Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_265358_test_samples_None_train_samples_None_weight_decay_0.0001.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: 0.09626980125904083
Inter Cos: 0.10563342273235321
Norm Quadratic Average: 58.313236236572266
Nearest Class Center Accuracy: 0.8452833333333334

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16678611934185028
Inter Cos: 0.13571928441524506
Norm Quadratic Average: 42.21550369262695
Nearest Class Center Accuracy: 0.8942

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19582438468933105
Inter Cos: 0.13535061478614807
Norm Quadratic Average: 41.244197845458984
Nearest Class Center Accuracy: 0.9231666666666667

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24414847791194916
Inter Cos: 0.10801681131124496
Norm Quadratic Average: 26.877368927001953
Nearest Class Center Accuracy: 0.97345

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.32042115926742554
Inter Cos: 0.09991352260112762
Norm Quadratic Average: 28.458694458007812
Nearest Class Center Accuracy: 0.98995

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4514351487159729
Inter Cos: 0.15443630516529083
Norm Quadratic Average: 22.23308563232422
Nearest Class Center Accuracy: 0.99915

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6964674592018127
Inter Cos: 0.15536239743232727
Norm Quadratic Average: 17.118492126464844
Nearest Class Center Accuracy: 0.9999833333333333

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.83421325683594
Linear Weight Rank: 4031
Intra Cos: 0.8813498616218567
Inter Cos: 0.06546583026647568
Norm Quadratic Average: 117.19253540039062
Nearest Class Center Accuracy: 1.0

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.968284606933594
Linear Weight Rank: 3671
Intra Cos: 0.97010737657547
Inter Cos: 0.008510144427418709
Norm Quadratic Average: 68.84986877441406
Nearest Class Center Accuracy: 1.0

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.4311740398406982
Linear Weight Rank: 10
Intra Cos: 0.9750367403030396
Inter Cos: 0.012643679976463318
Norm Quadratic Average: 33.974735260009766
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9969345331192017
Inter Cos: 0.18301986157894135
Norm Quadratic Average: 20.2287540435791
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.022912609499142856
Accuracy: 0.9953
NC1 Within Class Collapse: 0.34222713112831116
NC2 Equinorm: Features: 0.04050067812204361, Weights: 0.015166462399065495
NC2 Equiangle: Features: 0.05971922344631619, Weights: 0.07295502026875814
NC3 Self-Duality: 0.5742021203041077
NC4 NCC Mismatch: 0.00029999999999996696

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: 0.10599470138549805
Inter Cos: 0.10534671694040298
Norm Quadratic Average: 57.99351119995117
Nearest Class Center Accuracy: 0.8579

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17771250009536743
Inter Cos: 0.1335873007774353
Norm Quadratic Average: 41.92678451538086
Nearest Class Center Accuracy: 0.9058

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.206433966755867
Inter Cos: 0.1370789110660553
Norm Quadratic Average: 41.017433166503906
Nearest Class Center Accuracy: 0.9321

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2536500096321106
Inter Cos: 0.11810815334320068
Norm Quadratic Average: 26.768993377685547
Nearest Class Center Accuracy: 0.973

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3341831564903259
Inter Cos: 0.11332715302705765
Norm Quadratic Average: 28.383056640625
Nearest Class Center Accuracy: 0.986

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4564078152179718
Inter Cos: 0.15586204826831818
Norm Quadratic Average: 22.205392837524414
Nearest Class Center Accuracy: 0.9933

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6922305822372437
Inter Cos: 0.15517358481884003
Norm Quadratic Average: 17.11495018005371
Nearest Class Center Accuracy: 0.9944

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.83421325683594
Linear Weight Rank: 4031
Intra Cos: 0.8705571293830872
Inter Cos: 0.06380672752857208
Norm Quadratic Average: 117.25403594970703
Nearest Class Center Accuracy: 0.9945

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.968284606933594
Linear Weight Rank: 3671
Intra Cos: 0.9547289609909058
Inter Cos: 0.005425565876066685
Norm Quadratic Average: 68.84432983398438
Nearest Class Center Accuracy: 0.995

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.4311740398406982
Linear Weight Rank: 10
Intra Cos: 0.953532338142395
Inter Cos: 0.017859119921922684
Norm Quadratic Average: 33.98575973510742
Nearest Class Center Accuracy: 0.9953

Output Layer:
Intra Cos: 0.9821940660476685
Inter Cos: 0.19385746121406555
Norm Quadratic Average: 20.220178604125977
Nearest Class Center Accuracy: 0.9953

