Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.001.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09116754680871964
Inter Cos: 0.10967149585485458
Norm Quadratic Average: 23.567672729492188
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.05969102680683136
Inter Cos: 0.07376544177532196
Norm Quadratic Average: 6.736972332000732
Nearest Class Center Accuracy: 0.8181833333333334

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10119083523750305
Inter Cos: 0.0944388136267662
Norm Quadratic Average: 4.458428859710693
Nearest Class Center Accuracy: 0.88125

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09702154994010925
Inter Cos: 0.09262827783823013
Norm Quadratic Average: 4.1851959228515625
Nearest Class Center Accuracy: 0.8881166666666667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16636531054973602
Inter Cos: 0.11270903050899506
Norm Quadratic Average: 3.1568329334259033
Nearest Class Center Accuracy: 0.9400666666666667

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2112215906381607
Inter Cos: 0.11594893783330917
Norm Quadratic Average: 2.312190055847168
Nearest Class Center Accuracy: 0.9617333333333333

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.256343275308609
Inter Cos: 0.11392403393983841
Norm Quadratic Average: 2.1962804794311523
Nearest Class Center Accuracy: 0.9736666666666667

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2901827394962311
Inter Cos: 0.12527821958065033
Norm Quadratic Average: 2.1367299556732178
Nearest Class Center Accuracy: 0.9793

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.37374645471572876
Inter Cos: 0.11974926292896271
Norm Quadratic Average: 1.6540888547897339
Nearest Class Center Accuracy: 0.9934833333333334

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5773003697395325
Inter Cos: 0.11891042441129684
Norm Quadratic Average: 1.1587268114089966
Nearest Class Center Accuracy: 0.9985166666666667

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7334713339805603
Inter Cos: 0.08580222725868225
Norm Quadratic Average: 1.033181071281433
Nearest Class Center Accuracy: 0.9990833333333333

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8278377652168274
Inter Cos: 0.05694650113582611
Norm Quadratic Average: 0.8581681251525879
Nearest Class Center Accuracy: 0.9997833333333334

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9144086241722107
Inter Cos: 0.04708435386419296
Norm Quadratic Average: 0.7118076682090759
Nearest Class Center Accuracy: 0.9999833333333333

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9796121120452881
Inter Cos: 0.024711528792977333
Norm Quadratic Average: 0.6010589599609375
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9939659237861633
Inter Cos: -0.012373149394989014
Norm Quadratic Average: 0.6544104218482971
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9987820386886597
Inter Cos: -0.05135781317949295
Norm Quadratic Average: 1.0645228624343872
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.090716600418091
Linear Weight Rank: 4028
Intra Cos: 0.9996364116668701
Inter Cos: -0.0440252348780632
Norm Quadratic Average: 26.46121597290039
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.4775965213775635
Linear Weight Rank: 3637
Intra Cos: 0.9995906949043274
Inter Cos: 0.020876087248325348
Norm Quadratic Average: 18.776052474975586
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.300022602081299
Linear Weight Rank: 9
Intra Cos: 0.9995512366294861
Inter Cos: 0.04967385157942772
Norm Quadratic Average: 13.712296485900879
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9998754858970642
Inter Cos: 0.13612321019172668
Norm Quadratic Average: 10.828717231750488
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.024144969139667228
Accuracy: 0.9957
NC1 Within Class Collapse: 0.0838426798582077
NC2 Equinorm: Features: 0.017083128914237022, Weights: 0.01017400249838829
NC2 Equiangle: Features: 0.08473635779486763, Weights: 0.05543472501966688
NC3 Self-Duality: 0.014283712953329086
NC4 NCC Mismatch: 9.999999999998899e-05

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219583660364151
Inter Cos: 0.12048852443695068
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: 0.06725302338600159
Inter Cos: 0.07622309029102325
Norm Quadratic Average: 6.7113142013549805
Nearest Class Center Accuracy: 0.8297

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11056689918041229
Inter Cos: 0.09631966799497604
Norm Quadratic Average: 4.427108287811279
Nearest Class Center Accuracy: 0.8944

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1059495359659195
Inter Cos: 0.09381185472011566
Norm Quadratic Average: 4.164975166320801
Nearest Class Center Accuracy: 0.9002

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17718902230262756
Inter Cos: 0.11114733666181564
Norm Quadratic Average: 3.140291452407837
Nearest Class Center Accuracy: 0.9462

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2236626297235489
Inter Cos: 0.11392758786678314
Norm Quadratic Average: 2.3018996715545654
Nearest Class Center Accuracy: 0.9659

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26936012506484985
Inter Cos: 0.11224055290222168
Norm Quadratic Average: 2.187494993209839
Nearest Class Center Accuracy: 0.9742

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3038538694381714
Inter Cos: 0.12217403948307037
Norm Quadratic Average: 2.1289124488830566
Nearest Class Center Accuracy: 0.9784

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.38694629073143005
Inter Cos: 0.13266508281230927
Norm Quadratic Average: 1.6479460000991821
Nearest Class Center Accuracy: 0.9882

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5848824381828308
Inter Cos: 0.13061217963695526
Norm Quadratic Average: 1.1555757522583008
Nearest Class Center Accuracy: 0.992

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7359633445739746
Inter Cos: 0.09177954494953156
Norm Quadratic Average: 1.0309333801269531
Nearest Class Center Accuracy: 0.9929

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8265004754066467
Inter Cos: 0.0695771723985672
Norm Quadratic Average: 0.8569222092628479
Nearest Class Center Accuracy: 0.9945

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9084964394569397
Inter Cos: 0.04624378681182861
Norm Quadratic Average: 0.710263192653656
Nearest Class Center Accuracy: 0.9948

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9636337757110596
Inter Cos: 0.024203643202781677
Norm Quadratic Average: 0.5992358922958374
Nearest Class Center Accuracy: 0.9948

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9736104607582092
Inter Cos: -0.012327147647738457
Norm Quadratic Average: 0.6521190404891968
Nearest Class Center Accuracy: 0.9955

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9776715636253357
Inter Cos: -0.050756193697452545
Norm Quadratic Average: 1.0606262683868408
Nearest Class Center Accuracy: 0.9955

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.090716600418091
Linear Weight Rank: 4028
Intra Cos: 0.9799985289573669
Inter Cos: -0.036215364933013916
Norm Quadratic Average: 26.363649368286133
Nearest Class Center Accuracy: 0.9955

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.4775965213775635
Linear Weight Rank: 3637
Intra Cos: 0.9807307124137878
Inter Cos: 0.020855922251939774
Norm Quadratic Average: 18.707395553588867
Nearest Class Center Accuracy: 0.9956

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.300022602081299
Linear Weight Rank: 9
Intra Cos: 0.9812346696853638
Inter Cos: 0.06286155432462692
Norm Quadratic Average: 13.663558006286621
Nearest Class Center Accuracy: 0.9956

Output Layer:
Intra Cos: 0.9827357530593872
Inter Cos: 0.14884328842163086
Norm Quadratic Average: 10.78947639465332
Nearest Class Center Accuracy: 0.9956

