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_979323_weight_decay_0.003.pth.tar
Training Set:
Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 7.154818058013916
Linear Weight Rank: 16
Intra Cos: 0.014583897776901722
Inter Cos: 0.004829857032746077
Norm Quadratic Average: 3.9931399822235107
Nearest Class Center Accuracy: 0.447

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 7.219695091247559
Linear Weight Rank: 179
Intra Cos: 0.012534662149846554
Inter Cos: 0.004856074694544077
Norm Quadratic Average: 12.261175155639648
Nearest Class Center Accuracy: 0.465375

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 3.7356979846954346
Linear Weight Rank: 4
Intra Cos: 0.017637863755226135
Inter Cos: -9.501251042820513e-05
Norm Quadratic Average: 12.709868431091309
Nearest Class Center Accuracy: 0.650875

Output Layer:
Intra Cos: 0.5709667801856995
Inter Cos: -0.019537895917892456
Norm Quadratic Average: 4.0521626472473145
Nearest Class Center Accuracy: 0.931375

Test Set:
Average Loss: 1.1668355255126952
Accuracy: 0.5455
NC1 Within Class Collapse: 13.96237564086914
NC2 Equinorm: Features: 0.058939363807439804, Weights: 0.009828025475144386
NC2 Equiangle: Features: 0.2260122299194336, Weights: 0.04364944497744242
NC3 Self-Duality: 0.9281269311904907
NC4 NCC Mismatch: 0.38349999999999995

Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 7.154818058013916
Linear Weight Rank: 16
Intra Cos: 0.014108890667557716
Inter Cos: 0.005012288689613342
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: 7.219695091247559
Linear Weight Rank: 179
Intra Cos: 0.012296793051064014
Inter Cos: 0.004164937883615494
Norm Quadratic Average: 12.20483112335205
Nearest Class Center Accuracy: 0.4695

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 3.7356979846954346
Linear Weight Rank: 4
Intra Cos: 0.01090710237622261
Inter Cos: 0.0008474704809486866
Norm Quadratic Average: 12.693253517150879
Nearest Class Center Accuracy: 0.53

Output Layer:
Intra Cos: 0.16572415828704834
Inter Cos: 0.03523372486233711
Norm Quadratic Average: 3.9319121837615967
Nearest Class Center Accuracy: 0.543

