Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.02.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: 0.12534373998641968
Inter Cos: 0.1581660807132721
Norm Quadratic Average: 37.025596618652344
Nearest Class Center Accuracy: 0.7987333333333333

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15201951563358307
Inter Cos: 0.19289125502109528
Norm Quadratic Average: 41.68488311767578
Nearest Class Center Accuracy: 0.7684

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1872374713420868
Inter Cos: 0.2217639535665512
Norm Quadratic Average: 51.543277740478516
Nearest Class Center Accuracy: 0.78775

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16678467392921448
Inter Cos: 0.24475599825382233
Norm Quadratic Average: 31.73051643371582
Nearest Class Center Accuracy: 0.8323333333333334

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22360508143901825
Inter Cos: 0.3049372136592865
Norm Quadratic Average: 19.378694534301758
Nearest Class Center Accuracy: 0.8857166666666667

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.39796432852745056
Inter Cos: 0.36322203278541565
Norm Quadratic Average: 9.680412292480469
Nearest Class Center Accuracy: 0.9248333333333333

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5539060235023499
Inter Cos: 0.4199119806289673
Norm Quadratic Average: 8.8517427444458
Nearest Class Center Accuracy: 0.9506

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.6508821249008179
Linear Weight Rank: 5
Intra Cos: 0.7039795517921448
Inter Cos: 0.40938466787338257
Norm Quadratic Average: 41.49595260620117
Nearest Class Center Accuracy: 0.97395

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.6521400213241577
Linear Weight Rank: 2471
Intra Cos: 0.7791604995727539
Inter Cos: 0.43283510208129883
Norm Quadratic Average: 30.620817184448242
Nearest Class Center Accuracy: 0.9785

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.6486856937408447
Linear Weight Rank: 9
Intra Cos: 0.8115549683570862
Inter Cos: 0.378097802400589
Norm Quadratic Average: 20.595796585083008
Nearest Class Center Accuracy: 0.9798166666666667

Output Layer:
Intra Cos: 0.8438831567764282
Inter Cos: 0.40150055289268494
Norm Quadratic Average: 15.141704559326172
Nearest Class Center Accuracy: 0.9796166666666667

Test Set:
Average Loss: 0.06292694781720638
Accuracy: 0.9818
NC1 Within Class Collapse: 1.7222323417663574
NC2 Equinorm: Features: 0.11127742379903793, Weights: 0.039012689143419266
NC2 Equiangle: Features: 0.32713678148057723, Weights: 0.25148421393500436
NC3 Self-Duality: 0.07791157066822052
NC4 NCC Mismatch: 0.009199999999999986

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: 0.13878706097602844
Inter Cos: 0.173127219080925
Norm Quadratic Average: 37.122432708740234
Nearest Class Center Accuracy: 0.8156

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16843360662460327
Inter Cos: 0.21265552937984467
Norm Quadratic Average: 41.69881057739258
Nearest Class Center Accuracy: 0.788

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20315904915332794
Inter Cos: 0.2427513748407364
Norm Quadratic Average: 51.55695343017578
Nearest Class Center Accuracy: 0.8058

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17796079814434052
Inter Cos: 0.244077667593956
Norm Quadratic Average: 31.69179916381836
Nearest Class Center Accuracy: 0.8514

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2356208711862564
Inter Cos: 0.3025056719779968
Norm Quadratic Average: 19.383560180664062
Nearest Class Center Accuracy: 0.9009

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.407910019159317
Inter Cos: 0.35537004470825195
Norm Quadratic Average: 9.737896919250488
Nearest Class Center Accuracy: 0.9287

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5574998259544373
Inter Cos: 0.4364462196826935
Norm Quadratic Average: 8.945279121398926
Nearest Class Center Accuracy: 0.9503

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.6508821249008179
Linear Weight Rank: 5
Intra Cos: 0.698174774646759
Inter Cos: 0.42498692870140076
Norm Quadratic Average: 42.088619232177734
Nearest Class Center Accuracy: 0.974

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.6521400213241577
Linear Weight Rank: 2471
Intra Cos: 0.7685356736183167
Inter Cos: 0.45192456245422363
Norm Quadratic Average: 31.107667922973633
Nearest Class Center Accuracy: 0.978

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.6486856937408447
Linear Weight Rank: 9
Intra Cos: 0.8007848262786865
Inter Cos: 0.39470261335372925
Norm Quadratic Average: 20.9229679107666
Nearest Class Center Accuracy: 0.9799

Output Layer:
Intra Cos: 0.8315893411636353
Inter Cos: 0.41765454411506653
Norm Quadratic Average: 15.391011238098145
Nearest Class Center Accuracy: 0.9809

