Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_100_lr_0.001_model_type_vgg11_rand_seed_338327_weight_decay_0.007.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.08946067094802856
Inter Cos: 0.11311887949705124
Norm Quadratic Average: 23.53293800354004
Nearest Class Center Accuracy: 0.81175

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11700084060430527
Inter Cos: 0.1386115849018097
Norm Quadratic Average: 43.38516616821289
Nearest Class Center Accuracy: 0.817625

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15401777625083923
Inter Cos: 0.17613571882247925
Norm Quadratic Average: 45.27957534790039
Nearest Class Center Accuracy: 0.78825

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1652432680130005
Inter Cos: 0.1914660781621933
Norm Quadratic Average: 57.48352813720703
Nearest Class Center Accuracy: 0.78925

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1838671863079071
Inter Cos: 0.19663827121257782
Norm Quadratic Average: 34.551300048828125
Nearest Class Center Accuracy: 0.83125

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20628103613853455
Inter Cos: 0.2354937493801117
Norm Quadratic Average: 28.40260124206543
Nearest Class Center Accuracy: 0.877375

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28787267208099365
Inter Cos: 0.22611434757709503
Norm Quadratic Average: 14.687422752380371
Nearest Class Center Accuracy: 0.930125

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4320122003555298
Inter Cos: 0.25577446818351746
Norm Quadratic Average: 10.16061019897461
Nearest Class Center Accuracy: 0.9685

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 72.46636199951172
Linear Weight Rank: 4031
Intra Cos: 0.6410272121429443
Inter Cos: 0.2836374342441559
Norm Quadratic Average: 44.861759185791016
Nearest Class Center Accuracy: 0.993125

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 29.311365127563477
Linear Weight Rank: 3670
Intra Cos: 0.7362248301506042
Inter Cos: 0.28351330757141113
Norm Quadratic Average: 30.202495574951172
Nearest Class Center Accuracy: 0.998625

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2100327014923096
Linear Weight Rank: 10
Intra Cos: 0.7686622738838196
Inter Cos: 0.27267661690711975
Norm Quadratic Average: 24.03200340270996
Nearest Class Center Accuracy: 0.998625

Output Layer:
Intra Cos: 0.8009852766990662
Inter Cos: 0.3359665870666504
Norm Quadratic Average: 18.18848419189453
Nearest Class Center Accuracy: 0.997375

Test Set:
Average Loss: 0.07094428086280823
Accuracy: 0.9775
NC1 Within Class Collapse: 2.5089077949523926
NC2 Equinorm: Features: 0.12346598505973816, Weights: 0.023159554228186607
NC2 Equiangle: Features: 0.2703882005479601, Weights: 0.11597988340589735
NC3 Self-Duality: 0.37453609704971313
NC4 NCC Mismatch: 0.013000000000000012

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10133808106184006
Inter Cos: 0.11957792192697525
Norm Quadratic Average: 22.82430076599121
Nearest Class Center Accuracy: 0.809

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13773715496063232
Inter Cos: 0.15668728947639465
Norm Quadratic Average: 42.106327056884766
Nearest Class Center Accuracy: 0.8115

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17133082449436188
Inter Cos: 0.20995716750621796
Norm Quadratic Average: 43.923866271972656
Nearest Class Center Accuracy: 0.792

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18452896177768707
Inter Cos: 0.23663322627544403
Norm Quadratic Average: 55.64089584350586
Nearest Class Center Accuracy: 0.789

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16397561132907867
Inter Cos: 0.23581743240356445
Norm Quadratic Average: 33.57874298095703
Nearest Class Center Accuracy: 0.8315

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18311293423175812
Inter Cos: 0.27063319087028503
Norm Quadratic Average: 27.673931121826172
Nearest Class Center Accuracy: 0.875

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25588899850845337
Inter Cos: 0.25528469681739807
Norm Quadratic Average: 14.291586875915527
Nearest Class Center Accuracy: 0.923

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3830120861530304
Inter Cos: 0.29129868745803833
Norm Quadratic Average: 9.836614608764648
Nearest Class Center Accuracy: 0.9525

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 72.46636199951172
Linear Weight Rank: 4031
Intra Cos: 0.5672886967658997
Inter Cos: 0.30548515915870667
Norm Quadratic Average: 43.12675094604492
Nearest Class Center Accuracy: 0.9675

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 29.311365127563477
Linear Weight Rank: 3670
Intra Cos: 0.6542139053344727
Inter Cos: 0.2898380756378174
Norm Quadratic Average: 29.00043296813965
Nearest Class Center Accuracy: 0.975

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2100327014923096
Linear Weight Rank: 10
Intra Cos: 0.6824566721916199
Inter Cos: 0.25697779655456543
Norm Quadratic Average: 23.095165252685547
Nearest Class Center Accuracy: 0.9745

Output Layer:
Intra Cos: 0.7027359008789062
Inter Cos: 0.3110101521015167
Norm Quadratic Average: 17.450082778930664
Nearest Class Center Accuracy: 0.9725

