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_0.0003.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: 0.10967152565717697
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: 0.0671720877289772
Inter Cos: 0.08071420341730118
Norm Quadratic Average: 46.60966110229492
Nearest Class Center Accuracy: 0.8260333333333333

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10456584393978119
Inter Cos: 0.10074147582054138
Norm Quadratic Average: 26.98393440246582
Nearest Class Center Accuracy: 0.8656333333333334

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1038563996553421
Inter Cos: 0.10230933874845505
Norm Quadratic Average: 27.118722915649414
Nearest Class Center Accuracy: 0.8742

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17701812088489532
Inter Cos: 0.1249699741601944
Norm Quadratic Average: 17.67592430114746
Nearest Class Center Accuracy: 0.9273833333333333

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21429695188999176
Inter Cos: 0.1296442151069641
Norm Quadratic Average: 19.48228645324707
Nearest Class Center Accuracy: 0.9499833333333333

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.241534024477005
Inter Cos: 0.13143201172351837
Norm Quadratic Average: 19.6866512298584
Nearest Class Center Accuracy: 0.96295

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2704770863056183
Inter Cos: 0.13280339539051056
Norm Quadratic Average: 20.440046310424805
Nearest Class Center Accuracy: 0.9707

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.33210489153862
Inter Cos: 0.1531258374452591
Norm Quadratic Average: 13.872350692749023
Nearest Class Center Accuracy: 0.9918166666666667

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.47355809807777405
Inter Cos: 0.1802055686712265
Norm Quadratic Average: 14.643851280212402
Nearest Class Center Accuracy: 0.9975666666666667

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5886774063110352
Inter Cos: 0.15516985952854156
Norm Quadratic Average: 15.563972473144531
Nearest Class Center Accuracy: 0.9994833333333333

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6935566663742065
Inter Cos: 0.11035919934511185
Norm Quadratic Average: 15.84768295288086
Nearest Class Center Accuracy: 0.9999333333333333

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8082693815231323
Inter Cos: 0.16397038102149963
Norm Quadratic Average: 12.909088134765625
Nearest Class Center Accuracy: 0.9999833333333333

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9179492592811584
Inter Cos: 0.09490092098712921
Norm Quadratic Average: 8.310343742370605
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9570834040641785
Inter Cos: 0.06741000711917877
Norm Quadratic Average: 8.646476745605469
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9764143824577332
Inter Cos: 0.026055878028273582
Norm Quadratic Average: 8.852676391601562
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.6719913482666
Linear Weight Rank: 4031
Intra Cos: 0.9898403286933899
Inter Cos: -0.003906322177499533
Norm Quadratic Average: 73.92082214355469
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 12.968335151672363
Linear Weight Rank: 3670
Intra Cos: 0.9930894374847412
Inter Cos: 0.0174652598798275
Norm Quadratic Average: 41.551815032958984
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.9530353546142578
Linear Weight Rank: 10
Intra Cos: 0.9917457103729248
Inter Cos: 0.03231913223862648
Norm Quadratic Average: 23.417573928833008
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9987122416496277
Inter Cos: 0.1540174037218094
Norm Quadratic Average: 14.550337791442871
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.0183305815860178
Accuracy: 0.9964
NC1 Within Class Collapse: 0.13151368498802185
NC2 Equinorm: Features: 0.026073193177580833, Weights: 0.016330821439623833
NC2 Equiangle: Features: 0.09000203874376085, Weights: 0.06872907214694553
NC3 Self-Duality: 0.1713458150625229
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: 0.07549974322319031
Inter Cos: 0.08283351361751556
Norm Quadratic Average: 46.48958206176758
Nearest Class Center Accuracy: 0.8388

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11574076116085052
Inter Cos: 0.10195767879486084
Norm Quadratic Average: 26.78647804260254
Nearest Class Center Accuracy: 0.8755

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11435279250144958
Inter Cos: 0.10362465679645538
Norm Quadratic Average: 26.93749237060547
Nearest Class Center Accuracy: 0.8819

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1919279545545578
Inter Cos: 0.12550565600395203
Norm Quadratic Average: 17.54497718811035
Nearest Class Center Accuracy: 0.9339

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2261771410703659
Inter Cos: 0.12694981694221497
Norm Quadratic Average: 19.347244262695312
Nearest Class Center Accuracy: 0.9541

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25335341691970825
Inter Cos: 0.12804844975471497
Norm Quadratic Average: 19.562179565429688
Nearest Class Center Accuracy: 0.9652

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2819887697696686
Inter Cos: 0.12959280610084534
Norm Quadratic Average: 20.3226261138916
Nearest Class Center Accuracy: 0.9714

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.34336790442466736
Inter Cos: 0.15740296244621277
Norm Quadratic Average: 13.809174537658691
Nearest Class Center Accuracy: 0.9889

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.48231953382492065
Inter Cos: 0.18274223804473877
Norm Quadratic Average: 14.597980499267578
Nearest Class Center Accuracy: 0.9928

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5952789187431335
Inter Cos: 0.15571001172065735
Norm Quadratic Average: 15.533683776855469
Nearest Class Center Accuracy: 0.9944

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6962747573852539
Inter Cos: 0.1127154603600502
Norm Quadratic Average: 15.834671020507812
Nearest Class Center Accuracy: 0.9948

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8037930130958557
Inter Cos: 0.1622314751148224
Norm Quadratic Average: 12.906421661376953
Nearest Class Center Accuracy: 0.9944

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9091804623603821
Inter Cos: 0.09837786853313446
Norm Quadratic Average: 8.3065824508667
Nearest Class Center Accuracy: 0.9952

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9436917304992676
Inter Cos: 0.06962375342845917
Norm Quadratic Average: 8.63763427734375
Nearest Class Center Accuracy: 0.9955

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9563233256340027
Inter Cos: 0.027442127466201782
Norm Quadratic Average: 8.838375091552734
Nearest Class Center Accuracy: 0.9962

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.6719913482666
Linear Weight Rank: 4031
Intra Cos: 0.9684759378433228
Inter Cos: -0.008136188611388206
Norm Quadratic Average: 73.76500701904297
Nearest Class Center Accuracy: 0.9963

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 12.968335151672363
Linear Weight Rank: 3670
Intra Cos: 0.9722322821617126
Inter Cos: 0.012571487575769424
Norm Quadratic Average: 41.4625129699707
Nearest Class Center Accuracy: 0.9965

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.9530353546142578
Linear Weight Rank: 10
Intra Cos: 0.970140814781189
Inter Cos: 0.04260718822479248
Norm Quadratic Average: 23.374929428100586
Nearest Class Center Accuracy: 0.9963

Output Layer:
Intra Cos: 0.9851033091545105
Inter Cos: 0.15960249304771423
Norm Quadratic Average: 14.511762619018555
Nearest Class Center Accuracy: 0.9962

