Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.0005.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: 0.10967149585485458
Norm Quadratic Average: 23.567672729492188
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09450254589319229
Inter Cos: 0.11486953496932983
Norm Quadratic Average: 58.936153411865234
Nearest Class Center Accuracy: 0.8112666666666667

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12512627243995667
Inter Cos: 0.14491726458072662
Norm Quadratic Average: 70.63101196289062
Nearest Class Center Accuracy: 0.8265666666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1371207982301712
Inter Cos: 0.1536426544189453
Norm Quadratic Average: 108.57113647460938
Nearest Class Center Accuracy: 0.8371333333333333

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20459488034248352
Inter Cos: 0.16971443593502045
Norm Quadratic Average: 83.38177490234375
Nearest Class Center Accuracy: 0.8918166666666667

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23894408345222473
Inter Cos: 0.18414711952209473
Norm Quadratic Average: 78.19869995117188
Nearest Class Center Accuracy: 0.9195

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2731159031391144
Inter Cos: 0.1833745241165161
Norm Quadratic Average: 68.06965637207031
Nearest Class Center Accuracy: 0.9379333333333333

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.30214184522628784
Inter Cos: 0.17201372981071472
Norm Quadratic Average: 51.623111724853516
Nearest Class Center Accuracy: 0.9520833333333333

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.33386364579200745
Inter Cos: 0.16332079470157623
Norm Quadratic Average: 21.549419403076172
Nearest Class Center Accuracy: 0.9737166666666667

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.511766791343689
Inter Cos: 0.26068320870399475
Norm Quadratic Average: 13.569676399230957
Nearest Class Center Accuracy: 0.9857833333333333

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6007536053657532
Inter Cos: 0.3171784579753876
Norm Quadratic Average: 12.545870780944824
Nearest Class Center Accuracy: 0.9888166666666667

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6501199007034302
Inter Cos: 0.3308776915073395
Norm Quadratic Average: 12.738494873046875
Nearest Class Center Accuracy: 0.9920833333333333

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.693154513835907
Inter Cos: 0.2805764675140381
Norm Quadratic Average: 8.02414608001709
Nearest Class Center Accuracy: 0.98905

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8692114353179932
Inter Cos: 0.3303561508655548
Norm Quadratic Average: 6.914339065551758
Nearest Class Center Accuracy: 0.9920833333333333

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9020925760269165
Inter Cos: 0.36969372630119324
Norm Quadratic Average: 7.083847999572754
Nearest Class Center Accuracy: 0.9939666666666667

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9166684150695801
Inter Cos: 0.368031769990921
Norm Quadratic Average: 6.947036266326904
Nearest Class Center Accuracy: 0.9951166666666666

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.70848274230957
Linear Weight Rank: 4031
Intra Cos: 0.924084484577179
Inter Cos: 0.3283388614654541
Norm Quadratic Average: 38.391693115234375
Nearest Class Center Accuracy: 0.99675

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.347946643829346
Linear Weight Rank: 3667
Intra Cos: 0.935456395149231
Inter Cos: 0.3313771188259125
Norm Quadratic Average: 32.566810607910156
Nearest Class Center Accuracy: 0.9977666666666667

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.335588216781616
Linear Weight Rank: 10
Intra Cos: 0.9390350580215454
Inter Cos: 0.30962464213371277
Norm Quadratic Average: 26.897911071777344
Nearest Class Center Accuracy: 0.9986333333333334

Output Layer:
Intra Cos: 0.9716017246246338
Inter Cos: 0.3698749542236328
Norm Quadratic Average: 24.76255989074707
Nearest Class Center Accuracy: 0.9999

Test Set:
Average Loss: 0.02227879616842838
Accuracy: 0.9942
NC1 Within Class Collapse: 0.8584858179092407
NC2 Equinorm: Features: 0.11507857590913773, Weights: 0.03828060254454613
NC2 Equiangle: Features: 0.2602678722805447, Weights: 0.1656216091579861
NC3 Self-Duality: 0.15621496737003326
NC4 NCC Mismatch: 0.0047000000000000375

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219583660364151
Inter Cos: 0.12048852443695068
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: 0.10585060715675354
Inter Cos: 0.12652556598186493
Norm Quadratic Average: 59.06249237060547
Nearest Class Center Accuracy: 0.8236

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13932089507579803
Inter Cos: 0.15867586433887482
Norm Quadratic Average: 70.58821105957031
Nearest Class Center Accuracy: 0.8411

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1514243483543396
Inter Cos: 0.16773013770580292
Norm Quadratic Average: 108.61536407470703
Nearest Class Center Accuracy: 0.849

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22090379893779755
Inter Cos: 0.1840207278728485
Norm Quadratic Average: 83.35858154296875
Nearest Class Center Accuracy: 0.9014

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.25494858622550964
Inter Cos: 0.18250878155231476
Norm Quadratic Average: 78.20484924316406
Nearest Class Center Accuracy: 0.9302

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28699755668640137
Inter Cos: 0.18605652451515198
Norm Quadratic Average: 68.10674285888672
Nearest Class Center Accuracy: 0.9454

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3144535422325134
Inter Cos: 0.18616732954978943
Norm Quadratic Average: 51.69952392578125
Nearest Class Center Accuracy: 0.9566

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3476405739784241
Inter Cos: 0.17273958027362823
Norm Quadratic Average: 21.615535736083984
Nearest Class Center Accuracy: 0.9734

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5206999182701111
Inter Cos: 0.2819669842720032
Norm Quadratic Average: 13.638941764831543
Nearest Class Center Accuracy: 0.9821

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6079393029212952
Inter Cos: 0.3418634831905365
Norm Quadratic Average: 12.636055946350098
Nearest Class Center Accuracy: 0.9831

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6598104238510132
Inter Cos: 0.3559737801551819
Norm Quadratic Average: 12.844511985778809
Nearest Class Center Accuracy: 0.986

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6901255249977112
Inter Cos: 0.28574466705322266
Norm Quadratic Average: 8.093730926513672
Nearest Class Center Accuracy: 0.9833

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8709467053413391
Inter Cos: 0.3295442461967468
Norm Quadratic Average: 6.981170177459717
Nearest Class Center Accuracy: 0.9855

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9030129909515381
Inter Cos: 0.36737489700317383
Norm Quadratic Average: 7.15273904800415
Nearest Class Center Accuracy: 0.9881

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9142597317695618
Inter Cos: 0.3680035471916199
Norm Quadratic Average: 7.011588096618652
Nearest Class Center Accuracy: 0.9895

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.70848274230957
Linear Weight Rank: 4031
Intra Cos: 0.9214064478874207
Inter Cos: 0.3286770284175873
Norm Quadratic Average: 38.71274948120117
Nearest Class Center Accuracy: 0.9906

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.347946643829346
Linear Weight Rank: 3667
Intra Cos: 0.9333557486534119
Inter Cos: 0.33101484179496765
Norm Quadratic Average: 32.84113693237305
Nearest Class Center Accuracy: 0.9913

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.335588216781616
Linear Weight Rank: 10
Intra Cos: 0.9361171126365662
Inter Cos: 0.3092118501663208
Norm Quadratic Average: 27.12839126586914
Nearest Class Center Accuracy: 0.9921

Output Layer:
Intra Cos: 0.9638383984565735
Inter Cos: 0.36884722113609314
Norm Quadratic Average: 24.977319717407227
Nearest Class Center Accuracy: 0.9937

