Model save path: ./New_Models/bn_True_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg19_rand_seed_338327_test_samples_None_train_samples_None_weight_decay_0.001.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: 0.10967152565717697
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: 0.05841614678502083
Inter Cos: 0.07648655027151108
Norm Quadratic Average: 6.520430088043213
Nearest Class Center Accuracy: 0.8163333333333334

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09738641232252121
Inter Cos: 0.09317617118358612
Norm Quadratic Average: 4.565361022949219
Nearest Class Center Accuracy: 0.8735666666666667

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09501244872808456
Inter Cos: 0.09055334329605103
Norm Quadratic Average: 3.9909934997558594
Nearest Class Center Accuracy: 0.88315

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17251509428024292
Inter Cos: 0.1212206780910492
Norm Quadratic Average: 3.0499038696289062
Nearest Class Center Accuracy: 0.9365833333333333

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21126419305801392
Inter Cos: 0.11693436652421951
Norm Quadratic Average: 2.216757297515869
Nearest Class Center Accuracy: 0.9605333333333334

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24000173807144165
Inter Cos: 0.1005043089389801
Norm Quadratic Average: 2.0849478244781494
Nearest Class Center Accuracy: 0.9723833333333334

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2751820683479309
Inter Cos: 0.10302840173244476
Norm Quadratic Average: 1.9521745443344116
Nearest Class Center Accuracy: 0.9766333333333334

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.36893701553344727
Inter Cos: 0.11098042130470276
Norm Quadratic Average: 1.569496989250183
Nearest Class Center Accuracy: 0.9923833333333333

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5705389976501465
Inter Cos: 0.12204539030790329
Norm Quadratic Average: 1.1035887002944946
Nearest Class Center Accuracy: 0.9981166666666667

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.727455198764801
Inter Cos: 0.10778789967298508
Norm Quadratic Average: 0.9843418598175049
Nearest Class Center Accuracy: 0.9994666666666666

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8255889415740967
Inter Cos: 0.05395288020372391
Norm Quadratic Average: 0.8137658834457397
Nearest Class Center Accuracy: 0.9999666666666667

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9077774882316589
Inter Cos: 0.05784185603260994
Norm Quadratic Average: 0.6927480101585388
Nearest Class Center Accuracy: 1.0

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9814103841781616
Inter Cos: -0.03714001923799515
Norm Quadratic Average: 0.5920237302780151
Nearest Class Center Accuracy: 1.0

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9964957237243652
Inter Cos: -0.04509306699037552
Norm Quadratic Average: 0.6539005041122437
Nearest Class Center Accuracy: 1.0

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9988117218017578
Inter Cos: -0.054729022085666656
Norm Quadratic Average: 1.0635722875595093
Nearest Class Center Accuracy: 1.0

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.089160442352295
Linear Weight Rank: 4028
Intra Cos: 0.9995861053466797
Inter Cos: -0.0574355311691761
Norm Quadratic Average: 26.478824615478516
Nearest Class Center Accuracy: 1.0

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.473947048187256
Linear Weight Rank: 3638
Intra Cos: 0.9995285272598267
Inter Cos: -0.001577909104526043
Norm Quadratic Average: 18.755237579345703
Nearest Class Center Accuracy: 1.0

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2967922687530518
Linear Weight Rank: 9
Intra Cos: 0.9994885921478271
Inter Cos: 0.02336074784398079
Norm Quadratic Average: 13.648494720458984
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9998918175697327
Inter Cos: 0.09716511517763138
Norm Quadratic Average: 10.673623085021973
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.021597986721713095
Accuracy: 0.9961
NC1 Within Class Collapse: 0.07481972873210907
NC2 Equinorm: Features: 0.01719662919640541, Weights: 0.008585251867771149
NC2 Equiangle: Features: 0.07868782149420844, Weights: 0.04924478530883789
NC3 Self-Duality: 0.014067377895116806
NC4 NCC Mismatch: 0.0

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582170248032
Inter Cos: 0.12048853188753128
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: 0.06601300835609436
Inter Cos: 0.07895921915769577
Norm Quadratic Average: 6.493436813354492
Nearest Class Center Accuracy: 0.8266

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10697223991155624
Inter Cos: 0.095077745616436
Norm Quadratic Average: 4.528916835784912
Nearest Class Center Accuracy: 0.8864

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10450533032417297
Inter Cos: 0.0924844741821289
Norm Quadratic Average: 3.970527410507202
Nearest Class Center Accuracy: 0.8915

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1860685646533966
Inter Cos: 0.12063264101743698
Norm Quadratic Average: 3.0316808223724365
Nearest Class Center Accuracy: 0.9436

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22498540580272675
Inter Cos: 0.11669629067182541
Norm Quadratic Average: 2.2062859535217285
Nearest Class Center Accuracy: 0.9637

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.255058228969574
Inter Cos: 0.11031654477119446
Norm Quadratic Average: 2.0773589611053467
Nearest Class Center Accuracy: 0.9738

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2912716269493103
Inter Cos: 0.11509964615106583
Norm Quadratic Average: 1.9471900463104248
Nearest Class Center Accuracy: 0.9765

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.38287651538848877
Inter Cos: 0.11752110719680786
Norm Quadratic Average: 1.5667707920074463
Nearest Class Center Accuracy: 0.9888

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.580557644367218
Inter Cos: 0.12679630517959595
Norm Quadratic Average: 1.102648138999939
Nearest Class Center Accuracy: 0.9927

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7328371405601501
Inter Cos: 0.11663855612277985
Norm Quadratic Average: 0.9833506345748901
Nearest Class Center Accuracy: 0.9942

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8270741105079651
Inter Cos: 0.05645231157541275
Norm Quadratic Average: 0.8119935393333435
Nearest Class Center Accuracy: 0.9961

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9025407433509827
Inter Cos: 0.05611538141965866
Norm Quadratic Average: 0.6907466650009155
Nearest Class Center Accuracy: 0.9959

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9670933485031128
Inter Cos: -0.03080390766263008
Norm Quadratic Average: 0.590101957321167
Nearest Class Center Accuracy: 0.9962

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9765270948410034
Inter Cos: -0.04266565665602684
Norm Quadratic Average: 0.6515892744064331
Nearest Class Center Accuracy: 0.9962

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9789179563522339
Inter Cos: -0.0522783063352108
Norm Quadratic Average: 1.0598608255386353
Nearest Class Center Accuracy: 0.9961

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 3.089160442352295
Linear Weight Rank: 4028
Intra Cos: 0.9809441566467285
Inter Cos: -0.05399329960346222
Norm Quadratic Average: 26.3845272064209
Nearest Class Center Accuracy: 0.9961

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.473947048187256
Linear Weight Rank: 3638
Intra Cos: 0.9816628694534302
Inter Cos: -0.0002948497422039509
Norm Quadratic Average: 18.689464569091797
Nearest Class Center Accuracy: 0.9961

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.2967922687530518
Linear Weight Rank: 9
Intra Cos: 0.9822433590888977
Inter Cos: 0.03583444654941559
Norm Quadratic Average: 13.60191535949707
Nearest Class Center Accuracy: 0.9961

Output Layer:
Intra Cos: 0.9837173819541931
Inter Cos: 0.10934881120920181
Norm Quadratic Average: 10.63677978515625
Nearest Class Center Accuracy: 0.9962

