Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_314159_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: 0.09116753935813904
Inter Cos: 0.10967151075601578
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: 0.0711955577135086
Inter Cos: 0.0870402604341507
Norm Quadratic Average: 94.51315307617188
Nearest Class Center Accuracy: 0.8244

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.104328952729702
Inter Cos: 0.10893647372722626
Norm Quadratic Average: 60.663124084472656
Nearest Class Center Accuracy: 0.8634666666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11037763953208923
Inter Cos: 0.11313968151807785
Norm Quadratic Average: 61.7314453125
Nearest Class Center Accuracy: 0.8761666666666666

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18620778620243073
Inter Cos: 0.13702699542045593
Norm Quadratic Average: 39.88050842285156
Nearest Class Center Accuracy: 0.9263833333333333

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22176027297973633
Inter Cos: 0.1459658294916153
Norm Quadratic Average: 41.44325637817383
Nearest Class Center Accuracy: 0.9459166666666666

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2443174570798874
Inter Cos: 0.14835704863071442
Norm Quadratic Average: 41.65126419067383
Nearest Class Center Accuracy: 0.9589833333333333

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2664770483970642
Inter Cos: 0.14369116723537445
Norm Quadratic Average: 44.095481872558594
Nearest Class Center Accuracy: 0.9678

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.31344926357269287
Inter Cos: 0.1886041909456253
Norm Quadratic Average: 28.859230041503906
Nearest Class Center Accuracy: 0.9884833333333334

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.396193265914917
Inter Cos: 0.25071603059768677
Norm Quadratic Average: 29.452638626098633
Nearest Class Center Accuracy: 0.9952333333333333

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.483293354511261
Inter Cos: 0.2564341425895691
Norm Quadratic Average: 30.813037872314453
Nearest Class Center Accuracy: 0.9982

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5753424167633057
Inter Cos: 0.2463703751564026
Norm Quadratic Average: 31.327667236328125
Nearest Class Center Accuracy: 0.99945

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7080692052841187
Inter Cos: 0.3202000558376312
Norm Quadratic Average: 25.09193992614746
Nearest Class Center Accuracy: 0.9997333333333334

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.845731794834137
Inter Cos: 0.24743293225765228
Norm Quadratic Average: 16.675935745239258
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9079945683479309
Inter Cos: 0.1544904261827469
Norm Quadratic Average: 17.638429641723633
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9402232766151428
Inter Cos: 0.05684547871351242
Norm Quadratic Average: 18.67638397216797
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.80746459960938
Linear Weight Rank: 4031
Intra Cos: 0.9547910094261169
Inter Cos: 0.09231123328208923
Norm Quadratic Average: 124.01715087890625
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.946557998657227
Linear Weight Rank: 3670
Intra Cos: 0.982767641544342
Inter Cos: 0.022034423425793648
Norm Quadratic Average: 65.6098861694336
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.4173216819763184
Linear Weight Rank: 10
Intra Cos: 0.9837801456451416
Inter Cos: 0.04120972380042076
Norm Quadratic Average: 31.45696258544922
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9983634948730469
Inter Cos: 0.18651874363422394
Norm Quadratic Average: 18.699649810791016
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.023315428086756764
Accuracy: 0.9953
NC1 Within Class Collapse: 0.21321862936019897
NC2 Equinorm: Features: 0.06278397142887115, Weights: 0.015893813222646713
NC2 Equiangle: Features: 0.06919260025024414, Weights: 0.07543937895033094
NC3 Self-Duality: 0.5832602977752686
NC4 NCC Mismatch: 0.00019999999999997797

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
Inter Cos: 0.12048851698637009
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: 0.08016730844974518
Inter Cos: 0.08940672874450684
Norm Quadratic Average: 94.30241394042969
Nearest Class Center Accuracy: 0.8382

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11558669060468674
Inter Cos: 0.1093481108546257
Norm Quadratic Average: 60.18940734863281
Nearest Class Center Accuracy: 0.8761

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12109531462192535
Inter Cos: 0.11348896473646164
Norm Quadratic Average: 61.31507873535156
Nearest Class Center Accuracy: 0.8875

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2008964717388153
Inter Cos: 0.13517671823501587
Norm Quadratic Average: 39.5876579284668
Nearest Class Center Accuracy: 0.9362

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2368246167898178
Inter Cos: 0.14247220754623413
Norm Quadratic Average: 41.154579162597656
Nearest Class Center Accuracy: 0.9523

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2575787603855133
Inter Cos: 0.14413881301879883
Norm Quadratic Average: 41.4039306640625
Nearest Class Center Accuracy: 0.9619

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.28032514452934265
Inter Cos: 0.13923592865467072
Norm Quadratic Average: 43.863670349121094
Nearest Class Center Accuracy: 0.9686

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3243493139743805
Inter Cos: 0.1922040432691574
Norm Quadratic Average: 28.741443634033203
Nearest Class Center Accuracy: 0.9864

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.40512725710868835
Inter Cos: 0.2523157596588135
Norm Quadratic Average: 29.358409881591797
Nearest Class Center Accuracy: 0.9909

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.48999178409576416
Inter Cos: 0.25335341691970825
Norm Quadratic Average: 30.735881805419922
Nearest Class Center Accuracy: 0.9928

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5790279507637024
Inter Cos: 0.24092480540275574
Norm Quadratic Average: 31.275339126586914
Nearest Class Center Accuracy: 0.9939

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7051060795783997
Inter Cos: 0.31078657507896423
Norm Quadratic Average: 25.06966781616211
Nearest Class Center Accuracy: 0.9937

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.838581383228302
Inter Cos: 0.2376679927110672
Norm Quadratic Average: 16.66562271118164
Nearest Class Center Accuracy: 0.995

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8984514474868774
Inter Cos: 0.14538511633872986
Norm Quadratic Average: 17.622875213623047
Nearest Class Center Accuracy: 0.9953

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9194983839988708
Inter Cos: 0.04939280450344086
Norm Quadratic Average: 18.656925201416016
Nearest Class Center Accuracy: 0.9954

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.80746459960938
Linear Weight Rank: 4031
Intra Cos: 0.9237130880355835
Inter Cos: 0.08562366664409637
Norm Quadratic Average: 123.91044616699219
Nearest Class Center Accuracy: 0.9949

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.946557998657227
Linear Weight Rank: 3670
Intra Cos: 0.9558895230293274
Inter Cos: 0.018222901970148087
Norm Quadratic Average: 65.49382019042969
Nearest Class Center Accuracy: 0.9953

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.4173216819763184
Linear Weight Rank: 10
Intra Cos: 0.9558337330818176
Inter Cos: 0.043247684836387634
Norm Quadratic Average: 31.413455963134766
Nearest Class Center Accuracy: 0.9954

Output Layer:
Intra Cos: 0.9860848784446716
Inter Cos: 0.19272074103355408
Norm Quadratic Average: 18.657989501953125
Nearest Class Center Accuracy: 0.9955

