Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_314159_test_samples_None_train_samples_None_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: 0.09116753935813904
Inter Cos: 0.10967151075601578
Norm Quadratic Average: 23.567678451538086
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11857105791568756
Inter Cos: 0.1421971619129181
Norm Quadratic Average: 69.77672576904297
Nearest Class Center Accuracy: 0.7991

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14098641276359558
Inter Cos: 0.18231941759586334
Norm Quadratic Average: 139.669189453125
Nearest Class Center Accuracy: 0.77505

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1451360583305359
Inter Cos: 0.1877487301826477
Norm Quadratic Average: 259.7941589355469
Nearest Class Center Accuracy: 0.7754166666666666

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17204509675502777
Inter Cos: 0.18758612871170044
Norm Quadratic Average: 153.16799926757812
Nearest Class Center Accuracy: 0.8222166666666667

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2048627734184265
Inter Cos: 0.19942715764045715
Norm Quadratic Average: 92.74998474121094
Nearest Class Center Accuracy: 0.8615666666666667

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24380557239055634
Inter Cos: 0.24490641057491302
Norm Quadratic Average: 74.26692199707031
Nearest Class Center Accuracy: 0.8927166666666667

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2878337800502777
Inter Cos: 0.2837171256542206
Norm Quadratic Average: 78.09489440917969
Nearest Class Center Accuracy: 0.9158833333333334

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2678581476211548
Inter Cos: 0.27662089467048645
Norm Quadratic Average: 42.214317321777344
Nearest Class Center Accuracy: 0.90865

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.27489909529685974
Inter Cos: 0.30550867319107056
Norm Quadratic Average: 28.895078659057617
Nearest Class Center Accuracy: 0.88245

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.31660985946655273
Inter Cos: 0.3525424003601074
Norm Quadratic Average: 28.35356903076172
Nearest Class Center Accuracy: 0.9012333333333333

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3933742344379425
Inter Cos: 0.3271563947200775
Norm Quadratic Average: 29.385305404663086
Nearest Class Center Accuracy: 0.9292333333333334

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.45164236426353455
Inter Cos: 0.4243420362472534
Norm Quadratic Average: 18.94598960876465
Nearest Class Center Accuracy: 0.9366

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6149088740348816
Inter Cos: 0.5065194368362427
Norm Quadratic Average: 16.186992645263672
Nearest Class Center Accuracy: 0.9554666666666667

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6740593314170837
Inter Cos: 0.4635930061340332
Norm Quadratic Average: 17.60239601135254
Nearest Class Center Accuracy: 0.9703333333333334

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7088745832443237
Inter Cos: 0.4300437569618225
Norm Quadratic Average: 19.727584838867188
Nearest Class Center Accuracy: 0.9780666666666666

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.581957221031189
Linear Weight Rank: 22
Intra Cos: 0.7770975232124329
Inter Cos: 0.42081815004348755
Norm Quadratic Average: 80.52313995361328
Nearest Class Center Accuracy: 0.9868833333333333

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.5921095609664917
Linear Weight Rank: 2767
Intra Cos: 0.85077303647995
Inter Cos: 0.3917159140110016
Norm Quadratic Average: 49.37269973754883
Nearest Class Center Accuracy: 0.9911666666666666

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.5786076784133911
Linear Weight Rank: 9
Intra Cos: 0.8754281997680664
Inter Cos: 0.347606360912323
Norm Quadratic Average: 28.869834899902344
Nearest Class Center Accuracy: 0.9923666666666666

Output Layer:
Intra Cos: 0.9170944094657898
Inter Cos: 0.40168827772140503
Norm Quadratic Average: 18.879762649536133
Nearest Class Center Accuracy: 0.99425

Test Set:
Average Loss: 0.033487794906646016
Accuracy: 0.9894
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.1276925504207611, Weights: 0.046279530972242355
NC2 Equiangle: Features: 0.2677998224894206, Weights: 0.19898738861083984
NC3 Self-Duality: 0.10766200721263885
NC4 NCC Mismatch: 0.007900000000000018

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
Inter Cos: 0.12048851698637009
Norm Quadratic Average: 23.595197677612305
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13024871051311493
Inter Cos: 0.15567967295646667
Norm Quadratic Average: 70.2166976928711
Nearest Class Center Accuracy: 0.8155

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1553298383951187
Inter Cos: 0.19972169399261475
Norm Quadratic Average: 140.40060424804688
Nearest Class Center Accuracy: 0.796

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1593623161315918
Inter Cos: 0.2061883807182312
Norm Quadratic Average: 261.1930847167969
Nearest Class Center Accuracy: 0.7943

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18141168355941772
Inter Cos: 0.2068096101284027
Norm Quadratic Average: 153.62530517578125
Nearest Class Center Accuracy: 0.838

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21514299511909485
Inter Cos: 0.2196652591228485
Norm Quadratic Average: 93.06820678710938
Nearest Class Center Accuracy: 0.8744

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25467807054519653
Inter Cos: 0.2611199617385864
Norm Quadratic Average: 74.52559661865234
Nearest Class Center Accuracy: 0.9017

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3003213405609131
Inter Cos: 0.2935623228549957
Norm Quadratic Average: 78.56925964355469
Nearest Class Center Accuracy: 0.9226

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2800370454788208
Inter Cos: 0.2926381230354309
Norm Quadratic Average: 42.560997009277344
Nearest Class Center Accuracy: 0.9159

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28366604447364807
Inter Cos: 0.32849445939064026
Norm Quadratic Average: 29.146709442138672
Nearest Class Center Accuracy: 0.8901

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.32600417733192444
Inter Cos: 0.374329537153244
Norm Quadratic Average: 28.638158798217773
Nearest Class Center Accuracy: 0.9057

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.40549153089523315
Inter Cos: 0.34652259945869446
Norm Quadratic Average: 29.70425033569336
Nearest Class Center Accuracy: 0.9309

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4689890742301941
Inter Cos: 0.4549144506454468
Norm Quadratic Average: 19.15766143798828
Nearest Class Center Accuracy: 0.9368

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6253077983856201
Inter Cos: 0.5287871956825256
Norm Quadratic Average: 16.40096664428711
Nearest Class Center Accuracy: 0.9549

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6846863031387329
Inter Cos: 0.4848599433898926
Norm Quadratic Average: 17.847766876220703
Nearest Class Center Accuracy: 0.9678

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7244045734405518
Inter Cos: 0.4459719955921173
Norm Quadratic Average: 20.015613555908203
Nearest Class Center Accuracy: 0.9745

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.581957221031189
Linear Weight Rank: 22
Intra Cos: 0.7867097854614258
Inter Cos: 0.4358554184436798
Norm Quadratic Average: 81.7470703125
Nearest Class Center Accuracy: 0.9813

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.5921095609664917
Linear Weight Rank: 2767
Intra Cos: 0.8518494367599487
Inter Cos: 0.4069267213344574
Norm Quadratic Average: 50.15567398071289
Nearest Class Center Accuracy: 0.9847

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.5786076784133911
Linear Weight Rank: 9
Intra Cos: 0.8736358284950256
Inter Cos: 0.363229364156723
Norm Quadratic Average: 29.339317321777344
Nearest Class Center Accuracy: 0.9853

Output Layer:
Intra Cos: 0.9090104699134827
Inter Cos: 0.41663649678230286
Norm Quadratic Average: 19.194271087646484
Nearest Class Center Accuracy: 0.987

