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_0.0001.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: 0.107512466609478
Inter Cos: 0.12388279289007187
Norm Quadratic Average: 33.7105827331543
Nearest Class Center Accuracy: 0.8327

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1889297515153885
Inter Cos: 0.15863050520420074
Norm Quadratic Average: 28.940385818481445
Nearest Class Center Accuracy: 0.8916166666666666

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22059065103530884
Inter Cos: 0.1659311205148697
Norm Quadratic Average: 30.801597595214844
Nearest Class Center Accuracy: 0.9151

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2701380252838135
Inter Cos: 0.15281040966510773
Norm Quadratic Average: 15.89490032196045
Nearest Class Center Accuracy: 0.96395

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.363202840089798
Inter Cos: 0.1740688979625702
Norm Quadratic Average: 12.65072250366211
Nearest Class Center Accuracy: 0.9810833333333333

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5045691132545471
Inter Cos: 0.1873345971107483
Norm Quadratic Average: 6.882996082305908
Nearest Class Center Accuracy: 0.9947666666666667

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7658218145370483
Inter Cos: 0.22401690483093262
Norm Quadratic Average: 5.849661350250244
Nearest Class Center Accuracy: 0.9987666666666667

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.80171203613281
Linear Weight Rank: 4031
Intra Cos: 0.9041105508804321
Inter Cos: 0.17913345992565155
Norm Quadratic Average: 32.83787155151367
Nearest Class Center Accuracy: 0.9994166666666666

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.918193817138672
Linear Weight Rank: 3669
Intra Cos: 0.9316253662109375
Inter Cos: 0.16444794833660126
Norm Quadratic Average: 27.88699722290039
Nearest Class Center Accuracy: 0.9997666666666667

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.994596004486084
Linear Weight Rank: 10
Intra Cos: 0.9370457530021667
Inter Cos: 0.1598045974969864
Norm Quadratic Average: 25.612049102783203
Nearest Class Center Accuracy: 0.9999

Output Layer:
Intra Cos: 0.9654722809791565
Inter Cos: 0.25121697783470154
Norm Quadratic Average: 25.246137619018555
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.022352478161718318
Accuracy: 0.9947
NC1 Within Class Collapse: 0.47467944025993347
NC2 Equinorm: Features: 0.11523542553186417, Weights: 0.024717843160033226
NC2 Equiangle: Features: 0.18129215240478516, Weights: 0.0976748890346951
NC3 Self-Duality: 0.24955429136753082
NC4 NCC Mismatch: 0.0037000000000000366

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: 0.11954886466264725
Inter Cos: 0.12896017730236053
Norm Quadratic Average: 33.59818649291992
Nearest Class Center Accuracy: 0.844

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2032782882452011
Inter Cos: 0.16204753518104553
Norm Quadratic Average: 28.795682907104492
Nearest Class Center Accuracy: 0.9028

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2340194433927536
Inter Cos: 0.1794424206018448
Norm Quadratic Average: 30.671655654907227
Nearest Class Center Accuracy: 0.9255

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28216949105262756
Inter Cos: 0.16409635543823242
Norm Quadratic Average: 15.84919548034668
Nearest Class Center Accuracy: 0.9683

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3807469308376312
Inter Cos: 0.18725864589214325
Norm Quadratic Average: 12.633661270141602
Nearest Class Center Accuracy: 0.9809

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5156227946281433
Inter Cos: 0.20653969049453735
Norm Quadratic Average: 6.895991802215576
Nearest Class Center Accuracy: 0.9893

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7701640725135803
Inter Cos: 0.24609625339508057
Norm Quadratic Average: 5.8776373863220215
Nearest Class Center Accuracy: 0.9921

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.80171203613281
Linear Weight Rank: 4031
Intra Cos: 0.9040840268135071
Inter Cos: 0.1971970796585083
Norm Quadratic Average: 33.010215759277344
Nearest Class Center Accuracy: 0.9929

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.918193817138672
Linear Weight Rank: 3669
Intra Cos: 0.929914116859436
Inter Cos: 0.18145665526390076
Norm Quadratic Average: 28.027803421020508
Nearest Class Center Accuracy: 0.9935

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.994596004486084
Linear Weight Rank: 10
Intra Cos: 0.9343864321708679
Inter Cos: 0.16132666170597076
Norm Quadratic Average: 25.737031936645508
Nearest Class Center Accuracy: 0.9932

Output Layer:
Intra Cos: 0.9574363231658936
Inter Cos: 0.23656581342220306
Norm Quadratic Average: 25.353933334350586
Nearest Class Center Accuracy: 0.9937

