Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_265358_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: 0.09116753190755844
Inter Cos: 0.10967151820659637
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: 0.09498545527458191
Inter Cos: 0.11538056284189224
Norm Quadratic Average: 62.863616943359375
Nearest Class Center Accuracy: 0.8095666666666667

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12451948970556259
Inter Cos: 0.1411341428756714
Norm Quadratic Average: 71.68860626220703
Nearest Class Center Accuracy: 0.83165

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1346586048603058
Inter Cos: 0.1469249725341797
Norm Quadratic Average: 101.55341339111328
Nearest Class Center Accuracy: 0.8433166666666667

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1993354856967926
Inter Cos: 0.16790564358234406
Norm Quadratic Average: 70.9229507446289
Nearest Class Center Accuracy: 0.9014666666666666

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23700955510139465
Inter Cos: 0.1972522735595703
Norm Quadratic Average: 62.12843322753906
Nearest Class Center Accuracy: 0.9287166666666666

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26805609464645386
Inter Cos: 0.20939846336841583
Norm Quadratic Average: 51.27977752685547
Nearest Class Center Accuracy: 0.9456333333333333

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3000609874725342
Inter Cos: 0.1972314566373825
Norm Quadratic Average: 38.33012390136719
Nearest Class Center Accuracy: 0.95675

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3362235426902771
Inter Cos: 0.19868223369121552
Norm Quadratic Average: 16.047025680541992
Nearest Class Center Accuracy: 0.9762333333333333

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5320926904678345
Inter Cos: 0.3052798807621002
Norm Quadratic Average: 10.96245002746582
Nearest Class Center Accuracy: 0.9852

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6182412505149841
Inter Cos: 0.32402580976486206
Norm Quadratic Average: 11.640246391296387
Nearest Class Center Accuracy: 0.9867666666666667

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6778520345687866
Inter Cos: 0.289624959230423
Norm Quadratic Average: 13.317617416381836
Nearest Class Center Accuracy: 0.9902666666666666

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6696216464042664
Inter Cos: 0.22020874917507172
Norm Quadratic Average: 9.227787971496582
Nearest Class Center Accuracy: 0.9875166666666667

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8248056173324585
Inter Cos: 0.26872095465660095
Norm Quadratic Average: 7.9844279289245605
Nearest Class Center Accuracy: 0.9905833333333334

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8700088858604431
Inter Cos: 0.30030587315559387
Norm Quadratic Average: 8.077815055847168
Nearest Class Center Accuracy: 0.9916833333333334

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8904149532318115
Inter Cos: 0.3515359163284302
Norm Quadratic Average: 7.722751617431641
Nearest Class Center Accuracy: 0.9927833333333334

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.713495254516602
Linear Weight Rank: 4031
Intra Cos: 0.8990729451179504
Inter Cos: 0.32442277669906616
Norm Quadratic Average: 39.908050537109375
Nearest Class Center Accuracy: 0.9946

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.351972579956055
Linear Weight Rank: 3667
Intra Cos: 0.9208607077598572
Inter Cos: 0.2929830849170685
Norm Quadratic Average: 32.76658248901367
Nearest Class Center Accuracy: 0.9964166666666666

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.3562512397766113
Linear Weight Rank: 10
Intra Cos: 0.9274407625198364
Inter Cos: 0.23892490565776825
Norm Quadratic Average: 26.68933868408203
Nearest Class Center Accuracy: 0.9978

Output Layer:
Intra Cos: 0.9479392766952515
Inter Cos: 0.38251039385795593
Norm Quadratic Average: 24.00567626953125
Nearest Class Center Accuracy: 0.9995166666666667

Test Set:
Average Loss: 0.021389365992011154
Accuracy: 0.9943
NC1 Within Class Collapse: 0.7220801115036011
NC2 Equinorm: Features: 0.1423618644475937, Weights: 0.04815590754151344
NC2 Equiangle: Features: 0.2417827606201172, Weights: 0.15044502682156033
NC3 Self-Duality: 0.17145489156246185
NC4 NCC Mismatch: 0.005499999999999949

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
Inter Cos: 0.12048853933811188
Norm Quadratic Average: 23.595195770263672
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10614418238401413
Inter Cos: 0.12709937989711761
Norm Quadratic Average: 63.02543640136719
Nearest Class Center Accuracy: 0.8228

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13831278681755066
Inter Cos: 0.15411768853664398
Norm Quadratic Average: 71.65019226074219
Nearest Class Center Accuracy: 0.8447

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1481838822364807
Inter Cos: 0.15708570182323456
Norm Quadratic Average: 101.57744598388672
Nearest Class Center Accuracy: 0.8548

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21488167345523834
Inter Cos: 0.16693145036697388
Norm Quadratic Average: 70.89901733398438
Nearest Class Center Accuracy: 0.911

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2526416778564453
Inter Cos: 0.19226577877998352
Norm Quadratic Average: 62.13273239135742
Nearest Class Center Accuracy: 0.9378

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.283931165933609
Inter Cos: 0.20218460261821747
Norm Quadratic Average: 51.31365203857422
Nearest Class Center Accuracy: 0.9523

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3162989616394043
Inter Cos: 0.18759284913539886
Norm Quadratic Average: 38.38969039916992
Nearest Class Center Accuracy: 0.9605

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3475823402404785
Inter Cos: 0.21323078870773315
Norm Quadratic Average: 16.10101890563965
Nearest Class Center Accuracy: 0.9755

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5390772223472595
Inter Cos: 0.3232952058315277
Norm Quadratic Average: 11.027929306030273
Nearest Class Center Accuracy: 0.9819

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6246121525764465
Inter Cos: 0.34213587641716003
Norm Quadratic Average: 11.729759216308594
Nearest Class Center Accuracy: 0.9815

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6788040399551392
Inter Cos: 0.3068752884864807
Norm Quadratic Average: 13.429475784301758
Nearest Class Center Accuracy: 0.9844

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6655745506286621
Inter Cos: 0.23091886937618256
Norm Quadratic Average: 9.310717582702637
Nearest Class Center Accuracy: 0.9809

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8211161494255066
Inter Cos: 0.2776312828063965
Norm Quadratic Average: 8.068037986755371
Nearest Class Center Accuracy: 0.9824

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8654487729072571
Inter Cos: 0.29381564259529114
Norm Quadratic Average: 8.163634300231934
Nearest Class Center Accuracy: 0.9835

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8851168751716614
Inter Cos: 0.3455101549625397
Norm Quadratic Average: 7.802039623260498
Nearest Class Center Accuracy: 0.9846

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.713495254516602
Linear Weight Rank: 4031
Intra Cos: 0.8932470679283142
Inter Cos: 0.3197104036808014
Norm Quadratic Average: 40.304351806640625
Nearest Class Center Accuracy: 0.9869

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.351972579956055
Linear Weight Rank: 3667
Intra Cos: 0.9148260354995728
Inter Cos: 0.2883681058883667
Norm Quadratic Average: 33.089073181152344
Nearest Class Center Accuracy: 0.9888

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.3562512397766113
Linear Weight Rank: 10
Intra Cos: 0.9262667894363403
Inter Cos: 0.23739872872829437
Norm Quadratic Average: 26.948341369628906
Nearest Class Center Accuracy: 0.9907

Output Layer:
Intra Cos: 0.944481372833252
Inter Cos: 0.38269463181495667
Norm Quadratic Average: 24.212509155273438
Nearest Class Center Accuracy: 0.9926

