Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_846264_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: 0.09116754680871964
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567686080932617
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.07789336889982224
Inter Cos: 0.09420599788427353
Norm Quadratic Average: 54.92119216918945
Nearest Class Center Accuracy: 0.8188

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12061983346939087
Inter Cos: 0.123051218688488
Norm Quadratic Average: 48.86103057861328
Nearest Class Center Accuracy: 0.8566166666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13368384540081024
Inter Cos: 0.1373014897108078
Norm Quadratic Average: 63.47177505493164
Nearest Class Center Accuracy: 0.8680666666666667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22369401156902313
Inter Cos: 0.1812528520822525
Norm Quadratic Average: 47.94948959350586
Nearest Class Center Accuracy: 0.9170666666666667

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2463640421628952
Inter Cos: 0.1736336350440979
Norm Quadratic Average: 48.71979904174805
Nearest Class Center Accuracy: 0.9366833333333333

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2602861523628235
Inter Cos: 0.17069940268993378
Norm Quadratic Average: 44.573585510253906
Nearest Class Center Accuracy: 0.9519666666666666

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.284074604511261
Inter Cos: 0.18068912625312805
Norm Quadratic Average: 40.40338897705078
Nearest Class Center Accuracy: 0.9590166666666666

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3364672362804413
Inter Cos: 0.1691163033246994
Norm Quadratic Average: 20.033536911010742
Nearest Class Center Accuracy: 0.9828166666666667

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4425306022167206
Inter Cos: 0.17814040184020996
Norm Quadratic Average: 16.158031463623047
Nearest Class Center Accuracy: 0.9910333333333333

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5521754026412964
Inter Cos: 0.193807914853096
Norm Quadratic Average: 14.604881286621094
Nearest Class Center Accuracy: 0.9948666666666667

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6362003684043884
Inter Cos: 0.20529106259346008
Norm Quadratic Average: 13.786877632141113
Nearest Class Center Accuracy: 0.99675

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7018303275108337
Inter Cos: 0.14747509360313416
Norm Quadratic Average: 8.640700340270996
Nearest Class Center Accuracy: 0.9970333333333333

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.860807478427887
Inter Cos: 0.19670331478118896
Norm Quadratic Average: 7.981334686279297
Nearest Class Center Accuracy: 0.9977833333333334

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9187667965888977
Inter Cos: 0.26359823346138
Norm Quadratic Average: 7.70906400680542
Nearest Class Center Accuracy: 0.9981666666666666

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9386447668075562
Inter Cos: 0.31119996309280396
Norm Quadratic Average: 7.395028591156006
Nearest Class Center Accuracy: 0.9984166666666666

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.79541015625
Linear Weight Rank: 4031
Intra Cos: 0.9549726247787476
Inter Cos: 0.2929936349391937
Norm Quadratic Average: 42.37078857421875
Nearest Class Center Accuracy: 0.9992

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.897796630859375
Linear Weight Rank: 3671
Intra Cos: 0.9588874578475952
Inter Cos: 0.2981507182121277
Norm Quadratic Average: 33.71760940551758
Nearest Class Center Accuracy: 0.9994333333333333

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.810436487197876
Linear Weight Rank: 10
Intra Cos: 0.9589786529541016
Inter Cos: 0.2692314684391022
Norm Quadratic Average: 28.548315048217773
Nearest Class Center Accuracy: 0.9996166666666667

Output Layer:
Intra Cos: 0.9862983226776123
Inter Cos: 0.3482234477996826
Norm Quadratic Average: 26.202760696411133
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.024513658488589864
Accuracy: 0.9954
NC1 Within Class Collapse: 0.38657259941101074
NC2 Equinorm: Features: 0.08334962278604507, Weights: 0.025369293987751007
NC2 Equiangle: Features: 0.21507551405164932, Weights: 0.11165956921047634
NC3 Self-Duality: 0.29892978072166443
NC4 NCC Mismatch: 0.0031999999999999806

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: 23.59519386291504
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0872453898191452
Inter Cos: 0.09907370060682297
Norm Quadratic Average: 54.95248031616211
Nearest Class Center Accuracy: 0.8316

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1329715996980667
Inter Cos: 0.13119831681251526
Norm Quadratic Average: 48.7390022277832
Nearest Class Center Accuracy: 0.8686

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14653518795967102
Inter Cos: 0.14947840571403503
Norm Quadratic Average: 63.35171127319336
Nearest Class Center Accuracy: 0.8792

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24025709927082062
Inter Cos: 0.19496482610702515
Norm Quadratic Average: 47.812862396240234
Nearest Class Center Accuracy: 0.9282

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26192373037338257
Inter Cos: 0.18729835748672485
Norm Quadratic Average: 48.58843994140625
Nearest Class Center Accuracy: 0.9462

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2754724621772766
Inter Cos: 0.1851329356431961
Norm Quadratic Average: 44.49856185913086
Nearest Class Center Accuracy: 0.958

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2988901138305664
Inter Cos: 0.19597625732421875
Norm Quadratic Average: 40.36307907104492
Nearest Class Center Accuracy: 0.9636

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.35057637095451355
Inter Cos: 0.18236473202705383
Norm Quadratic Average: 20.0313777923584
Nearest Class Center Accuracy: 0.9827

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4551231265068054
Inter Cos: 0.19071520864963531
Norm Quadratic Average: 16.179033279418945
Nearest Class Center Accuracy: 0.9891

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5626407265663147
Inter Cos: 0.20368348062038422
Norm Quadratic Average: 14.641599655151367
Nearest Class Center Accuracy: 0.9901

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6441983580589294
Inter Cos: 0.21234624087810516
Norm Quadratic Average: 13.82946491241455
Nearest Class Center Accuracy: 0.9908

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7054724097251892
Inter Cos: 0.15169478952884674
Norm Quadratic Average: 8.678065299987793
Nearest Class Center Accuracy: 0.991

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8596233129501343
Inter Cos: 0.19016188383102417
Norm Quadratic Average: 8.020079612731934
Nearest Class Center Accuracy: 0.9915

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9151726365089417
Inter Cos: 0.2598702609539032
Norm Quadratic Average: 7.744898796081543
Nearest Class Center Accuracy: 0.9919

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9346715807914734
Inter Cos: 0.30610907077789307
Norm Quadratic Average: 7.425693035125732
Nearest Class Center Accuracy: 0.9927

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.79541015625
Linear Weight Rank: 4031
Intra Cos: 0.9499626755714417
Inter Cos: 0.28862276673316956
Norm Quadratic Average: 42.53263854980469
Nearest Class Center Accuracy: 0.9933

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.897796630859375
Linear Weight Rank: 3671
Intra Cos: 0.9541566967964172
Inter Cos: 0.29394108057022095
Norm Quadratic Average: 33.84489822387695
Nearest Class Center Accuracy: 0.994

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.810436487197876
Linear Weight Rank: 10
Intra Cos: 0.9561915993690491
Inter Cos: 0.26563316583633423
Norm Quadratic Average: 28.655757904052734
Nearest Class Center Accuracy: 0.9942

Output Layer:
Intra Cos: 0.9800330996513367
Inter Cos: 0.3443569540977478
Norm Quadratic Average: 26.30250358581543
Nearest Class Center Accuracy: 0.9951

