Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_338327_test_samples_None_train_samples_None_weight_decay_0.003.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: 0.10967152565717697
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: 0.05933517962694168
Inter Cos: 0.0796177089214325
Norm Quadratic Average: 2.702589511871338
Nearest Class Center Accuracy: 0.8085333333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10505624860525131
Inter Cos: 0.09710725396871567
Norm Quadratic Average: 1.630576252937317
Nearest Class Center Accuracy: 0.87585

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10007178038358688
Inter Cos: 0.09311574697494507
Norm Quadratic Average: 1.349827527999878
Nearest Class Center Accuracy: 0.8816166666666667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17625655233860016
Inter Cos: 0.11600233614444733
Norm Quadratic Average: 0.9838293790817261
Nearest Class Center Accuracy: 0.9399666666666666

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2249259501695633
Inter Cos: 0.11821479350328445
Norm Quadratic Average: 0.7582390308380127
Nearest Class Center Accuracy: 0.9647

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28250619769096375
Inter Cos: 0.11583391577005386
Norm Quadratic Average: 0.621623158454895
Nearest Class Center Accuracy: 0.9769166666666667

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.32851535081863403
Inter Cos: 0.12189656496047974
Norm Quadratic Average: 0.5517318248748779
Nearest Class Center Accuracy: 0.9809333333333333

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4037449061870575
Inter Cos: 0.13937753438949585
Norm Quadratic Average: 0.40913447737693787
Nearest Class Center Accuracy: 0.9938333333333333

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6627828478813171
Inter Cos: 0.22049671411514282
Norm Quadratic Average: 0.2888084053993225
Nearest Class Center Accuracy: 0.9985166666666667

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7977249622344971
Inter Cos: 0.17350175976753235
Norm Quadratic Average: 0.2460377812385559
Nearest Class Center Accuracy: 0.9992166666666666

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8650476336479187
Inter Cos: 0.09834760427474976
Norm Quadratic Average: 0.22169138491153717
Nearest Class Center Accuracy: 0.9999833333333333

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9000067114830017
Inter Cos: 0.08449447154998779
Norm Quadratic Average: 0.28206485509872437
Nearest Class Center Accuracy: 0.9999833333333333

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9872439503669739
Inter Cos: 0.015533648431301117
Norm Quadratic Average: 0.324309766292572
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.996196985244751
Inter Cos: -0.0022285617887973785
Norm Quadratic Average: 0.5656729340553284
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9983001947402954
Inter Cos: -0.043670982122421265
Norm Quadratic Average: 1.0892863273620605
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.19797945022583
Linear Weight Rank: 135
Intra Cos: 0.9992871284484863
Inter Cos: -0.0391523651778698
Norm Quadratic Average: 25.82036590576172
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.2045016288757324
Linear Weight Rank: 1412
Intra Cos: 0.9993904829025269
Inter Cos: 0.006766807287931442
Norm Quadratic Average: 17.63780975341797
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2012641429901123
Linear Weight Rank: 9
Intra Cos: 0.9994508028030396
Inter Cos: 0.04192515090107918
Norm Quadratic Average: 12.367718696594238
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9995684027671814
Inter Cos: 0.07015721499919891
Norm Quadratic Average: 9.170680046081543
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.01843245713012293
Accuracy: 0.9954
NC1 Within Class Collapse: 0.07799392938613892
NC2 Equinorm: Features: 0.016073599457740784, Weights: 0.006030412390828133
NC2 Equiangle: Features: 0.06812930636935764, Weights: 0.02988591194152832
NC3 Self-Duality: 0.009193750098347664
NC4 NCC Mismatch: 0.00029999999999996696

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582170248032
Inter Cos: 0.12048853188753128
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: 0.0675308108329773
Inter Cos: 0.08271512389183044
Norm Quadratic Average: 2.6929068565368652
Nearest Class Center Accuracy: 0.8184

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11498230695724487
Inter Cos: 0.09911813586950302
Norm Quadratic Average: 1.618272304534912
Nearest Class Center Accuracy: 0.8883

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10985153913497925
Inter Cos: 0.09574541449546814
Norm Quadratic Average: 1.343814492225647
Nearest Class Center Accuracy: 0.8919

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18778392672538757
Inter Cos: 0.12442249804735184
Norm Quadratic Average: 0.9793104529380798
Nearest Class Center Accuracy: 0.946

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23964053392410278
Inter Cos: 0.13032282888889313
Norm Quadratic Average: 0.7568193674087524
Nearest Class Center Accuracy: 0.9677

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.29856541752815247
Inter Cos: 0.128750279545784
Norm Quadratic Average: 0.621593177318573
Nearest Class Center Accuracy: 0.9749

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.34493568539619446
Inter Cos: 0.13431291282176971
Norm Quadratic Average: 0.5515756011009216
Nearest Class Center Accuracy: 0.9784

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.41811612248420715
Inter Cos: 0.15485604107379913
Norm Quadratic Average: 0.40889477729797363
Nearest Class Center Accuracy: 0.9895

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6740566492080688
Inter Cos: 0.23285333812236786
Norm Quadratic Average: 0.2893526256084442
Nearest Class Center Accuracy: 0.9933

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7899233102798462
Inter Cos: 0.1813211292028427
Norm Quadratic Average: 0.24681901931762695
Nearest Class Center Accuracy: 0.9947

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8678790330886841
Inter Cos: 0.09559629112482071
Norm Quadratic Average: 0.22183290123939514
Nearest Class Center Accuracy: 0.9952

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8972991704940796
Inter Cos: 0.0927770808339119
Norm Quadratic Average: 0.28164657950401306
Nearest Class Center Accuracy: 0.9956

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9735794067382812
Inter Cos: 0.02909757010638714
Norm Quadratic Average: 0.3234190344810486
Nearest Class Center Accuracy: 0.9953

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.976844847202301
Inter Cos: 0.011318568140268326
Norm Quadratic Average: 0.5640977621078491
Nearest Class Center Accuracy: 0.9953

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9784877896308899
Inter Cos: -0.030495576560497284
Norm Quadratic Average: 1.0861538648605347
Nearest Class Center Accuracy: 0.9953

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.19797945022583
Linear Weight Rank: 135
Intra Cos: 0.980462908744812
Inter Cos: -0.03081873431801796
Norm Quadratic Average: 25.739532470703125
Nearest Class Center Accuracy: 0.9952

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.2045016288757324
Linear Weight Rank: 1412
Intra Cos: 0.981250524520874
Inter Cos: 0.016875574365258217
Norm Quadratic Average: 17.583290100097656
Nearest Class Center Accuracy: 0.9952

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2012641429901123
Linear Weight Rank: 9
Intra Cos: 0.9817858934402466
Inter Cos: 0.05213213711977005
Norm Quadratic Average: 12.33026123046875
Nearest Class Center Accuracy: 0.9953

Output Layer:
Intra Cos: 0.9824056029319763
Inter Cos: 0.08032691478729248
Norm Quadratic Average: 9.142708778381348
Nearest Class Center Accuracy: 0.9953

