Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_265358_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: 0.09116753190755844
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567670822143555
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12203904986381531
Inter Cos: 0.14823605120182037
Norm Quadratic Average: 40.62078094482422
Nearest Class Center Accuracy: 0.8093

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17963054776191711
Inter Cos: 0.17503122985363007
Norm Quadratic Average: 41.715850830078125
Nearest Class Center Accuracy: 0.8280666666666666

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2126777321100235
Inter Cos: 0.20251591503620148
Norm Quadratic Average: 40.41639709472656
Nearest Class Center Accuracy: 0.8740166666666667

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21346494555473328
Inter Cos: 0.20877705514431
Norm Quadratic Average: 19.35365104675293
Nearest Class Center Accuracy: 0.9153666666666667

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.307102233171463
Inter Cos: 0.268259197473526
Norm Quadratic Average: 11.348520278930664
Nearest Class Center Accuracy: 0.9498166666666666

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4981216788291931
Inter Cos: 0.34193381667137146
Norm Quadratic Average: 6.65035343170166
Nearest Class Center Accuracy: 0.9800833333333333

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7245307564735413
Inter Cos: 0.38761061429977417
Norm Quadratic Average: 6.539748668670654
Nearest Class Center Accuracy: 0.9925333333333334

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.1118428707122803
Linear Weight Rank: 359
Intra Cos: 0.8168690204620361
Inter Cos: 0.3249706029891968
Norm Quadratic Average: 31.75285530090332
Nearest Class Center Accuracy: 0.9959666666666667

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.1235225200653076
Linear Weight Rank: 2708
Intra Cos: 0.8882178068161011
Inter Cos: 0.34185367822647095
Norm Quadratic Average: 26.338558197021484
Nearest Class Center Accuracy: 0.9982333333333333

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.108957052230835
Linear Weight Rank: 9
Intra Cos: 0.9084826707839966
Inter Cos: 0.329601913690567
Norm Quadratic Average: 21.126937866210938
Nearest Class Center Accuracy: 0.9987166666666667

Output Layer:
Intra Cos: 0.9395649433135986
Inter Cos: 0.40640905499458313
Norm Quadratic Average: 19.252397537231445
Nearest Class Center Accuracy: 0.999

Test Set:
Average Loss: 0.026117408740520476
Accuracy: 0.9915
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.1030217707157135, Weights: 0.05049870163202286
NC2 Equiangle: Features: 0.2585716671413846, Weights: 0.22937893337673612
NC3 Self-Duality: 0.06187182664871216
NC4 NCC Mismatch: 0.0034999999999999476

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, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1356479823589325
Inter Cos: 0.16250737011432648
Norm Quadratic Average: 40.6505241394043
Nearest Class Center Accuracy: 0.8235

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1956673115491867
Inter Cos: 0.18861804902553558
Norm Quadratic Average: 41.62760925292969
Nearest Class Center Accuracy: 0.8454

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22698169946670532
Inter Cos: 0.22024983167648315
Norm Quadratic Average: 40.36984634399414
Nearest Class Center Accuracy: 0.8857

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22460223734378815
Inter Cos: 0.22159667313098907
Norm Quadratic Average: 19.326488494873047
Nearest Class Center Accuracy: 0.9261

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3171674907207489
Inter Cos: 0.2750820815563202
Norm Quadratic Average: 11.358731269836426
Nearest Class Center Accuracy: 0.9536

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5083057880401611
Inter Cos: 0.3694150447845459
Norm Quadratic Average: 6.687739849090576
Nearest Class Center Accuracy: 0.9783

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7347540855407715
Inter Cos: 0.40093034505844116
Norm Quadratic Average: 6.596505641937256
Nearest Class Center Accuracy: 0.9866

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 2.1118428707122803
Linear Weight Rank: 359
Intra Cos: 0.8148186802864075
Inter Cos: 0.345284640789032
Norm Quadratic Average: 32.04230499267578
Nearest Class Center Accuracy: 0.9878

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 2.1235225200653076
Linear Weight Rank: 2708
Intra Cos: 0.8821125626564026
Inter Cos: 0.3663995862007141
Norm Quadratic Average: 26.584196090698242
Nearest Class Center Accuracy: 0.9896

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.108957052230835
Linear Weight Rank: 9
Intra Cos: 0.9009025692939758
Inter Cos: 0.352152556180954
Norm Quadratic Average: 21.323083877563477
Nearest Class Center Accuracy: 0.9904

Output Layer:
Intra Cos: 0.9289838671684265
Inter Cos: 0.42779839038848877
Norm Quadratic Average: 19.429716110229492
Nearest Class Center Accuracy: 0.99

