Model save path: ./New_Models/bn_True_dataset_CIFAR10_epochs_200_lr_0.01_model_type_vgg11_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.0001.pth.tar
Training Set:
Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0198909230530262
Inter Cos: 0.10477276891469955
Norm Quadratic Average: 27.59718894958496
Nearest Class Center Accuracy: 0.26968

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.019148878753185272
Inter Cos: 0.07630296796560287
Norm Quadratic Average: 69.82423400878906
Nearest Class Center Accuracy: 0.40138

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.019947310909628868
Inter Cos: 0.06000898778438568
Norm Quadratic Average: 39.960044860839844
Nearest Class Center Accuracy: 0.52622

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.01676914468407631
Inter Cos: 0.04826855659484863
Norm Quadratic Average: 42.82469177246094
Nearest Class Center Accuracy: 0.60002

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.023408902809023857
Inter Cos: 0.043469369411468506
Norm Quadratic Average: 29.561344146728516
Nearest Class Center Accuracy: 0.70636

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.038036346435546875
Inter Cos: 0.046015094965696335
Norm Quadratic Average: 33.6662712097168
Nearest Class Center Accuracy: 0.79318

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11976274847984314
Inter Cos: 0.10768713057041168
Norm Quadratic Average: 24.14811897277832
Nearest Class Center Accuracy: 0.91946

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.39867207407951355
Inter Cos: 0.22519275546073914
Norm Quadratic Average: 19.190231323242188
Nearest Class Center Accuracy: 0.98268

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 73.6056900024414
Linear Weight Rank: 4031
Intra Cos: 0.7079725861549377
Inter Cos: 0.4465778172016144
Norm Quadratic Average: 104.73748016357422
Nearest Class Center Accuracy: 0.9613

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 30.561424255371094
Linear Weight Rank: 3668
Intra Cos: 0.9451073408126831
Inter Cos: 0.0953545868396759
Norm Quadratic Average: 71.8122787475586
Nearest Class Center Accuracy: 0.99244

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 4.0504889488220215
Linear Weight Rank: 10
Intra Cos: 0.9329882264137268
Inter Cos: 0.020288635045289993
Norm Quadratic Average: 31.40041732788086
Nearest Class Center Accuracy: 0.99896

Output Layer:
Intra Cos: 0.9841921329498291
Inter Cos: 0.2866480052471161
Norm Quadratic Average: 22.549448013305664
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 1.0649251384735108
Accuracy: 0.8359
NC1 Within Class Collapse: 5.273629188537598
NC2 Equinorm: Features: 0.30160850286483765, Weights: 0.012969319708645344
NC2 Equiangle: Features: 0.08368414772881402, Weights: 0.1281678729587131
NC3 Self-Duality: 0.9691553711891174
NC4 NCC Mismatch: 0.1069

Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.018526077270507812
Inter Cos: 0.1042705774307251
Norm Quadratic Average: 27.550140380859375
Nearest Class Center Accuracy: 0.284

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.017977433279156685
Inter Cos: 0.0780535340309143
Norm Quadratic Average: 69.77174377441406
Nearest Class Center Accuracy: 0.4176

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.018576988950371742
Inter Cos: 0.06130580976605415
Norm Quadratic Average: 39.96678161621094
Nearest Class Center Accuracy: 0.5341

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0154429255053401
Inter Cos: 0.04947800934314728
Norm Quadratic Average: 42.85117721557617
Nearest Class Center Accuracy: 0.6047

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.020346228033304214
Inter Cos: 0.04489373415708542
Norm Quadratic Average: 29.57697296142578
Nearest Class Center Accuracy: 0.6779

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.030476201325654984
Inter Cos: 0.048819806426763535
Norm Quadratic Average: 33.61954116821289
Nearest Class Center Accuracy: 0.7282

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.08517313003540039
Inter Cos: 0.11276935786008835
Norm Quadratic Average: 23.998987197875977
Nearest Class Center Accuracy: 0.7741

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.29987168312072754
Inter Cos: 0.2417495846748352
Norm Quadratic Average: 18.739788055419922
Nearest Class Center Accuracy: 0.8061

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 73.6056900024414
Linear Weight Rank: 4031
Intra Cos: 0.5341451168060303
Inter Cos: 0.5590664744377136
Norm Quadratic Average: 101.29816436767578
Nearest Class Center Accuracy: 0.7716

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 30.561424255371094
Linear Weight Rank: 3668
Intra Cos: 0.5939735770225525
Inter Cos: 0.32205522060394287
Norm Quadratic Average: 65.43973541259766
Nearest Class Center Accuracy: 0.7801

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 4.0504889488220215
Linear Weight Rank: 10
Intra Cos: 0.530587375164032
Inter Cos: 0.24458323419094086
Norm Quadratic Average: 28.68707847595215
Nearest Class Center Accuracy: 0.7923

Output Layer:
Intra Cos: 0.5780782699584961
Inter Cos: 0.3526311218738556
Norm Quadratic Average: 20.1554012298584
Nearest Class Center Accuracy: 0.831

