Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_979323_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: 0.10967151820659637
Norm Quadratic Average: 23.567676544189453
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09781768172979355
Inter Cos: 0.11365357786417007
Norm Quadratic Average: 60.40780258178711
Nearest Class Center Accuracy: 0.8403666666666667

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16909965872764587
Inter Cos: 0.1313202679157257
Norm Quadratic Average: 42.53044509887695
Nearest Class Center Accuracy: 0.89625

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19335375726222992
Inter Cos: 0.13142739236354828
Norm Quadratic Average: 40.69278335571289
Nearest Class Center Accuracy: 0.9247166666666666

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25275611877441406
Inter Cos: 0.10940626263618469
Norm Quadratic Average: 26.893789291381836
Nearest Class Center Accuracy: 0.9750166666666666

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.34048140048980713
Inter Cos: 0.10032007098197937
Norm Quadratic Average: 29.00954246520996
Nearest Class Center Accuracy: 0.99085

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4700040817260742
Inter Cos: 0.1436397284269333
Norm Quadratic Average: 22.380006790161133
Nearest Class Center Accuracy: 0.9991

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7101598381996155
Inter Cos: 0.15718552470207214
Norm Quadratic Average: 17.24327278137207
Nearest Class Center Accuracy: 0.9999833333333333

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.84241485595703
Linear Weight Rank: 4031
Intra Cos: 0.8790872097015381
Inter Cos: 0.11784295737743378
Norm Quadratic Average: 118.71342468261719
Nearest Class Center Accuracy: 0.9999833333333333

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.990203857421875
Linear Weight Rank: 3670
Intra Cos: 0.9748396873474121
Inter Cos: 0.01045949012041092
Norm Quadratic Average: 68.95858764648438
Nearest Class Center Accuracy: 1.0

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.5836904048919678
Linear Weight Rank: 10
Intra Cos: 0.9756700396537781
Inter Cos: 0.026257479563355446
Norm Quadratic Average: 33.992210388183594
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9971718192100525
Inter Cos: 0.17429529130458832
Norm Quadratic Average: 20.699094772338867
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.022833374788879836
Accuracy: 0.9952
NC1 Within Class Collapse: 0.3401952385902405
NC2 Equinorm: Features: 0.07302039861679077, Weights: 0.0722924992442131
NC2 Equiangle: Features: 0.06369217766655816, Weights: 0.07361470858256022
NC3 Self-Duality: 0.6363006234169006
NC4 NCC Mismatch: 0.0009000000000000119

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: 23.595197677612305
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10748178511857986
Inter Cos: 0.11356423795223236
Norm Quadratic Average: 60.07197952270508
Nearest Class Center Accuracy: 0.8532

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18106107413768768
Inter Cos: 0.1293915957212448
Norm Quadratic Average: 42.241607666015625
Nearest Class Center Accuracy: 0.9078

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20841564238071442
Inter Cos: 0.12958021461963654
Norm Quadratic Average: 40.43737030029297
Nearest Class Center Accuracy: 0.9318

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2651204466819763
Inter Cos: 0.1041911393404007
Norm Quadratic Average: 26.757709503173828
Nearest Class Center Accuracy: 0.9752

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.35370486974716187
Inter Cos: 0.11051100492477417
Norm Quadratic Average: 28.91100311279297
Nearest Class Center Accuracy: 0.9871

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.47726795077323914
Inter Cos: 0.1509924679994583
Norm Quadratic Average: 22.35148811340332
Nearest Class Center Accuracy: 0.9924

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7088902592658997
Inter Cos: 0.1637723445892334
Norm Quadratic Average: 17.245609283447266
Nearest Class Center Accuracy: 0.9942

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.84241485595703
Linear Weight Rank: 4031
Intra Cos: 0.8716598153114319
Inter Cos: 0.12155580520629883
Norm Quadratic Average: 118.86155700683594
Nearest Class Center Accuracy: 0.9937

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.990203857421875
Linear Weight Rank: 3670
Intra Cos: 0.9577378034591675
Inter Cos: 0.021140428259968758
Norm Quadratic Average: 69.05400085449219
Nearest Class Center Accuracy: 0.9944

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.5836904048919678
Linear Weight Rank: 10
Intra Cos: 0.954721987247467
Inter Cos: 0.03634856641292572
Norm Quadratic Average: 34.04730987548828
Nearest Class Center Accuracy: 0.9949

Output Layer:
Intra Cos: 0.9835768342018127
Inter Cos: 0.18441928923130035
Norm Quadratic Average: 20.712060928344727
Nearest Class Center Accuracy: 0.9953

