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_0.01.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: 0.05959192290902138
Inter Cos: 0.07992494851350784
Norm Quadratic Average: 2.412020206451416
Nearest Class Center Accuracy: 0.8084

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10393951833248138
Inter Cos: 0.10044167190790176
Norm Quadratic Average: 1.3649011850357056
Nearest Class Center Accuracy: 0.8745

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10125618427991867
Inter Cos: 0.10012346506118774
Norm Quadratic Average: 1.0847331285476685
Nearest Class Center Accuracy: 0.8787833333333334

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18188786506652832
Inter Cos: 0.13038675487041473
Norm Quadratic Average: 0.6782410144805908
Nearest Class Center Accuracy: 0.9344

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22937242686748505
Inter Cos: 0.1356283575296402
Norm Quadratic Average: 0.49942001700401306
Nearest Class Center Accuracy: 0.95895

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.31313154101371765
Inter Cos: 0.15340685844421387
Norm Quadratic Average: 0.4140895903110504
Nearest Class Center Accuracy: 0.9728666666666667

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3670526146888733
Inter Cos: 0.14732126891613007
Norm Quadratic Average: 0.3551417589187622
Nearest Class Center Accuracy: 0.97675

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4317186772823334
Inter Cos: 0.17319190502166748
Norm Quadratic Average: 0.2104717493057251
Nearest Class Center Accuracy: 0.9933333333333333

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6591373682022095
Inter Cos: 0.221165269613266
Norm Quadratic Average: 0.13999296724796295
Nearest Class Center Accuracy: 0.9978333333333333

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7936064600944519
Inter Cos: 0.30416277050971985
Norm Quadratic Average: 0.13119055330753326
Nearest Class Center Accuracy: 0.9989833333333333

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8470082879066467
Inter Cos: 0.24070002138614655
Norm Quadratic Average: 0.14217637479305267
Nearest Class Center Accuracy: 0.9999666666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9105743169784546
Inter Cos: 0.14364971220493317
Norm Quadratic Average: 0.13370612263679504
Nearest Class Center Accuracy: 0.9999666666666667

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.980601966381073
Inter Cos: 0.2331458330154419
Norm Quadratic Average: 0.16552267968654633
Nearest Class Center Accuracy: 0.9999666666666667

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9962692260742188
Inter Cos: 0.19912000000476837
Norm Quadratic Average: 0.39712002873420715
Nearest Class Center Accuracy: 0.9999666666666667

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9986094832420349
Inter Cos: 0.16546137630939484
Norm Quadratic Average: 1.005921721458435
Nearest Class Center Accuracy: 0.9999666666666667

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.055389165878296
Linear Weight Rank: 9
Intra Cos: 0.999055802822113
Inter Cos: 0.2550342082977295
Norm Quadratic Average: 24.470779418945312
Nearest Class Center Accuracy: 0.9999666666666667

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.0567944049835205
Linear Weight Rank: 1492
Intra Cos: 0.9993054866790771
Inter Cos: 0.2392912209033966
Norm Quadratic Average: 16.651891708374023
Nearest Class Center Accuracy: 0.9999666666666667

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.0579159259796143
Linear Weight Rank: 9
Intra Cos: 0.9994457960128784
Inter Cos: 0.20112961530685425
Norm Quadratic Average: 11.557868003845215
Nearest Class Center Accuracy: 0.9999666666666667

Output Layer:
Intra Cos: 0.999514639377594
Inter Cos: 0.13751430809497833
Norm Quadratic Average: 8.466974258422852
Nearest Class Center Accuracy: 0.9999666666666667

Test Set:
Average Loss: 0.022491552134975792
Accuracy: 0.9948
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.024836698547005653, Weights: 0.010938928462564945
NC2 Equiangle: Features: 0.12175335354275174, Weights: 0.09596233367919922
NC3 Self-Duality: 0.0335334911942482
NC4 NCC Mismatch: 0.00019999999999997797

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: 0.06794969737529755
Inter Cos: 0.08263642340898514
Norm Quadratic Average: 2.4028682708740234
Nearest Class Center Accuracy: 0.8173

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11394605785608292
Inter Cos: 0.10231386125087738
Norm Quadratic Average: 1.3555611371994019
Nearest Class Center Accuracy: 0.8865

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11146066337823868
Inter Cos: 0.10253316909074783
Norm Quadratic Average: 1.0817806720733643
Nearest Class Center Accuracy: 0.8868

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1953459084033966
Inter Cos: 0.12864218652248383
Norm Quadratic Average: 0.6766194105148315
Nearest Class Center Accuracy: 0.9389

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24648237228393555
Inter Cos: 0.13256831467151642
Norm Quadratic Average: 0.4994153380393982
Nearest Class Center Accuracy: 0.9604

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3319888412952423
Inter Cos: 0.1483156532049179
Norm Quadratic Average: 0.4139825999736786
Nearest Class Center Accuracy: 0.9726

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3862164616584778
Inter Cos: 0.15553338825702667
Norm Quadratic Average: 0.35459068417549133
Nearest Class Center Accuracy: 0.9753

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.44692355394363403
Inter Cos: 0.1868550032377243
Norm Quadratic Average: 0.21030448377132416
Nearest Class Center Accuracy: 0.9891

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6766430139541626
Inter Cos: 0.2317005842924118
Norm Quadratic Average: 0.14020052552223206
Nearest Class Center Accuracy: 0.9927

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7922839522361755
Inter Cos: 0.3139358460903168
Norm Quadratic Average: 0.13144119083881378
Nearest Class Center Accuracy: 0.9937

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.848665714263916
Inter Cos: 0.2533405125141144
Norm Quadratic Average: 0.14222434163093567
Nearest Class Center Accuracy: 0.9946

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9063105583190918
Inter Cos: 0.15341678261756897
Norm Quadratic Average: 0.13355140388011932
Nearest Class Center Accuracy: 0.9949

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9711191058158875
Inter Cos: 0.23794826865196228
Norm Quadratic Average: 0.16514664888381958
Nearest Class Center Accuracy: 0.9945

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.976648211479187
Inter Cos: 0.20391589403152466
Norm Quadratic Average: 0.3962242007255554
Nearest Class Center Accuracy: 0.9948

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.977362871170044
Inter Cos: 0.17069615423679352
Norm Quadratic Average: 1.0031362771987915
Nearest Class Center Accuracy: 0.9947

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.055389165878296
Linear Weight Rank: 9
Intra Cos: 0.978567361831665
Inter Cos: 0.2573600113391876
Norm Quadratic Average: 24.404010772705078
Nearest Class Center Accuracy: 0.9947

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.0567944049835205
Linear Weight Rank: 1492
Intra Cos: 0.9794968962669373
Inter Cos: 0.24183250963687897
Norm Quadratic Average: 16.602296829223633
Nearest Class Center Accuracy: 0.9947

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.0579159259796143
Linear Weight Rank: 9
Intra Cos: 0.9796586036682129
Inter Cos: 0.20447109639644623
Norm Quadratic Average: 11.519610404968262
Nearest Class Center Accuracy: 0.9947

Output Layer:
Intra Cos: 0.9801784753799438
Inter Cos: 0.14919006824493408
Norm Quadratic Average: 8.43647289276123
Nearest Class Center Accuracy: 0.9947

