Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_False_dataset_mlp3_epochs_100_lr_0.001_model_depth_MLP_6_model_type_MLP_rand_seed_979323_weight_decay_0.03.pth.tar
Training Set:
Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 1.9454264640808105
Linear Weight Rank: 16
Intra Cos: 0.03299567475914955
Inter Cos: -0.006222445052117109
Norm Quadratic Average: 3.9938879013061523
Nearest Class Center Accuracy: 0.583375

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 1.9250108003616333
Linear Weight Rank: 180
Intra Cos: 0.023499373346567154
Inter Cos: -0.004835592582821846
Norm Quadratic Average: 1.1620216369628906
Nearest Class Center Accuracy: 0.564375

Layer 2: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 1.9231802225112915
Linear Weight Rank: 180
Intra Cos: 0.024107342585921288
Inter Cos: -0.0027206488884985447
Norm Quadratic Average: 0.10380617529153824
Nearest Class Center Accuracy: 0.53975

Layer 3: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 1.9358229637145996
Linear Weight Rank: 179
Intra Cos: 0.023592591285705566
Inter Cos: -0.002411632565781474
Norm Quadratic Average: 0.00993256364017725
Nearest Class Center Accuracy: 0.522375

Layer 4: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 1.9270598888397217
Linear Weight Rank: 180
Intra Cos: 0.026207737624645233
Inter Cos: 0.0022452250123023987
Norm Quadratic Average: 0.000995624577626586
Nearest Class Center Accuracy: 0.504

Layer 5: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 0.29103177785873413
Linear Weight Rank: 4
Intra Cos: 0.028008442372083664
Inter Cos: 0.0016969620482996106
Norm Quadratic Average: 0.00010309126082574949
Nearest Class Center Accuracy: 0.478625

Output Layer:
Intra Cos: 0.04550742357969284
Inter Cos: 0.012241062708199024
Norm Quadratic Average: 4.356124463811284e-06
Nearest Class Center Accuracy: 0.444

Test Set:
Average Loss: 1.3862349700927734
Accuracy: 0.255
NC1 Within Class Collapse: 20.00733184814453
NC2 Equinorm: Features: 0.3032669425010681, Weights: 0.06568116694688797
NC2 Equiangle: Features: 0.2130586306254069, Weights: 0.30386171738306683
NC3 Self-Duality: 1.3228199481964111
NC4 NCC Mismatch: 0.8325

Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 1.9454264640808105
Linear Weight Rank: 16
Intra Cos: 0.030431468039751053
Inter Cos: -0.00454972218722105
Norm Quadratic Average: 3.9951601028442383
Nearest Class Center Accuracy: 0.583

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 1.9250108003616333
Linear Weight Rank: 180
Intra Cos: 0.022245826199650764
Inter Cos: -0.004234869033098221
Norm Quadratic Average: 1.1592352390289307
Nearest Class Center Accuracy: 0.581

Layer 2: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 1.9231802225112915
Linear Weight Rank: 180
Intra Cos: 0.023393472656607628
Inter Cos: -0.0035592387430369854
Norm Quadratic Average: 0.10325049608945847
Nearest Class Center Accuracy: 0.5495

Layer 3: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 1.9358229637145996
Linear Weight Rank: 179
Intra Cos: 0.02312428131699562
Inter Cos: -0.0027258528862148523
Norm Quadratic Average: 0.009880386292934418
Nearest Class Center Accuracy: 0.5225

Layer 4: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 1.9270598888397217
Linear Weight Rank: 180
Intra Cos: 0.025572525337338448
Inter Cos: 0.0014880250673741102
Norm Quadratic Average: 0.0009877869160845876
Nearest Class Center Accuracy: 0.5145

Layer 5: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 0.29103177785873413
Linear Weight Rank: 4
Intra Cos: 0.026998315006494522
Inter Cos: -3.601838761824183e-05
Norm Quadratic Average: 0.0001025154342642054
Nearest Class Center Accuracy: 0.4935

Output Layer:
Intra Cos: 0.044667668640613556
Inter Cos: 0.00475424574688077
Norm Quadratic Average: 4.326048838265706e-06
Nearest Class Center Accuracy: 0.4585

