Model save path: ./New_Models/bn_False_dataset_CIFAR10_epochs_200_lr_0.01_model_type_vgg19_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.005.pth.tar
Training Set:
Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0198909193277359
Inter Cos: 0.10477277636528015
Norm Quadratic Average: 27.597179412841797
Nearest Class Center Accuracy: 0.26968

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.020691117271780968
Inter Cos: 0.09169657528400421
Norm Quadratic Average: 30.589900970458984
Nearest Class Center Accuracy: 0.33492

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02444772981107235
Inter Cos: 0.09667439013719559
Norm Quadratic Average: 24.888608932495117
Nearest Class Center Accuracy: 0.43626

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.028084643185138702
Inter Cos: 0.08322826772928238
Norm Quadratic Average: 23.873014450073242
Nearest Class Center Accuracy: 0.51838

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.021769389510154724
Inter Cos: 0.06310126930475235
Norm Quadratic Average: 8.980867385864258
Nearest Class Center Accuracy: 0.59974

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.029270075261592865
Inter Cos: 0.058900848031044006
Norm Quadratic Average: 3.001297950744629
Nearest Class Center Accuracy: 0.64888

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.08888561278581619
Inter Cos: 0.13544978201389313
Norm Quadratic Average: 0.896576464176178
Nearest Class Center Accuracy: 0.74006

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1775052845478058
Inter Cos: 0.24511271715164185
Norm Quadratic Average: 0.5315032601356506
Nearest Class Center Accuracy: 0.82634

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3008052110671997
Inter Cos: 0.34963464736938477
Norm Quadratic Average: 0.33527088165283203
Nearest Class Center Accuracy: 0.86348

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.47992727160453796
Inter Cos: 0.5371252298355103
Norm Quadratic Average: 0.31207019090652466
Nearest Class Center Accuracy: 0.9306

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5601944327354431
Inter Cos: 0.6082112193107605
Norm Quadratic Average: 0.45015937089920044
Nearest Class Center Accuracy: 0.9525

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6242194175720215
Inter Cos: 0.6170563697814941
Norm Quadratic Average: 0.7275891900062561
Nearest Class Center Accuracy: 0.9648

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7522394061088562
Inter Cos: 0.6677065491676331
Norm Quadratic Average: 0.8630557060241699
Nearest Class Center Accuracy: 0.97502

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8314146995544434
Inter Cos: 0.6852700114250183
Norm Quadratic Average: 1.3261884450912476
Nearest Class Center Accuracy: 0.98564

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8403779864311218
Inter Cos: 0.6251162886619568
Norm Quadratic Average: 2.2068183422088623
Nearest Class Center Accuracy: 0.9928

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8376511335372925
Inter Cos: 0.5988720655441284
Norm Quadratic Average: 3.6642515659332275
Nearest Class Center Accuracy: 0.99618

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.2829058170318604
Linear Weight Rank: 43
Intra Cos: 0.8441771268844604
Inter Cos: 0.5996007323265076
Norm Quadratic Average: 23.783824920654297
Nearest Class Center Accuracy: 0.99836

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.313793659210205
Linear Weight Rank: 2788
Intra Cos: 0.8529903888702393
Inter Cos: 0.5469256639480591
Norm Quadratic Average: 22.718584060668945
Nearest Class Center Accuracy: 0.99936

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.287198066711426
Linear Weight Rank: 9
Intra Cos: 0.8485076427459717
Inter Cos: 0.43177133798599243
Norm Quadratic Average: 18.968046188354492
Nearest Class Center Accuracy: 0.99988

Output Layer:
Intra Cos: 0.8714192509651184
Inter Cos: 0.4277881979942322
Norm Quadratic Average: 17.242809295654297
Nearest Class Center Accuracy: 0.99998

Test Set:
Average Loss: 0.7433887726783752
Accuracy: 0.8322
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.20173554122447968, Weights: 0.06565604358911514
NC2 Equiangle: Features: 0.3300570170084635, Weights: 0.17186516655815973
NC3 Self-Duality: 0.2230946123600006
NC4 NCC Mismatch: 0.04179999999999995

Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.018526077270507812
Inter Cos: 0.1042705774307251
Norm Quadratic Average: 27.550132751464844
Nearest Class Center Accuracy: 0.284

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.017812956124544144
Inter Cos: 0.09205901622772217
Norm Quadratic Average: 30.559856414794922
Nearest Class Center Accuracy: 0.3512

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02359602227807045
Inter Cos: 0.09727422893047333
Norm Quadratic Average: 24.888408660888672
Nearest Class Center Accuracy: 0.4524

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02627003751695156
Inter Cos: 0.08386021107435226
Norm Quadratic Average: 23.88698387145996
Nearest Class Center Accuracy: 0.5291

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.020031537860631943
Inter Cos: 0.06367634236812592
Norm Quadratic Average: 8.993470191955566
Nearest Class Center Accuracy: 0.6013

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.026992658153176308
Inter Cos: 0.058847397565841675
Norm Quadratic Average: 3.0071821212768555
Nearest Class Center Accuracy: 0.647

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.08382131904363632
Inter Cos: 0.13843291997909546
Norm Quadratic Average: 0.8964894413948059
Nearest Class Center Accuracy: 0.7137

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16487973928451538
Inter Cos: 0.24909743666648865
Norm Quadratic Average: 0.5299656391143799
Nearest Class Center Accuracy: 0.7692

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26391613483428955
Inter Cos: 0.35233795642852783
Norm Quadratic Average: 0.3330078721046448
Nearest Class Center Accuracy: 0.7769

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3839110732078552
Inter Cos: 0.49198734760284424
Norm Quadratic Average: 0.3085123598575592
Nearest Class Center Accuracy: 0.7989

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.42344850301742554
Inter Cos: 0.5504851341247559
Norm Quadratic Average: 0.4442666471004486
Nearest Class Center Accuracy: 0.8039

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4601331651210785
Inter Cos: 0.558246374130249
Norm Quadratic Average: 0.7172775864601135
Nearest Class Center Accuracy: 0.8035

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5685338973999023
Inter Cos: 0.6179791688919067
Norm Quadratic Average: 0.849943995475769
Nearest Class Center Accuracy: 0.8027

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.619140625
Inter Cos: 0.6396551132202148
Norm Quadratic Average: 1.3046590089797974
Nearest Class Center Accuracy: 0.8127

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5918813943862915
Inter Cos: 0.6144475340843201
Norm Quadratic Average: 2.1692888736724854
Nearest Class Center Accuracy: 0.8197

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.558249831199646
Inter Cos: 0.5940030217170715
Norm Quadratic Average: 3.5963830947875977
Nearest Class Center Accuracy: 0.8265

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.2829058170318604
Linear Weight Rank: 43
Intra Cos: 0.5403090119361877
Inter Cos: 0.5959168076515198
Norm Quadratic Average: 23.318397521972656
Nearest Class Center Accuracy: 0.8265

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.313793659210205
Linear Weight Rank: 2788
Intra Cos: 0.5364765524864197
Inter Cos: 0.5539035797119141
Norm Quadratic Average: 22.259157180786133
Nearest Class Center Accuracy: 0.8275

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.287198066711426
Linear Weight Rank: 9
Intra Cos: 0.5366513133049011
Inter Cos: 0.45795753598213196
Norm Quadratic Average: 18.55916404724121
Nearest Class Center Accuracy: 0.8289

Output Layer:
Intra Cos: 0.5239882469177246
Inter Cos: 0.45879414677619934
Norm Quadratic Average: 16.82626724243164
Nearest Class Center Accuracy: 0.8287

