Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_True_dataset_CIFAR10_epochs_100_lr_0.001_model_type_vgg11_train_samples_None_weight_decay_0.0007.pth.tar
Training Set:
Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0198909230530262
Inter Cos: 0.10477277636528015
Norm Quadratic Average: 27.597179412841797
Nearest Class Center Accuracy: 0.26968

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.020912593230605125
Inter Cos: 0.07799948006868362
Norm Quadratic Average: 73.7671890258789
Nearest Class Center Accuracy: 0.36216

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.023543493822216988
Inter Cos: 0.08085056394338608
Norm Quadratic Average: 51.79608917236328
Nearest Class Center Accuracy: 0.43434

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.021543394774198532
Inter Cos: 0.0649670660495758
Norm Quadratic Average: 55.28752899169922
Nearest Class Center Accuracy: 0.4873

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02759638987481594
Inter Cos: 0.05937406048178673
Norm Quadratic Average: 36.32380676269531
Nearest Class Center Accuracy: 0.57832

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.030895225703716278
Inter Cos: 0.05484750494360924
Norm Quadratic Average: 37.88834762573242
Nearest Class Center Accuracy: 0.62998

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.04832705110311508
Inter Cos: 0.06871174275875092
Norm Quadratic Average: 24.4060115814209
Nearest Class Center Accuracy: 0.72778

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0927659347653389
Inter Cos: 0.09828443825244904
Norm Quadratic Average: 17.5947322845459
Nearest Class Center Accuracy: 0.94332

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 82.09098052978516
Linear Weight Rank: 4031
Intra Cos: 0.29376623034477234
Inter Cos: 0.14032834768295288
Norm Quadratic Average: 100.30616760253906
Nearest Class Center Accuracy: 0.99992

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 33.20356750488281
Linear Weight Rank: 3670
Intra Cos: 0.6769782900810242
Inter Cos: 0.13885679841041565
Norm Quadratic Average: 55.677425384521484
Nearest Class Center Accuracy: 1.0

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.4355051517486572
Linear Weight Rank: 10
Intra Cos: 0.8099952936172485
Inter Cos: 0.1633628010749817
Norm Quadratic Average: 37.225914001464844
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9008151888847351
Inter Cos: 0.4333634376525879
Norm Quadratic Average: 25.396961212158203
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 1.468010841369629
Accuracy: 0.771
NC1 Within Class Collapse: 8.527978897094727
NC2 Equinorm: Features: 0.2177441269159317, Weights: 0.012691570445895195
NC2 Equiangle: Features: 0.2109733157687717, Weights: 0.07610677083333334
NC3 Self-Duality: 0.41074153780937195
NC4 NCC Mismatch: 0.0665

Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.018480461090803146
Inter Cos: 0.09352367371320724
Norm Quadratic Average: 27.530664443969727
Nearest Class Center Accuracy: 0.2895

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.020215975120663643
Inter Cos: 0.0723886489868164
Norm Quadratic Average: 73.71317291259766
Nearest Class Center Accuracy: 0.397

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02315402962267399
Inter Cos: 0.07551921904087067
Norm Quadratic Average: 51.68972396850586
Nearest Class Center Accuracy: 0.478

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.020226072520017624
Inter Cos: 0.06105148792266846
Norm Quadratic Average: 55.22639083862305
Nearest Class Center Accuracy: 0.5425

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.023942388594150543
Inter Cos: 0.055532056838274
Norm Quadratic Average: 36.29975891113281
Nearest Class Center Accuracy: 0.607

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.025758616626262665
Inter Cos: 0.0515313558280468
Norm Quadratic Average: 37.847434997558594
Nearest Class Center Accuracy: 0.637

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.035517435520887375
Inter Cos: 0.06703128665685654
Norm Quadratic Average: 24.3436336517334
Nearest Class Center Accuracy: 0.671

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.057311657816171646
Inter Cos: 0.10075370967388153
Norm Quadratic Average: 17.44536018371582
Nearest Class Center Accuracy: 0.7375

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 82.09098052978516
Linear Weight Rank: 4031
Intra Cos: 0.13706816732883453
Inter Cos: 0.16700084507465363
Norm Quadratic Average: 96.57723999023438
Nearest Class Center Accuracy: 0.773

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 33.20356750488281
Linear Weight Rank: 3670
Intra Cos: 0.29245394468307495
Inter Cos: 0.22612877190113068
Norm Quadratic Average: 50.82449722290039
Nearest Class Center Accuracy: 0.768

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.4355051517486572
Linear Weight Rank: 10
Intra Cos: 0.3732176721096039
Inter Cos: 0.27514833211898804
Norm Quadratic Average: 33.484371185302734
Nearest Class Center Accuracy: 0.7665

Output Layer:
Intra Cos: 0.45452335476875305
Inter Cos: 0.4472281336784363
Norm Quadratic Average: 22.75673484802246
Nearest Class Center Accuracy: 0.7715

