Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_265358_test_samples_None_train_samples_None_weight_decay_0.0003.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: 23.567678451538086
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.06855394691228867
Inter Cos: 0.08688580989837646
Norm Quadratic Average: 43.23653030395508
Nearest Class Center Accuracy: 0.8260833333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10833527147769928
Inter Cos: 0.10677851736545563
Norm Quadratic Average: 25.133041381835938
Nearest Class Center Accuracy: 0.8726666666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10820338875055313
Inter Cos: 0.10626410692930222
Norm Quadratic Average: 27.295177459716797
Nearest Class Center Accuracy: 0.8806833333333334

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17595519125461578
Inter Cos: 0.1156168133020401
Norm Quadratic Average: 16.63074493408203
Nearest Class Center Accuracy: 0.9337666666666666

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20223112404346466
Inter Cos: 0.11740611493587494
Norm Quadratic Average: 19.3236141204834
Nearest Class Center Accuracy: 0.9554333333333334

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2284124195575714
Inter Cos: 0.11625194549560547
Norm Quadratic Average: 19.577428817749023
Nearest Class Center Accuracy: 0.9672833333333334

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25153860449790955
Inter Cos: 0.10994629561901093
Norm Quadratic Average: 20.639408111572266
Nearest Class Center Accuracy: 0.9752333333333333

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.32114294171333313
Inter Cos: 0.13336603343486786
Norm Quadratic Average: 13.9003267288208
Nearest Class Center Accuracy: 0.9922666666666666

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4505898356437683
Inter Cos: 0.16991795599460602
Norm Quadratic Average: 14.903143882751465
Nearest Class Center Accuracy: 0.9972166666666666

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5699268579483032
Inter Cos: 0.1783166378736496
Norm Quadratic Average: 15.731985092163086
Nearest Class Center Accuracy: 0.9989333333333333

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6703420877456665
Inter Cos: 0.14312660694122314
Norm Quadratic Average: 16.026578903198242
Nearest Class Center Accuracy: 0.9998

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7997995018959045
Inter Cos: 0.17373837530612946
Norm Quadratic Average: 13.159231185913086
Nearest Class Center Accuracy: 0.9999

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9275960326194763
Inter Cos: 0.07421967387199402
Norm Quadratic Average: 8.479875564575195
Nearest Class Center Accuracy: 0.9999833333333333

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.961332380771637
Inter Cos: 0.01924377866089344
Norm Quadratic Average: 8.788519859313965
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9756894707679749
Inter Cos: -0.012591036036610603
Norm Quadratic Average: 8.851723670959473
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.675586700439453
Linear Weight Rank: 4031
Intra Cos: 0.9894616007804871
Inter Cos: 0.006849439814686775
Norm Quadratic Average: 73.47218322753906
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 12.969688415527344
Linear Weight Rank: 3670
Intra Cos: 0.9923603534698486
Inter Cos: 0.020095834508538246
Norm Quadratic Average: 41.47777557373047
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.9574470520019531
Linear Weight Rank: 10
Intra Cos: 0.9907501339912415
Inter Cos: 0.05018211156129837
Norm Quadratic Average: 23.59662437438965
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9989923238754272
Inter Cos: 0.16843825578689575
Norm Quadratic Average: 14.788546562194824
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.021362950712312884
Accuracy: 0.9955
NC1 Within Class Collapse: 0.14119073748588562
NC2 Equinorm: Features: 0.020438438281416893, Weights: 0.014764875173568726
NC2 Equiangle: Features: 0.09664276970757378, Weights: 0.07575872209337023
NC3 Self-Duality: 0.17423595488071442
NC4 NCC Mismatch: 0.00029999999999996696

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
Inter Cos: 0.12048853933811188
Norm Quadratic Average: 23.595195770263672
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.07686351984739304
Inter Cos: 0.08817014843225479
Norm Quadratic Average: 43.13823699951172
Nearest Class Center Accuracy: 0.8386

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11888197809457779
Inter Cos: 0.10722371190786362
Norm Quadratic Average: 24.960813522338867
Nearest Class Center Accuracy: 0.8847

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11819642037153244
Inter Cos: 0.10694112628698349
Norm Quadratic Average: 27.143640518188477
Nearest Class Center Accuracy: 0.8919

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18875476717948914
Inter Cos: 0.12590160965919495
Norm Quadratic Average: 16.52971649169922
Nearest Class Center Accuracy: 0.9407

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21595816314220428
Inter Cos: 0.12298629432916641
Norm Quadratic Average: 19.21526527404785
Nearest Class Center Accuracy: 0.9599

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24240867793560028
Inter Cos: 0.1129733994603157
Norm Quadratic Average: 19.48355484008789
Nearest Class Center Accuracy: 0.97

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.264767050743103
Inter Cos: 0.11188679188489914
Norm Quadratic Average: 20.54688835144043
Nearest Class Center Accuracy: 0.9752

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3314248323440552
Inter Cos: 0.13572163879871368
Norm Quadratic Average: 13.846467018127441
Nearest Class Center Accuracy: 0.9892

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.45728421211242676
Inter Cos: 0.16952526569366455
Norm Quadratic Average: 14.85609245300293
Nearest Class Center Accuracy: 0.9923

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5747472643852234
Inter Cos: 0.17626872658729553
Norm Quadratic Average: 15.688648223876953
Nearest Class Center Accuracy: 0.9936

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6730117797851562
Inter Cos: 0.1390828937292099
Norm Quadratic Average: 15.985957145690918
Nearest Class Center Accuracy: 0.9949

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.797131359577179
Inter Cos: 0.16802608966827393
Norm Quadratic Average: 13.133868217468262
Nearest Class Center Accuracy: 0.994

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9171397089958191
Inter Cos: 0.06808207184076309
Norm Quadratic Average: 8.463313102722168
Nearest Class Center Accuracy: 0.9943

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.942887008190155
Inter Cos: 0.018734076991677284
Norm Quadratic Average: 8.77058219909668
Nearest Class Center Accuracy: 0.9951

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9565414190292358
Inter Cos: -0.0005688858218491077
Norm Quadratic Average: 8.833416938781738
Nearest Class Center Accuracy: 0.9956

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.675586700439453
Linear Weight Rank: 4031
Intra Cos: 0.9678691625595093
Inter Cos: 0.00422601867467165
Norm Quadratic Average: 73.29309844970703
Nearest Class Center Accuracy: 0.9955

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 12.969688415527344
Linear Weight Rank: 3670
Intra Cos: 0.9720103740692139
Inter Cos: 0.019405348226428032
Norm Quadratic Average: 41.376869201660156
Nearest Class Center Accuracy: 0.9954

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.9574470520019531
Linear Weight Rank: 10
Intra Cos: 0.9690027236938477
Inter Cos: 0.05699295923113823
Norm Quadratic Average: 23.546581268310547
Nearest Class Center Accuracy: 0.9954

Output Layer:
Intra Cos: 0.9855653643608093
Inter Cos: 0.17554154992103577
Norm Quadratic Average: 14.746000289916992
Nearest Class Center Accuracy: 0.9956

