Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_265358_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: 0.09116753190755844
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567670822143555
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11621850728988647
Inter Cos: 0.13540621101856232
Norm Quadratic Average: 37.943580627441406
Nearest Class Center Accuracy: 0.8234666666666667

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18882541358470917
Inter Cos: 0.16956377029418945
Norm Quadratic Average: 36.8466796875
Nearest Class Center Accuracy: 0.87085

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22431638836860657
Inter Cos: 0.18299640715122223
Norm Quadratic Average: 34.4753303527832
Nearest Class Center Accuracy: 0.90875

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2504141628742218
Inter Cos: 0.17843037843704224
Norm Quadratic Average: 15.647616386413574
Nearest Class Center Accuracy: 0.9539833333333333

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.40459126234054565
Inter Cos: 0.23563888669013977
Norm Quadratic Average: 9.708271026611328
Nearest Class Center Accuracy: 0.9791

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5564334392547607
Inter Cos: 0.3001367747783661
Norm Quadratic Average: 5.625764846801758
Nearest Class Center Accuracy: 0.9940166666666667

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7794695496559143
Inter Cos: 0.3576398193836212
Norm Quadratic Average: 5.056631088256836
Nearest Class Center Accuracy: 0.9986

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.770201683044434
Linear Weight Rank: 4031
Intra Cos: 0.8820497989654541
Inter Cos: 0.2776598632335663
Norm Quadratic Average: 26.909923553466797
Nearest Class Center Accuracy: 0.9992833333333333

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.457459926605225
Linear Weight Rank: 3667
Intra Cos: 0.9189810752868652
Inter Cos: 0.23654700815677643
Norm Quadratic Average: 23.321365356445312
Nearest Class Center Accuracy: 0.9996833333333334

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.6292433738708496
Linear Weight Rank: 10
Intra Cos: 0.9298158288002014
Inter Cos: 0.1900022178888321
Norm Quadratic Average: 20.136516571044922
Nearest Class Center Accuracy: 0.9998

Output Layer:
Intra Cos: 0.9513828158378601
Inter Cos: 0.32585209608078003
Norm Quadratic Average: 19.365903854370117
Nearest Class Center Accuracy: 0.9999666666666667

Test Set:
Average Loss: 0.020828593245428056
Accuracy: 0.9943
NC1 Within Class Collapse: 0.5685235261917114
NC2 Equinorm: Features: 0.09330781549215317, Weights: 0.031101947650313377
NC2 Equiangle: Features: 0.2031532499525282, Weights: 0.1017448001437717
NC3 Self-Duality: 0.10751736909151077
NC4 NCC Mismatch: 0.0032999999999999696

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, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12918327748775482
Inter Cos: 0.14868633449077606
Norm Quadratic Average: 37.896575927734375
Nearest Class Center Accuracy: 0.8367

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20266415178775787
Inter Cos: 0.17726896703243256
Norm Quadratic Average: 36.72930908203125
Nearest Class Center Accuracy: 0.8831

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23795545101165771
Inter Cos: 0.19471636414527893
Norm Quadratic Average: 34.39740753173828
Nearest Class Center Accuracy: 0.9184

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2620130479335785
Inter Cos: 0.19378602504730225
Norm Quadratic Average: 15.61406135559082
Nearest Class Center Accuracy: 0.9606

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.420783132314682
Inter Cos: 0.25128158926963806
Norm Quadratic Average: 9.707243919372559
Nearest Class Center Accuracy: 0.9816

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5617061257362366
Inter Cos: 0.31840336322784424
Norm Quadratic Average: 5.651486396789551
Nearest Class Center Accuracy: 0.989

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7819042205810547
Inter Cos: 0.37812766432762146
Norm Quadratic Average: 5.095012187957764
Nearest Class Center Accuracy: 0.9917

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.770201683044434
Linear Weight Rank: 4031
Intra Cos: 0.8788821697235107
Inter Cos: 0.296053409576416
Norm Quadratic Average: 27.102718353271484
Nearest Class Center Accuracy: 0.9922

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.457459926605225
Linear Weight Rank: 3667
Intra Cos: 0.9145455360412598
Inter Cos: 0.25547894835472107
Norm Quadratic Average: 23.48528289794922
Nearest Class Center Accuracy: 0.9926

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.6292433738708496
Linear Weight Rank: 10
Intra Cos: 0.9243181943893433
Inter Cos: 0.20858147740364075
Norm Quadratic Average: 20.2768611907959
Nearest Class Center Accuracy: 0.9923

Output Layer:
Intra Cos: 0.9421652555465698
Inter Cos: 0.3273356556892395
Norm Quadratic Average: 19.504621505737305
Nearest Class Center Accuracy: 0.9933

