Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_846264_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: 0.09116754680871964
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567686080932617
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.05521067976951599
Inter Cos: 0.0715516209602356
Norm Quadratic Average: 2.577057123184204
Nearest Class Center Accuracy: 0.8081666666666667

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09996522963047028
Inter Cos: 0.09454239904880524
Norm Quadratic Average: 1.5438872575759888
Nearest Class Center Accuracy: 0.8717

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09999123215675354
Inter Cos: 0.09338304400444031
Norm Quadratic Average: 1.1954193115234375
Nearest Class Center Accuracy: 0.8784

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18152479827404022
Inter Cos: 0.1236116886138916
Norm Quadratic Average: 0.8464571237564087
Nearest Class Center Accuracy: 0.9348166666666666

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22938258945941925
Inter Cos: 0.1274135410785675
Norm Quadratic Average: 0.6548827290534973
Nearest Class Center Accuracy: 0.96325

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2879928648471832
Inter Cos: 0.13334691524505615
Norm Quadratic Average: 0.5405762195587158
Nearest Class Center Accuracy: 0.9755833333333334

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.34705522656440735
Inter Cos: 0.15692324936389923
Norm Quadratic Average: 0.49199479818344116
Nearest Class Center Accuracy: 0.9807833333333333

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4366839826107025
Inter Cos: 0.18064512312412262
Norm Quadratic Average: 0.3563573658466339
Nearest Class Center Accuracy: 0.99485

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6739358305931091
Inter Cos: 0.2510547637939453
Norm Quadratic Average: 0.2602645456790924
Nearest Class Center Accuracy: 0.9988666666666667

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8059471845626831
Inter Cos: 0.21946805715560913
Norm Quadratic Average: 0.22700533270835876
Nearest Class Center Accuracy: 0.99955

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8398864269256592
Inter Cos: 0.15548619627952576
Norm Quadratic Average: 0.21536268293857574
Nearest Class Center Accuracy: 0.9999666666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8865567445755005
Inter Cos: 0.0982121005654335
Norm Quadratic Average: 0.27991238236427307
Nearest Class Center Accuracy: 1.0

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9844729900360107
Inter Cos: -0.003286705818027258
Norm Quadratic Average: 0.3253857493400574
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9966627955436707
Inter Cos: -0.017004363238811493
Norm Quadratic Average: 0.5607596039772034
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9980734586715698
Inter Cos: -0.043735526502132416
Norm Quadratic Average: 1.0839382410049438
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.1966397762298584
Linear Weight Rank: 145
Intra Cos: 0.9990239143371582
Inter Cos: -0.019535422325134277
Norm Quadratic Average: 25.7247257232666
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.2032947540283203
Linear Weight Rank: 1499
Intra Cos: 0.9991410970687866
Inter Cos: 0.02911127172410488
Norm Quadratic Average: 17.59732437133789
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.1998205184936523
Linear Weight Rank: 9
Intra Cos: 0.9992076754570007
Inter Cos: 0.05443232133984566
Norm Quadratic Average: 12.329109191894531
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9993511438369751
Inter Cos: 0.07317867130041122
Norm Quadratic Average: 9.13273811340332
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.017716645275615155
Accuracy: 0.9961
NC1 Within Class Collapse: 0.08053772896528244
NC2 Equinorm: Features: 0.018199488520622253, Weights: 0.005655465181916952
NC2 Equiangle: Features: 0.06723895072937011, Weights: 0.028822580973307293
NC3 Self-Duality: 0.008899902924895287
NC4 NCC Mismatch: 0.0

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582170248032
Inter Cos: 0.12048851698637009
Norm Quadratic Average: 23.59519386291504
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.06315780431032181
Inter Cos: 0.073915995657444
Norm Quadratic Average: 2.566772699356079
Nearest Class Center Accuracy: 0.8195

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11005928367376328
Inter Cos: 0.09598858654499054
Norm Quadratic Average: 1.5333287715911865
Nearest Class Center Accuracy: 0.8816

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11003952473402023
Inter Cos: 0.09503007680177689
Norm Quadratic Average: 1.1916906833648682
Nearest Class Center Accuracy: 0.8876

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19345860183238983
Inter Cos: 0.13412219285964966
Norm Quadratic Average: 0.8433972001075745
Nearest Class Center Accuracy: 0.9419

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24446485936641693
Inter Cos: 0.13893026113510132
Norm Quadratic Average: 0.653867781162262
Nearest Class Center Accuracy: 0.9657

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.30365025997161865
Inter Cos: 0.14641810953617096
Norm Quadratic Average: 0.5401787161827087
Nearest Class Center Accuracy: 0.9748

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3623315095901489
Inter Cos: 0.17106881737709045
Norm Quadratic Average: 0.4908297657966614
Nearest Class Center Accuracy: 0.9801

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4498733580112457
Inter Cos: 0.1923278123140335
Norm Quadratic Average: 0.3552789092063904
Nearest Class Center Accuracy: 0.9908

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6822171211242676
Inter Cos: 0.26169171929359436
Norm Quadratic Average: 0.2598162591457367
Nearest Class Center Accuracy: 0.9941

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8103052377700806
Inter Cos: 0.2283395230770111
Norm Quadratic Average: 0.2268374264240265
Nearest Class Center Accuracy: 0.9949

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8413077592849731
Inter Cos: 0.16328275203704834
Norm Quadratic Average: 0.2150573432445526
Nearest Class Center Accuracy: 0.9952

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.881070077419281
Inter Cos: 0.09814675897359848
Norm Quadratic Average: 0.27922308444976807
Nearest Class Center Accuracy: 0.9955

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9750552177429199
Inter Cos: -0.009832760319113731
Norm Quadratic Average: 0.32437729835510254
Nearest Class Center Accuracy: 0.996

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9778461456298828
Inter Cos: -0.014643317088484764
Norm Quadratic Average: 0.558976948261261
Nearest Class Center Accuracy: 0.9959

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9795462489128113
Inter Cos: -0.041752345860004425
Norm Quadratic Average: 1.0805648565292358
Nearest Class Center Accuracy: 0.996

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.1966397762298584
Linear Weight Rank: 145
Intra Cos: 0.9812607169151306
Inter Cos: -0.00885077565908432
Norm Quadratic Average: 25.6451416015625
Nearest Class Center Accuracy: 0.996

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.2032947540283203
Linear Weight Rank: 1499
Intra Cos: 0.982241690158844
Inter Cos: 0.03896458074450493
Norm Quadratic Average: 17.543636322021484
Nearest Class Center Accuracy: 0.9961

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.1998205184936523
Linear Weight Rank: 9
Intra Cos: 0.9826995134353638
Inter Cos: 0.06397519260644913
Norm Quadratic Average: 12.292013168334961
Nearest Class Center Accuracy: 0.9961

Output Layer:
Intra Cos: 0.9835793375968933
Inter Cos: 0.08229833841323853
Norm Quadratic Average: 9.105499267578125
Nearest Class Center Accuracy: 0.9961

