Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.005.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09116753935813904
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567676544189453
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12321604043245316
Inter Cos: 0.15080943703651428
Norm Quadratic Average: 40.522239685058594
Nearest Class Center Accuracy: 0.8053333333333333

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17474393546581268
Inter Cos: 0.17725799977779388
Norm Quadratic Average: 41.56135177612305
Nearest Class Center Accuracy: 0.8166833333333333

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2080041766166687
Inter Cos: 0.20965009927749634
Norm Quadratic Average: 41.04698181152344
Nearest Class Center Accuracy: 0.8616

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19301584362983704
Inter Cos: 0.22357264161109924
Norm Quadratic Average: 20.982194900512695
Nearest Class Center Accuracy: 0.9038166666666667

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2516297698020935
Inter Cos: 0.276957243680954
Norm Quadratic Average: 12.66594409942627
Nearest Class Center Accuracy: 0.9359

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.49250301718711853
Inter Cos: 0.31814730167388916
Norm Quadratic Average: 7.133996486663818
Nearest Class Center Accuracy: 0.9726166666666667

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7118170857429504
Inter Cos: 0.4058099687099457
Norm Quadratic Average: 7.15601921081543
Nearest Class Center Accuracy: 0.9885333333333334

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.9844180345535278
Linear Weight Rank: 52
Intra Cos: 0.809423565864563
Inter Cos: 0.4129142463207245
Norm Quadratic Average: 34.5255241394043
Nearest Class Center Accuracy: 0.9933

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.9867615699768066
Linear Weight Rank: 2745
Intra Cos: 0.8857876062393188
Inter Cos: 0.41097885370254517
Norm Quadratic Average: 27.378583908081055
Nearest Class Center Accuracy: 0.9961666666666666

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.9800063371658325
Linear Weight Rank: 9
Intra Cos: 0.9103206396102905
Inter Cos: 0.371857225894928
Norm Quadratic Average: 21.02596092224121
Nearest Class Center Accuracy: 0.99695

Output Layer:
Intra Cos: 0.9306923747062683
Inter Cos: 0.4330408275127411
Norm Quadratic Average: 18.368127822875977
Nearest Class Center Accuracy: 0.9970666666666667

Test Set:
Average Loss: 0.026471363544277846
Accuracy: 0.9912
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.10749663412570953, Weights: 0.04480446130037308
NC2 Equiangle: Features: 0.2696933110555013, Weights: 0.24407278696695964
NC3 Self-Duality: 0.055359501391649246
NC4 NCC Mismatch: 0.0047000000000000375

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582170248032
Inter Cos: 0.12048851698637009
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: 0.1367543786764145
Inter Cos: 0.16514480113983154
Norm Quadratic Average: 40.5731315612793
Nearest Class Center Accuracy: 0.8198

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19100557267665863
Inter Cos: 0.19193914532661438
Norm Quadratic Average: 41.49928665161133
Nearest Class Center Accuracy: 0.8326

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22234226763248444
Inter Cos: 0.2282680869102478
Norm Quadratic Average: 41.016090393066406
Nearest Class Center Accuracy: 0.8759

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20253372192382812
Inter Cos: 0.2287999391555786
Norm Quadratic Average: 20.946613311767578
Nearest Class Center Accuracy: 0.9182

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25973159074783325
Inter Cos: 0.28937405347824097
Norm Quadratic Average: 12.659317016601562
Nearest Class Center Accuracy: 0.9418

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5017956495285034
Inter Cos: 0.3460712730884552
Norm Quadratic Average: 7.147251605987549
Nearest Class Center Accuracy: 0.9702

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7236143946647644
Inter Cos: 0.4317944049835205
Norm Quadratic Average: 7.201704502105713
Nearest Class Center Accuracy: 0.9835

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.9844180345535278
Linear Weight Rank: 52
Intra Cos: 0.8114630579948425
Inter Cos: 0.43796712160110474
Norm Quadratic Average: 34.80740737915039
Nearest Class Center Accuracy: 0.9882

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.9867615699768066
Linear Weight Rank: 2745
Intra Cos: 0.8828531503677368
Inter Cos: 0.4345117509365082
Norm Quadratic Average: 27.62370491027832
Nearest Class Center Accuracy: 0.9901

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.9800063371658325
Linear Weight Rank: 9
Intra Cos: 0.9026435613632202
Inter Cos: 0.3945882022380829
Norm Quadratic Average: 21.211143493652344
Nearest Class Center Accuracy: 0.9897

Output Layer:
Intra Cos: 0.9164019823074341
Inter Cos: 0.4540029466152191
Norm Quadratic Average: 18.534074783325195
Nearest Class Center Accuracy: 0.9897

