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_0.003.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: 0.10424203425645828
Inter Cos: 0.12610352039337158
Norm Quadratic Average: 61.25193786621094
Nearest Class Center Accuracy: 0.8051333333333334

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12997554242610931
Inter Cos: 0.15681549906730652
Norm Quadratic Average: 76.10960388183594
Nearest Class Center Accuracy: 0.8070833333333334

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15101106464862823
Inter Cos: 0.1803167164325714
Norm Quadratic Average: 125.54335021972656
Nearest Class Center Accuracy: 0.8118

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19344356656074524
Inter Cos: 0.19968296587467194
Norm Quadratic Average: 105.7988510131836
Nearest Class Center Accuracy: 0.8390666666666666

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20922262966632843
Inter Cos: 0.19975167512893677
Norm Quadratic Average: 91.49137878417969
Nearest Class Center Accuracy: 0.8674333333333333

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22952617704868317
Inter Cos: 0.21944363415241241
Norm Quadratic Average: 86.40436553955078
Nearest Class Center Accuracy: 0.8923666666666666

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.27112650871276855
Inter Cos: 0.22324782609939575
Norm Quadratic Average: 74.6190414428711
Nearest Class Center Accuracy: 0.9262333333333334

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2546845078468323
Inter Cos: 0.22874850034713745
Norm Quadratic Average: 28.03229331970215
Nearest Class Center Accuracy: 0.9415833333333333

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3499126136302948
Inter Cos: 0.2965485453605652
Norm Quadratic Average: 14.85600757598877
Nearest Class Center Accuracy: 0.9365166666666667

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.45946004986763
Inter Cos: 0.3319670557975769
Norm Quadratic Average: 14.923333168029785
Nearest Class Center Accuracy: 0.9498833333333333

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5520995855331421
Inter Cos: 0.3704078495502472
Norm Quadratic Average: 17.572887420654297
Nearest Class Center Accuracy: 0.9666166666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5616674423217773
Inter Cos: 0.4194134771823883
Norm Quadratic Average: 12.659408569335938
Nearest Class Center Accuracy: 0.9522333333333334

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7206016778945923
Inter Cos: 0.4382385015487671
Norm Quadratic Average: 10.980796813964844
Nearest Class Center Accuracy: 0.9623

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7704058885574341
Inter Cos: 0.36821722984313965
Norm Quadratic Average: 12.166555404663086
Nearest Class Center Accuracy: 0.9743333333333334

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8109813928604126
Inter Cos: 0.44613271951675415
Norm Quadratic Average: 13.668649673461914
Nearest Class Center Accuracy: 0.9823

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.763435959815979
Linear Weight Rank: 560
Intra Cos: 0.8359563946723938
Inter Cos: 0.3673763573169708
Norm Quadratic Average: 60.561275482177734
Nearest Class Center Accuracy: 0.9888666666666667

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.78980553150177
Linear Weight Rank: 2723
Intra Cos: 0.8880875706672668
Inter Cos: 0.3707146942615509
Norm Quadratic Average: 42.504764556884766
Nearest Class Center Accuracy: 0.9953166666666666

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.7686101198196411
Linear Weight Rank: 9
Intra Cos: 0.9064781069755554
Inter Cos: 0.35544705390930176
Norm Quadratic Average: 28.49363136291504
Nearest Class Center Accuracy: 0.9967833333333334

Output Layer:
Intra Cos: 0.9462805986404419
Inter Cos: 0.43851330876350403
Norm Quadratic Average: 21.396574020385742
Nearest Class Center Accuracy: 0.9985666666666667

Test Set:
Average Loss: 0.025025078054890036
Accuracy: 0.9933
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.10293924808502197, Weights: 0.032932646572589874
NC2 Equiangle: Features: 0.2778216256035699, Weights: 0.21466060214572483
NC3 Self-Duality: 0.08588044345378876
NC4 NCC Mismatch: 0.006000000000000005

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: 0.11571474373340607
Inter Cos: 0.13852961361408234
Norm Quadratic Average: 61.511741638183594
Nearest Class Center Accuracy: 0.8194

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.14392602443695068
Inter Cos: 0.17159220576286316
Norm Quadratic Average: 76.26593017578125
Nearest Class Center Accuracy: 0.8228

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16589567065238953
Inter Cos: 0.1970074623823166
Norm Quadratic Average: 125.84343719482422
Nearest Class Center Accuracy: 0.8277

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20714548230171204
Inter Cos: 0.21904437243938446
Norm Quadratic Average: 105.8163833618164
Nearest Class Center Accuracy: 0.8565

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2236652821302414
Inter Cos: 0.21108879148960114
Norm Quadratic Average: 91.51378631591797
Nearest Class Center Accuracy: 0.8822

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24247556924819946
Inter Cos: 0.21243026852607727
Norm Quadratic Average: 86.47785949707031
Nearest Class Center Accuracy: 0.9038

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2826399505138397
Inter Cos: 0.22320552170276642
Norm Quadratic Average: 74.86727142333984
Nearest Class Center Accuracy: 0.9346

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2667251527309418
Inter Cos: 0.22041934728622437
Norm Quadratic Average: 28.141918182373047
Nearest Class Center Accuracy: 0.9489

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.36852502822875977
Inter Cos: 0.3203328847885132
Norm Quadratic Average: 14.922636985778809
Nearest Class Center Accuracy: 0.9388

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4804129898548126
Inter Cos: 0.3552734851837158
Norm Quadratic Average: 15.026057243347168
Nearest Class Center Accuracy: 0.9483

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5739131569862366
Inter Cos: 0.3901982307434082
Norm Quadratic Average: 17.733020782470703
Nearest Class Center Accuracy: 0.963

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5709685683250427
Inter Cos: 0.43540364503860474
Norm Quadratic Average: 12.80256175994873
Nearest Class Center Accuracy: 0.9499

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7304078936576843
Inter Cos: 0.4505425989627838
Norm Quadratic Average: 11.108492851257324
Nearest Class Center Accuracy: 0.9596

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7753233313560486
Inter Cos: 0.3903881013393402
Norm Quadratic Average: 12.31050968170166
Nearest Class Center Accuracy: 0.9698

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.814162015914917
Inter Cos: 0.46732476353645325
Norm Quadratic Average: 13.835043907165527
Nearest Class Center Accuracy: 0.9775

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.763435959815979
Linear Weight Rank: 560
Intra Cos: 0.8355875611305237
Inter Cos: 0.38850846886634827
Norm Quadratic Average: 61.26264953613281
Nearest Class Center Accuracy: 0.9818

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.78980553150177
Linear Weight Rank: 2723
Intra Cos: 0.8852697014808655
Inter Cos: 0.39001336693763733
Norm Quadratic Average: 43.009037017822266
Nearest Class Center Accuracy: 0.9884

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.7686101198196411
Linear Weight Rank: 9
Intra Cos: 0.9045351147651672
Inter Cos: 0.373460054397583
Norm Quadratic Average: 28.832012176513672
Nearest Class Center Accuracy: 0.9901

Output Layer:
Intra Cos: 0.9382990598678589
Inter Cos: 0.4549564719200134
Norm Quadratic Average: 21.654584884643555
Nearest Class Center Accuracy: 0.9921

