Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_False_dataset_mlp3_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: 8.954665184020996
Linear Weight Rank: 16
Intra Cos: 0.03299567848443985
Inter Cos: -0.006222444586455822
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: 8.854998588562012
Linear Weight Rank: 179
Intra Cos: 0.032310616225004196
Inter Cos: -0.002853184472769499
Norm Quadratic Average: 5.302076816558838
Nearest Class Center Accuracy: 0.616625

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 3.7076363563537598
Linear Weight Rank: 4
Intra Cos: 0.1232985407114029
Inter Cos: -0.023815440014004707
Norm Quadratic Average: 3.2670252323150635
Nearest Class Center Accuracy: 0.73975

Output Layer:
Intra Cos: 0.32766100764274597
Inter Cos: -0.07094137370586395
Norm Quadratic Average: 4.064411163330078
Nearest Class Center Accuracy: 0.762125

Test Set:
Average Loss: 0.638997127532959
Accuracy: 0.7285
NC1 Within Class Collapse: 5.415771961212158
NC2 Equinorm: Features: 0.1263972520828247, Weights: 0.0445699580013752
NC2 Equiangle: Features: 0.18629562854766846, Weights: 0.035698890686035156
NC3 Self-Duality: 0.219407856464386
NC4 NCC Mismatch: 0.11050000000000004

Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 8.954665184020996
Linear Weight Rank: 16
Intra Cos: 0.030431468039751053
Inter Cos: -0.0045497240498661995
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: 8.854998588562012
Linear Weight Rank: 179
Intra Cos: 0.028653638437390327
Inter Cos: -0.002916717901825905
Norm Quadratic Average: 5.301656246185303
Nearest Class Center Accuracy: 0.615

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 3.7076363563537598
Linear Weight Rank: 4
Intra Cos: 0.10975006967782974
Inter Cos: -0.012560557574033737
Norm Quadratic Average: 3.2593955993652344
Nearest Class Center Accuracy: 0.714

Output Layer:
Intra Cos: 0.2960062325000763
Inter Cos: -0.0494672916829586
Norm Quadratic Average: 4.048253059387207
Nearest Class Center Accuracy: 0.73

