Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_False_dataset_conic_epochs_200_lr_0.001_model_depth_MLP_3_model_type_MLP_rand_seed_265358_weight_decay_0.0001.pth.tar
Training Set:
Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 9.018196105957031
Linear Weight Rank: 16
Intra Cos: 0.07988916337490082
Inter Cos: 0.0017584804445505142
Norm Quadratic Average: 4.001680374145508
Nearest Class Center Accuracy: 0.948875

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 9.025965690612793
Linear Weight Rank: 179
Intra Cos: 0.17373788356781006
Inter Cos: 0.007054587826132774
Norm Quadratic Average: 5.577291011810303
Nearest Class Center Accuracy: 0.954875

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 4.106637954711914
Linear Weight Rank: 4
Intra Cos: 0.623231053352356
Inter Cos: 0.0337786041200161
Norm Quadratic Average: 7.117480754852295
Nearest Class Center Accuracy: 0.9805

Output Layer:
Intra Cos: 0.8099321126937866
Inter Cos: 0.033238623291254044
Norm Quadratic Average: 18.20486068725586
Nearest Class Center Accuracy: 0.98225

Test Set:
Average Loss: 0.08683721399307251
Accuracy: 0.9755
NC1 Within Class Collapse: 2.735999822616577
NC2 Equinorm: Features: 0.02657366916537285, Weights: 0.023434730246663094
NC2 Equiangle: Features: 0.4320613145828247, Weights: 0.3870615561803182
NC3 Self-Duality: 0.05573245882987976
NC4 NCC Mismatch: 0.023499999999999965

Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 9.018196105957031
Linear Weight Rank: 16
Intra Cos: 0.0803302749991417
Inter Cos: 0.0033129025250673294
Norm Quadratic Average: 3.9883358478546143
Nearest Class Center Accuracy: 0.9165

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 9.025965690612793
Linear Weight Rank: 179
Intra Cos: 0.17616155743598938
Inter Cos: 0.003380928188562393
Norm Quadratic Average: 5.583043098449707
Nearest Class Center Accuracy: 0.953

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 4.106637954711914
Linear Weight Rank: 4
Intra Cos: 0.6200748085975647
Inter Cos: 0.02853388898074627
Norm Quadratic Average: 7.180741786956787
Nearest Class Center Accuracy: 0.965

Output Layer:
Intra Cos: 0.7968484163284302
Inter Cos: 0.03482567146420479
Norm Quadratic Average: 18.35464859008789
Nearest Class Center Accuracy: 0.969

