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_0.005.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: 0.06102675199508667
Inter Cos: 0.07986453175544739
Norm Quadratic Average: 2.574183464050293
Nearest Class Center Accuracy: 0.8106

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10366223007440567
Inter Cos: 0.10152264684438705
Norm Quadratic Average: 1.4065560102462769
Nearest Class Center Accuracy: 0.8713666666666666

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09934885054826736
Inter Cos: 0.09760928899049759
Norm Quadratic Average: 1.0868539810180664
Nearest Class Center Accuracy: 0.8785833333333334

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1826523244380951
Inter Cos: 0.12515921890735626
Norm Quadratic Average: 0.6824647784233093
Nearest Class Center Accuracy: 0.9360333333333334

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24437828361988068
Inter Cos: 0.14529594779014587
Norm Quadratic Average: 0.48699867725372314
Nearest Class Center Accuracy: 0.9618333333333333

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3202973008155823
Inter Cos: 0.1592317521572113
Norm Quadratic Average: 0.39233121275901794
Nearest Class Center Accuracy: 0.9729666666666666

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3657270073890686
Inter Cos: 0.15186583995819092
Norm Quadratic Average: 0.356320321559906
Nearest Class Center Accuracy: 0.9780333333333333

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4143621325492859
Inter Cos: 0.17109917104244232
Norm Quadratic Average: 0.2194570004940033
Nearest Class Center Accuracy: 0.9935

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6361872553825378
Inter Cos: 0.2393530309200287
Norm Quadratic Average: 0.1404947191476822
Nearest Class Center Accuracy: 0.9987

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8311023116111755
Inter Cos: 0.30771586298942566
Norm Quadratic Average: 0.13764318823814392
Nearest Class Center Accuracy: 0.9999333333333333

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8792455196380615
Inter Cos: 0.27039098739624023
Norm Quadratic Average: 0.16156978905200958
Nearest Class Center Accuracy: 0.9999666666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9288680553436279
Inter Cos: 0.16276079416275024
Norm Quadratic Average: 0.20462706685066223
Nearest Class Center Accuracy: 0.9999833333333333

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9844647645950317
Inter Cos: 0.14030413329601288
Norm Quadratic Average: 0.25848957896232605
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9952760338783264
Inter Cos: 0.06369238346815109
Norm Quadratic Average: 0.5234771370887756
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9981405138969421
Inter Cos: 0.1423487365245819
Norm Quadratic Average: 1.096859335899353
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.131274938583374
Linear Weight Rank: 10
Intra Cos: 0.9987848401069641
Inter Cos: 0.2293437123298645
Norm Quadratic Average: 24.667842864990234
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.1335957050323486
Linear Weight Rank: 1394
Intra Cos: 0.999154806137085
Inter Cos: 0.2210220843553543
Norm Quadratic Average: 17.23933219909668
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.133876323699951
Linear Weight Rank: 9
Intra Cos: 0.9993337988853455
Inter Cos: 0.18912747502326965
Norm Quadratic Average: 12.328205108642578
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9995203614234924
Inter Cos: 0.11799423396587372
Norm Quadratic Average: 9.295479774475098
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.01950403793361038
Accuracy: 0.9956
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.026342226192355156, Weights: 0.00810367800295353
NC2 Equiangle: Features: 0.11205247243245443, Weights: 0.08972805870903863
NC3 Self-Duality: 0.03264426440000534
NC4 NCC Mismatch: 9.999999999998899e-05

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: 0.0694013461470604
Inter Cos: 0.0825972855091095
Norm Quadratic Average: 2.5651695728302
Nearest Class Center Accuracy: 0.8199

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11403646320104599
Inter Cos: 0.10286536812782288
Norm Quadratic Average: 1.3974262475967407
Nearest Class Center Accuracy: 0.8839

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10902746766805649
Inter Cos: 0.09917107969522476
Norm Quadratic Average: 1.0832685232162476
Nearest Class Center Accuracy: 0.8857

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19469407200813293
Inter Cos: 0.13229866325855255
Norm Quadratic Average: 0.6800082921981812
Nearest Class Center Accuracy: 0.9414

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25973525643348694
Inter Cos: 0.14195823669433594
Norm Quadratic Average: 0.4861500859260559
Nearest Class Center Accuracy: 0.9616

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3352081775665283
Inter Cos: 0.1536102443933487
Norm Quadratic Average: 0.3917776942253113
Nearest Class Center Accuracy: 0.9728

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.37969255447387695
Inter Cos: 0.16042248904705048
Norm Quadratic Average: 0.35570982098579407
Nearest Class Center Accuracy: 0.9774

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4262593686580658
Inter Cos: 0.1852126121520996
Norm Quadratic Average: 0.2192496359348297
Nearest Class Center Accuracy: 0.9889

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6453902125358582
Inter Cos: 0.252062052488327
Norm Quadratic Average: 0.14071311056613922
Nearest Class Center Accuracy: 0.9921

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8235688209533691
Inter Cos: 0.319411039352417
Norm Quadratic Average: 0.1381089836359024
Nearest Class Center Accuracy: 0.9948

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8803748488426208
Inter Cos: 0.28094542026519775
Norm Quadratic Average: 0.16154025495052338
Nearest Class Center Accuracy: 0.995

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.922431468963623
Inter Cos: 0.17274923622608185
Norm Quadratic Average: 0.2043209969997406
Nearest Class Center Accuracy: 0.9956

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9738180637359619
Inter Cos: 0.15175826847553253
Norm Quadratic Average: 0.25773516297340393
Nearest Class Center Accuracy: 0.9955

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9768450260162354
Inter Cos: 0.07124630361795425
Norm Quadratic Average: 0.5219014883041382
Nearest Class Center Accuracy: 0.9956

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9779899716377258
Inter Cos: 0.14757172763347626
Norm Quadratic Average: 1.093327283859253
Nearest Class Center Accuracy: 0.9955

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.131274938583374
Linear Weight Rank: 10
Intra Cos: 0.979221522808075
Inter Cos: 0.23142388463020325
Norm Quadratic Average: 24.592193603515625
Nearest Class Center Accuracy: 0.9955

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.1335957050323486
Linear Weight Rank: 1394
Intra Cos: 0.9797576069831848
Inter Cos: 0.2231566160917282
Norm Quadratic Average: 17.181848526000977
Nearest Class Center Accuracy: 0.9955

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.133876323699951
Linear Weight Rank: 9
Intra Cos: 0.9798228740692139
Inter Cos: 0.19200454652309418
Norm Quadratic Average: 12.28378677368164
Nearest Class Center Accuracy: 0.9955

Output Layer:
Intra Cos: 0.9800796508789062
Inter Cos: 0.12943297624588013
Norm Quadratic Average: 9.259176254272461
Nearest Class Center Accuracy: 0.9955

