Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_314159_test_samples_None_train_samples_None_weight_decay_0.001.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: 0.10967153310775757
Norm Quadratic Average: 23.567678451538086
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11909889429807663
Inter Cos: 0.14262887835502625
Norm Quadratic Average: 40.54779052734375
Nearest Class Center Accuracy: 0.8152166666666667

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18662714958190918
Inter Cos: 0.17322658002376556
Norm Quadratic Average: 41.6810302734375
Nearest Class Center Accuracy: 0.8477833333333333

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21926428377628326
Inter Cos: 0.19053654372692108
Norm Quadratic Average: 40.836238861083984
Nearest Class Center Accuracy: 0.8921833333333333

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.237264946103096
Inter Cos: 0.18028445541858673
Norm Quadratic Average: 19.386947631835938
Nearest Class Center Accuracy: 0.93895

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.364311158657074
Inter Cos: 0.2502676844596863
Norm Quadratic Average: 11.687453269958496
Nearest Class Center Accuracy: 0.9666833333333333

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5105552077293396
Inter Cos: 0.33166220784187317
Norm Quadratic Average: 6.551181793212891
Nearest Class Center Accuracy: 0.9863833333333333

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7759956121444702
Inter Cos: 0.3662213683128357
Norm Quadratic Average: 5.722416400909424
Nearest Class Center Accuracy: 0.99525

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.1679158210754395
Linear Weight Rank: 4028
Intra Cos: 0.8542529344558716
Inter Cos: 0.3005197048187256
Norm Quadratic Average: 29.59555435180664
Nearest Class Center Accuracy: 0.9974833333333334

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.5719358921051025
Linear Weight Rank: 3637
Intra Cos: 0.9051042199134827
Inter Cos: 0.2636008858680725
Norm Quadratic Average: 25.416154861450195
Nearest Class Center Accuracy: 0.9989

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.374629497528076
Linear Weight Rank: 9
Intra Cos: 0.916205108165741
Inter Cos: 0.21345467865467072
Norm Quadratic Average: 21.31865692138672
Nearest Class Center Accuracy: 0.99935

Output Layer:
Intra Cos: 0.9431458115577698
Inter Cos: 0.30286306142807007
Norm Quadratic Average: 20.302566528320312
Nearest Class Center Accuracy: 0.9998333333333334

Test Set:
Average Loss: 0.021385505011631174
Accuracy: 0.9929
NC1 Within Class Collapse: 1.0158203840255737
NC2 Equinorm: Features: 0.13322162628173828, Weights: 0.04810921847820282
NC2 Equiangle: Features: 0.223615476820204, Weights: 0.16283067067464194
NC3 Self-Duality: 0.09814472496509552
NC4 NCC Mismatch: 0.0033999999999999586

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
Inter Cos: 0.12048852443695068
Norm Quadratic Average: 23.595197677612305
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.132613867521286
Inter Cos: 0.15639711916446686
Norm Quadratic Average: 40.53949737548828
Nearest Class Center Accuracy: 0.8301

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20264360308647156
Inter Cos: 0.18356594443321228
Norm Quadratic Average: 41.564476013183594
Nearest Class Center Accuracy: 0.8645

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23387868702411652
Inter Cos: 0.20098787546157837
Norm Quadratic Average: 40.75595474243164
Nearest Class Center Accuracy: 0.9033

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24919651448726654
Inter Cos: 0.1903301626443863
Norm Quadratic Average: 19.356449127197266
Nearest Class Center Accuracy: 0.9493

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3774570822715759
Inter Cos: 0.2670693099498749
Norm Quadratic Average: 11.699365615844727
Nearest Class Center Accuracy: 0.9702

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5200235843658447
Inter Cos: 0.3491540849208832
Norm Quadratic Average: 6.591981410980225
Nearest Class Center Accuracy: 0.9837

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7811129689216614
Inter Cos: 0.3857574164867401
Norm Quadratic Average: 5.7808356285095215
Nearest Class Center Accuracy: 0.9889

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.1679158210754395
Linear Weight Rank: 4028
Intra Cos: 0.8549505472183228
Inter Cos: 0.3173007369041443
Norm Quadratic Average: 29.88580322265625
Nearest Class Center Accuracy: 0.9902

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.5719358921051025
Linear Weight Rank: 3637
Intra Cos: 0.9032162427902222
Inter Cos: 0.28037041425704956
Norm Quadratic Average: 25.65976905822754
Nearest Class Center Accuracy: 0.9917

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.374629497528076
Linear Weight Rank: 9
Intra Cos: 0.912996232509613
Inter Cos: 0.2276441901922226
Norm Quadratic Average: 21.521371841430664
Nearest Class Center Accuracy: 0.9915

Output Layer:
Intra Cos: 0.9349298477172852
Inter Cos: 0.30122464895248413
Norm Quadratic Average: 20.493728637695312
Nearest Class Center Accuracy: 0.9924

