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_0.005.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: 0.10967151075601578
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: 0.11623565107584
Inter Cos: 0.1399361938238144
Norm Quadratic Average: 68.41864776611328
Nearest Class Center Accuracy: 0.7997333333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13717171549797058
Inter Cos: 0.1785355806350708
Norm Quadratic Average: 121.87239837646484
Nearest Class Center Accuracy: 0.7802166666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14444482326507568
Inter Cos: 0.18963955342769623
Norm Quadratic Average: 226.52049255371094
Nearest Class Center Accuracy: 0.7816666666666666

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17059506475925446
Inter Cos: 0.19451439380645752
Norm Quadratic Average: 148.2255401611328
Nearest Class Center Accuracy: 0.8216

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1970098912715912
Inter Cos: 0.2389427274465561
Norm Quadratic Average: 106.57595825195312
Nearest Class Center Accuracy: 0.85165

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22075557708740234
Inter Cos: 0.2715767025947571
Norm Quadratic Average: 83.08497619628906
Nearest Class Center Accuracy: 0.8878

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2740565836429596
Inter Cos: 0.24898405373096466
Norm Quadratic Average: 59.391014099121094
Nearest Class Center Accuracy: 0.9239666666666667

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24196122586727142
Inter Cos: 0.2777319848537445
Norm Quadratic Average: 22.830303192138672
Nearest Class Center Accuracy: 0.8945666666666666

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28602081537246704
Inter Cos: 0.29251939058303833
Norm Quadratic Average: 14.571313858032227
Nearest Class Center Accuracy: 0.81595

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3963966369628906
Inter Cos: 0.24432355165481567
Norm Quadratic Average: 14.813432693481445
Nearest Class Center Accuracy: 0.8663166666666666

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5703747272491455
Inter Cos: 0.36802759766578674
Norm Quadratic Average: 18.247295379638672
Nearest Class Center Accuracy: 0.9404166666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6092661023139954
Inter Cos: 0.3077586889266968
Norm Quadratic Average: 14.086968421936035
Nearest Class Center Accuracy: 0.9340166666666667

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7097543478012085
Inter Cos: 0.390511155128479
Norm Quadratic Average: 12.64302921295166
Nearest Class Center Accuracy: 0.94085

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7675952911376953
Inter Cos: 0.3877842426300049
Norm Quadratic Average: 14.546445846557617
Nearest Class Center Accuracy: 0.9579666666666666

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7956772446632385
Inter Cos: 0.4563189446926117
Norm Quadratic Average: 17.322166442871094
Nearest Class Center Accuracy: 0.9682833333333334

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.5883793830871582
Linear Weight Rank: 16
Intra Cos: 0.8366716504096985
Inter Cos: 0.5076692700386047
Norm Quadratic Average: 79.73475646972656
Nearest Class Center Accuracy: 0.9843833333333334

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.5996092557907104
Linear Weight Rank: 2525
Intra Cos: 0.8758138418197632
Inter Cos: 0.5000173449516296
Norm Quadratic Average: 54.24079132080078
Nearest Class Center Accuracy: 0.9941166666666666

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.5886982679367065
Linear Weight Rank: 9
Intra Cos: 0.8993076086044312
Inter Cos: 0.4403138756752014
Norm Quadratic Average: 33.027713775634766
Nearest Class Center Accuracy: 0.9972833333333333

Output Layer:
Intra Cos: 0.9457588791847229
Inter Cos: 0.4927632212638855
Norm Quadratic Average: 21.685964584350586
Nearest Class Center Accuracy: 0.9988833333333333

Test Set:
Average Loss: 0.03740056258710101
Accuracy: 0.9893
NC1 Within Class Collapse: 1.2250268459320068
NC2 Equinorm: Features: 0.15726105868816376, Weights: 0.06577004492282867
NC2 Equiangle: Features: 0.3062538570827908, Weights: 0.19448841942681205
NC3 Self-Duality: 0.13380734622478485
NC4 NCC Mismatch: 0.007499999999999951

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: 0.12780757248401642
Inter Cos: 0.1532648354768753
Norm Quadratic Average: 68.83941650390625
Nearest Class Center Accuracy: 0.8165

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15140701830387115
Inter Cos: 0.19547110795974731
Norm Quadratic Average: 122.46682739257812
Nearest Class Center Accuracy: 0.7995

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1593376100063324
Inter Cos: 0.2079218477010727
Norm Quadratic Average: 227.6534881591797
Nearest Class Center Accuracy: 0.8026

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18215303122997284
Inter Cos: 0.2140277475118637
Norm Quadratic Average: 148.63563537597656
Nearest Class Center Accuracy: 0.8363

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21016950905323029
Inter Cos: 0.2609962224960327
Norm Quadratic Average: 106.88219451904297
Nearest Class Center Accuracy: 0.8676

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23457373678684235
Inter Cos: 0.29411134123802185
Norm Quadratic Average: 83.42070007324219
Nearest Class Center Accuracy: 0.9006

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2885560691356659
Inter Cos: 0.26061367988586426
Norm Quadratic Average: 59.82763671875
Nearest Class Center Accuracy: 0.9288

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2554358243942261
Inter Cos: 0.27456414699554443
Norm Quadratic Average: 22.99481773376465
Nearest Class Center Accuracy: 0.9

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3015837073326111
Inter Cos: 0.27415868639945984
Norm Quadratic Average: 14.639983177185059
Nearest Class Center Accuracy: 0.8298

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.41525858640670776
Inter Cos: 0.24411378800868988
Norm Quadratic Average: 14.891403198242188
Nearest Class Center Accuracy: 0.8783

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5894079804420471
Inter Cos: 0.38526031374931335
Norm Quadratic Average: 18.378551483154297
Nearest Class Center Accuracy: 0.9361

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6254824995994568
Inter Cos: 0.314812570810318
Norm Quadratic Average: 14.168486595153809
Nearest Class Center Accuracy: 0.9305

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7261320352554321
Inter Cos: 0.3910691440105438
Norm Quadratic Average: 12.736989974975586
Nearest Class Center Accuracy: 0.9353

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7787299752235413
Inter Cos: 0.4122174382209778
Norm Quadratic Average: 14.677619934082031
Nearest Class Center Accuracy: 0.9505

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8004990816116333
Inter Cos: 0.48139116168022156
Norm Quadratic Average: 17.502641677856445
Nearest Class Center Accuracy: 0.9606

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.5883793830871582
Linear Weight Rank: 16
Intra Cos: 0.8380774855613708
Inter Cos: 0.5285282135009766
Norm Quadratic Average: 80.71165466308594
Nearest Class Center Accuracy: 0.9736

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.5996092557907104
Linear Weight Rank: 2525
Intra Cos: 0.8744723796844482
Inter Cos: 0.5165839791297913
Norm Quadratic Average: 54.89861297607422
Nearest Class Center Accuracy: 0.9839

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.5886982679367065
Linear Weight Rank: 9
Intra Cos: 0.8970258235931396
Inter Cos: 0.45669975876808167
Norm Quadratic Average: 33.42413330078125
Nearest Class Center Accuracy: 0.987

Output Layer:
Intra Cos: 0.9380533695220947
Inter Cos: 0.5071014761924744
Norm Quadratic Average: 21.954551696777344
Nearest Class Center Accuracy: 0.9892

