Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_314159_test_samples_None_train_samples_None_weight_decay_0.0001.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09116753935813904
Inter Cos: 0.10967151075601578
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: 0.08328370004892349
Inter Cos: 0.10290471464395523
Norm Quadratic Average: 58.39792251586914
Nearest Class Center Accuracy: 0.8176166666666667

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12321485579013824
Inter Cos: 0.12641742825508118
Norm Quadratic Average: 54.89103698730469
Nearest Class Center Accuracy: 0.854

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12585672736167908
Inter Cos: 0.12974469363689423
Norm Quadratic Average: 74.02613830566406
Nearest Class Center Accuracy: 0.8596333333333334

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19911204278469086
Inter Cos: 0.1609581708908081
Norm Quadratic Average: 53.1900520324707
Nearest Class Center Accuracy: 0.9170333333333334

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22666673362255096
Inter Cos: 0.1624070256948471
Norm Quadratic Average: 51.42737579345703
Nearest Class Center Accuracy: 0.9409

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25917848944664
Inter Cos: 0.15630300343036652
Norm Quadratic Average: 46.29342269897461
Nearest Class Center Accuracy: 0.9554333333333334

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2924486994743347
Inter Cos: 0.17228879034519196
Norm Quadratic Average: 41.53569793701172
Nearest Class Center Accuracy: 0.9643666666666667

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3414733111858368
Inter Cos: 0.16217432916164398
Norm Quadratic Average: 20.2619571685791
Nearest Class Center Accuracy: 0.9846666666666667

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4633413553237915
Inter Cos: 0.1930263787508011
Norm Quadratic Average: 16.102676391601562
Nearest Class Center Accuracy: 0.9927833333333334

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5695356130599976
Inter Cos: 0.22586481273174286
Norm Quadratic Average: 14.483175277709961
Nearest Class Center Accuracy: 0.9959166666666667

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6530041098594666
Inter Cos: 0.22639353573322296
Norm Quadratic Average: 13.533649444580078
Nearest Class Center Accuracy: 0.99745

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.724889874458313
Inter Cos: 0.15978460013866425
Norm Quadratic Average: 8.50601863861084
Nearest Class Center Accuracy: 0.9977333333333334

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8757745027542114
Inter Cos: 0.17942401766777039
Norm Quadratic Average: 7.517477035522461
Nearest Class Center Accuracy: 0.9986333333333334

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.921288251876831
Inter Cos: 0.2189556509256363
Norm Quadratic Average: 6.93848180770874
Nearest Class Center Accuracy: 0.9988833333333333

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9378916025161743
Inter Cos: 0.28584015369415283
Norm Quadratic Average: 6.5181097984313965
Nearest Class Center Accuracy: 0.9991333333333333

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.78863525390625
Linear Weight Rank: 4031
Intra Cos: 0.9534989595413208
Inter Cos: 0.27734237909317017
Norm Quadratic Average: 38.53137969970703
Nearest Class Center Accuracy: 0.99955

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.905441284179688
Linear Weight Rank: 3670
Intra Cos: 0.9560568332672119
Inter Cos: 0.2728482186794281
Norm Quadratic Average: 31.76215934753418
Nearest Class Center Accuracy: 0.9996833333333334

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.912264823913574
Linear Weight Rank: 10
Intra Cos: 0.9571341276168823
Inter Cos: 0.26720142364501953
Norm Quadratic Average: 28.343524932861328
Nearest Class Center Accuracy: 0.9996833333333334

Output Layer:
Intra Cos: 0.9850223660469055
Inter Cos: 0.38782429695129395
Norm Quadratic Average: 28.096221923828125
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.025203063978233284
Accuracy: 0.9945
NC1 Within Class Collapse: 0.4634290337562561
NC2 Equinorm: Features: 0.12079733610153198, Weights: 0.04273286834359169
NC2 Equiangle: Features: 0.22074411180284287, Weights: 0.12706960042317708
NC3 Self-Duality: 0.3083203434944153
NC4 NCC Mismatch: 0.0033999999999999586

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
Inter Cos: 0.12048851698637009
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: 0.09341645240783691
Inter Cos: 0.10874360054731369
Norm Quadratic Average: 58.417911529541016
Nearest Class Center Accuracy: 0.8305

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13618530333042145
Inter Cos: 0.13885167241096497
Norm Quadratic Average: 54.733699798583984
Nearest Class Center Accuracy: 0.8661

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13814200460910797
Inter Cos: 0.1425432711839676
Norm Quadratic Average: 73.90528106689453
Nearest Class Center Accuracy: 0.8707

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21569962799549103
Inter Cos: 0.1750672608613968
Norm Quadratic Average: 53.0711669921875
Nearest Class Center Accuracy: 0.9272

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24467718601226807
Inter Cos: 0.1759660392999649
Norm Quadratic Average: 51.38926315307617
Nearest Class Center Accuracy: 0.9491

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.27810660004615784
Inter Cos: 0.1700402796268463
Norm Quadratic Average: 46.29224395751953
Nearest Class Center Accuracy: 0.9608

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3095445930957794
Inter Cos: 0.1869165599346161
Norm Quadratic Average: 41.56822204589844
Nearest Class Center Accuracy: 0.9686

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3582388460636139
Inter Cos: 0.16590411961078644
Norm Quadratic Average: 20.30506706237793
Nearest Class Center Accuracy: 0.9846

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4781213402748108
Inter Cos: 0.2024347186088562
Norm Quadratic Average: 16.163997650146484
Nearest Class Center Accuracy: 0.99

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5811662673950195
Inter Cos: 0.24709105491638184
Norm Quadratic Average: 14.56084156036377
Nearest Class Center Accuracy: 0.9909

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6619163155555725
Inter Cos: 0.24802711606025696
Norm Quadratic Average: 13.620928764343262
Nearest Class Center Accuracy: 0.9916

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7293993234634399
Inter Cos: 0.16288037598133087
Norm Quadratic Average: 8.573628425598145
Nearest Class Center Accuracy: 0.9916

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8739356994628906
Inter Cos: 0.1805737018585205
Norm Quadratic Average: 7.580699443817139
Nearest Class Center Accuracy: 0.9922

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9168980717658997
Inter Cos: 0.20878642797470093
Norm Quadratic Average: 6.994537353515625
Nearest Class Center Accuracy: 0.9923

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9328904151916504
Inter Cos: 0.2751944959163666
Norm Quadratic Average: 6.567235469818115
Nearest Class Center Accuracy: 0.9922

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.78863525390625
Linear Weight Rank: 4031
Intra Cos: 0.9458492398262024
Inter Cos: 0.2669842541217804
Norm Quadratic Average: 38.80146026611328
Nearest Class Center Accuracy: 0.9929

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.905441284179688
Linear Weight Rank: 3670
Intra Cos: 0.9481934905052185
Inter Cos: 0.26226529479026794
Norm Quadratic Average: 31.981409072875977
Nearest Class Center Accuracy: 0.993

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.912264823913574
Linear Weight Rank: 10
Intra Cos: 0.9519025683403015
Inter Cos: 0.25721275806427
Norm Quadratic Average: 28.539112091064453
Nearest Class Center Accuracy: 0.993

Output Layer:
Intra Cos: 0.9761412739753723
Inter Cos: 0.3861250877380371
Norm Quadratic Average: 28.280330657958984
Nearest Class Center Accuracy: 0.9942

