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_265358_weight_decay_0.05.pth.tar
Training Set:
Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 0.7442929744720459
Linear Weight Rank: 16
Intra Cos: 0.03299567475914955
Inter Cos: -0.006222445052117109
Norm Quadratic Average: 3.9938881397247314
Nearest Class Center Accuracy: 0.583375

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 0.7346653938293457
Linear Weight Rank: 179
Intra Cos: 0.028664425015449524
Inter Cos: -0.0039609260857105255
Norm Quadratic Average: 0.4364066421985626
Nearest Class Center Accuracy: 0.5775

Layer 2: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 0.7329845428466797
Linear Weight Rank: 179
Intra Cos: 0.03269634023308754
Inter Cos: -0.00387707632035017
Norm Quadratic Average: 0.016640981659293175
Nearest Class Center Accuracy: 0.53225

Layer 3: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 0.7352515459060669
Linear Weight Rank: 180
Intra Cos: 0.03270944580435753
Inter Cos: -0.0014426927082240582
Norm Quadratic Average: 0.00058176921447739
Nearest Class Center Accuracy: 0.520375

Layer 4: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 0.7327384352684021
Linear Weight Rank: 180
Intra Cos: 0.03503952547907829
Inter Cos: 0.001441589673049748
Norm Quadratic Average: 2.067306195385754e-05
Nearest Class Center Accuracy: 0.506125

Layer 5: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 0.11001203209161758
Linear Weight Rank: 4
Intra Cos: 0.031902529299259186
Inter Cos: -0.0007194594945758581
Norm Quadratic Average: 7.305638405341597e-07
Nearest Class Center Accuracy: 0.49875

Output Layer:
Intra Cos: 0.9914316534996033
Inter Cos: 0.9915264248847961
Norm Quadratic Average: 6.452019363223371e-08
Nearest Class Center Accuracy: 0.2805

Test Set:
Average Loss: 1.3859435234069823
Accuracy: 0.255
NC1 Within Class Collapse: 18.43944549560547
NC2 Equinorm: Features: 0.04269411787390709, Weights: 0.058984167873859406
NC2 Equiangle: Features: 0.19799745082855225, Weights: 0.26480331023534137
NC3 Self-Duality: 1.7120141983032227
NC4 NCC Mismatch: 0.7344999999999999

Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 0.7442929744720459
Linear Weight Rank: 16
Intra Cos: 0.030431468039751053
Inter Cos: -0.004549724515527487
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: 0.7346653938293457
Linear Weight Rank: 179
Intra Cos: 0.025955993682146072
Inter Cos: -0.0036797651555389166
Norm Quadratic Average: 0.43703535199165344
Nearest Class Center Accuracy: 0.5825

Layer 2: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 0.7329845428466797
Linear Weight Rank: 179
Intra Cos: 0.03201119974255562
Inter Cos: -0.0049909004010260105
Norm Quadratic Average: 0.01665644906461239
Nearest Class Center Accuracy: 0.537

Layer 3: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 0.7352515459060669
Linear Weight Rank: 180
Intra Cos: 0.029494166374206543
Inter Cos: -0.0012214318849146366
Norm Quadratic Average: 0.0005800699000246823
Nearest Class Center Accuracy: 0.5445

Layer 4: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 0.7327384352684021
Linear Weight Rank: 180
Intra Cos: 0.030489526689052582
Inter Cos: 0.0010831218678504229
Norm Quadratic Average: 2.053179377980996e-05
Nearest Class Center Accuracy: 0.514

Layer 5: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 0.11001203209161758
Linear Weight Rank: 4
Intra Cos: 0.028083382174372673
Inter Cos: -0.0009138062596321106
Norm Quadratic Average: 7.253051990119275e-07
Nearest Class Center Accuracy: 0.4985

Output Layer:
Intra Cos: 0.9899903535842896
Inter Cos: 0.9919521808624268
Norm Quadratic Average: 6.304690458591722e-08
Nearest Class Center Accuracy: 0.29

