
Norm Quadratic Average: 4.211606502532959
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9869304895401001
Inter Cos: -0.03849390521645546
Norm Quadratic Average: 4.308701992034912
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.67750072479248
Linear Weight Rank: 4031
Intra Cos: 0.9975607991218567
Inter Cos: -0.035268329083919525
Norm Quadratic Average: 43.55943298339844
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.265104293823242
Linear Weight Rank: 3667
Intra Cos: 0.998341977596283
Inter Cos: 0.028062153607606888
Norm Quadratic Average: 27.24095344543457
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.063333511352539
Linear Weight Rank: 10
Intra Cos: 0.9980096817016602
Inter Cos: 0.05740538239479065
Norm Quadratic Average: 17.783292770385742
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.999647855758667
Inter Cos: 0.15330184996128082
Norm Quadratic Average: 12.57651424407959
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.02008619699467672
Accuracy: 0.9964
NC1 Within Class Collapse: 0.08561104536056519
NC2 Equinorm: Features: 0.020716380327939987, Weights: 0.014756742864847183
NC2 Equiangle: Features: 0.11058177947998046, Weights: 0.07352601687113444
NC3 Self-Duality: 0.049229130148887634
NC4 NCC Mismatch: 9.999999999998899e-05

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: 0.06894367188215256
Inter Cos: 0.07300034165382385
Norm Quadratic Average: 19.826662063598633
Nearest Class Center Accuracy: 0.839

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10862953215837479
Inter Cos: 0.09305307269096375
Norm Quadratic Average: 12.75386905670166
Nearest Class Center Accuracy: 0.8836

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10881180316209793
Inter Cos: 0.09466202557086945
Norm Quadratic Average: 12.779824256896973
Nearest Class Center Accuracy: 0.8906

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18584570288658142
Inter Cos: 0.12352095544338226
Norm Quadratic Average: 8.170811653137207
Nearest Class Center Accuracy: 0.9394

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22041526436805725
Inter Cos: 0.13747864961624146
Norm Quadratic Average: 8.535462379455566
Nearest Class Center Accuracy: 0.9573

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25406280159950256
Inter Cos: 0.12938307225704193
Norm Quadratic Average: 8.825281143188477
Nearest Class Center Accuracy: 0.9699

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28440219163894653
Inter Cos: 0.13334308564662933
Norm Quadratic Average: 9.177007675170898
Nearest Class Center Accuracy: 0.9764

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3559674620628357
Inter Cos: 0.12964719533920288
Norm Quadratic Average: 6.5361504554748535
Nearest Class Center Accuracy: 0.9911

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5268173813819885
Inter Cos: 0.1609453707933426
Norm Quadratic Average: 7.072624206542969
Nearest Class Center Accuracy: 0.9933

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6489690542221069
Inter Cos: 0.12359021604061127
Norm Quadratic Average: 7.563188552856445
Nearest Class Center Accuracy: 0.9944

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7418248057365417
Inter Cos: 0.08459383994340897
Norm Quadratic Average: 7.810988426208496
Nearest Class Center Accuracy: 0.9947

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8421347737312317
Inter Cos: 0.14343485236167908
Norm Quadratic Average: 6.754451751708984
Nearest Class Center Accuracy: 0.9947

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9455950856208801
Inter Cos: 0.03218666464090347
Norm Quadratic Average: 4.1657304763793945
Nearest Class Center Accuracy: 0.9951

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9595307111740112
Inter Cos: 0.023151731118559837
Norm Quadratic Average: 4.2104315757751465
Nearest Class Center Accuracy: 0.9963

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9712120890617371
Inter Cos: -0.03444833308458328
Norm Quadratic Average: 4.302189350128174
Nearest Class Center Accuracy: 0.9964

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.67750072479248
Linear Weight Rank: 4031
Intra Cos: 0.9790233969688416
Inter Cos: -0.03353223204612732
Norm Quadratic Average: 43.463104248046875
Nearest Class Center Accuracy: 0.9963

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.265104293823242
Linear Weight Rank: 3667
Intra Cos: 0.9802868962287903
Inter Cos: 0.028043970465660095
Norm Quadratic Average: 27.177867889404297
Nearest Class Center Accuracy: 0.9964

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.063333511352539
Linear Weight Rank: 10
Intra Cos: 0.9804168939590454
Inter Cos: 0.06380690634250641
Norm Quadratic Average: 17.745365142822266
Nearest Class Center Accuracy: 0.9963

Output Layer:
Intra Cos: 0.9852278232574463
Inter Cos: 0.1586102694272995
Norm Quadratic Average: 12.5474853515625
Nearest Class Center Accuracy: 0.9963

