Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_True_dataset_mlp6_epochs_100_lr_0.001_model_depth_MLP_3_model_type_MLP_rand_seed_314159_weight_decay_0.02.pth.tar
Training Set:
Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 3.1982312202453613
Linear Weight Rank: 16
Intra Cos: 0.014583898708224297
Inter Cos: 0.004829856567084789
Norm Quadratic Average: 3.993140459060669
Nearest Class Center Accuracy: 0.447

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 3.454054594039917
Linear Weight Rank: 179
Intra Cos: 0.012448621913790703
Inter Cos: 0.0047308048233389854
Norm Quadratic Average: 5.418351650238037
Nearest Class Center Accuracy: 0.467125

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 3.1163153648376465
Linear Weight Rank: 4
Intra Cos: 0.04004846513271332
Inter Cos: -0.003547294996678829
Norm Quadratic Average: 6.132286071777344
Nearest Class Center Accuracy: 0.85725

Output Layer:
Intra Cos: 0.6167632937431335
Inter Cos: -0.03573692589998245
Norm Quadratic Average: 2.8981947898864746
Nearest Class Center Accuracy: 0.92925

Test Set:
Average Loss: 1.048278797149658
Accuracy: 0.551
NC1 Within Class Collapse: 9.586751937866211
NC2 Equinorm: Features: 0.07633659243583679, Weights: 0.023438194766640663
NC2 Equiangle: Features: 0.25755516688028973, Weights: 0.03751099109649658
NC3 Self-Duality: 0.42007389664649963
NC4 NCC Mismatch: 0.20999999999999996

Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 3.1982312202453613
Linear Weight Rank: 16
Intra Cos: 0.014108894392848015
Inter Cos: 0.005012289620935917
Norm Quadratic Average: 3.9761204719543457
Nearest Class Center Accuracy: 0.4485

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 3.454054594039917
Linear Weight Rank: 179
Intra Cos: 0.013677422888576984
Inter Cos: 0.004685812629759312
Norm Quadratic Average: 5.3954176902771
Nearest Class Center Accuracy: 0.4705

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 3.1163153648376465
Linear Weight Rank: 4
Intra Cos: 0.01556444726884365
Inter Cos: 0.0017796556930989027
Norm Quadratic Average: 6.1940836906433105
Nearest Class Center Accuracy: 0.577

Output Layer:
Intra Cos: 0.18298116326332092
Inter Cos: 0.03163691610097885
Norm Quadratic Average: 2.7532336711883545
Nearest Class Center Accuracy: 0.551

