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_0.001.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: 0.11960361152887344
Inter Cos: 0.1418500542640686
Norm Quadratic Average: 39.70741271972656
Nearest Class Center Accuracy: 0.8167166666666666

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1878657191991806
Inter Cos: 0.1740090548992157
Norm Quadratic Average: 40.0329704284668
Nearest Class Center Accuracy: 0.8514666666666667

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22014857828617096
Inter Cos: 0.19414925575256348
Norm Quadratic Average: 37.69197463989258
Nearest Class Center Accuracy: 0.8956333333333333

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2342018187046051
Inter Cos: 0.17581529915332794
Norm Quadratic Average: 17.098644256591797
Nearest Class Center Accuracy: 0.93925

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3765473961830139
Inter Cos: 0.261584609746933
Norm Quadratic Average: 9.91656494140625
Nearest Class Center Accuracy: 0.96575

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5653692483901978
Inter Cos: 0.31861424446105957
Norm Quadratic Average: 5.859135627746582
Nearest Class Center Accuracy: 0.9889166666666667

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7851366400718689
Inter Cos: 0.39439377188682556
Norm Quadratic Average: 5.377895355224609
Nearest Class Center Accuracy: 0.9965166666666667

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.1520988941192627
Linear Weight Rank: 4028
Intra Cos: 0.8633216619491577
Inter Cos: 0.3015262484550476
Norm Quadratic Average: 28.207151412963867
Nearest Class Center Accuracy: 0.9981333333333333

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.554718494415283
Linear Weight Rank: 3639
Intra Cos: 0.9068521857261658
Inter Cos: 0.24866634607315063
Norm Quadratic Average: 25.078027725219727
Nearest Class Center Accuracy: 0.9993166666666666

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.3579907417297363
Linear Weight Rank: 9
Intra Cos: 0.9155690670013428
Inter Cos: 0.22714565694332123
Norm Quadratic Average: 21.502822875976562
Nearest Class Center Accuracy: 0.99975

Output Layer:
Intra Cos: 0.9400017857551575
Inter Cos: 0.2805067300796509
Norm Quadratic Average: 20.82320785522461
Nearest Class Center Accuracy: 0.9999

Test Set:
Average Loss: 0.02303700282125501
Accuracy: 0.9935
NC1 Within Class Collapse: 1.1319788694381714
NC2 Equinorm: Features: 0.10621080547571182, Weights: 0.04802461713552475
NC2 Equiangle: Features: 0.24166929456922742, Weights: 0.17998814053005643
NC3 Self-Duality: 0.08778547495603561
NC4 NCC Mismatch: 0.0033999999999999586

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: 0.13301347196102142
Inter Cos: 0.155237078666687
Norm Quadratic Average: 39.68806457519531
Nearest Class Center Accuracy: 0.8313

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20174860954284668
Inter Cos: 0.1793937385082245
Norm Quadratic Average: 39.910133361816406
Nearest Class Center Accuracy: 0.8704

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2347547858953476
Inter Cos: 0.20085924863815308
Norm Quadratic Average: 37.62131881713867
Nearest Class Center Accuracy: 0.9073

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24744568765163422
Inter Cos: 0.18983088433742523
Norm Quadratic Average: 17.08071517944336
Nearest Class Center Accuracy: 0.9501

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.39092788100242615
Inter Cos: 0.28071215748786926
Norm Quadratic Average: 9.930651664733887
Nearest Class Center Accuracy: 0.9699

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5731189250946045
Inter Cos: 0.3390423059463501
Norm Quadratic Average: 5.8933305740356445
Nearest Class Center Accuracy: 0.9855

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7837598323822021
Inter Cos: 0.40740880370140076
Norm Quadratic Average: 5.426527500152588
Nearest Class Center Accuracy: 0.9895

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.1520988941192627
Linear Weight Rank: 4028
Intra Cos: 0.8574987649917603
Inter Cos: 0.3073754906654358
Norm Quadratic Average: 28.44512176513672
Nearest Class Center Accuracy: 0.9899

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.554718494415283
Linear Weight Rank: 3639
Intra Cos: 0.899788498878479
Inter Cos: 0.2668566405773163
Norm Quadratic Average: 25.2889461517334
Nearest Class Center Accuracy: 0.9914

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.3579907417297363
Linear Weight Rank: 9
Intra Cos: 0.9076353907585144
Inter Cos: 0.2455880492925644
Norm Quadratic Average: 21.683332443237305
Nearest Class Center Accuracy: 0.9923

Output Layer:
Intra Cos: 0.9288054704666138
Inter Cos: 0.2844846546649933
Norm Quadratic Average: 20.99685287475586
Nearest Class Center Accuracy: 0.9929

