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_265358_weight_decay_0.0001.pth.tar
Training Set:
Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 8.265368461608887
Linear Weight Rank: 16
Intra Cos: 0.014583894982933998
Inter Cos: 0.004829855635762215
Norm Quadratic Average: 3.9931397438049316
Nearest Class Center Accuracy: 0.447

Layer 1: Linear(in_features=200, out_features=200, bias=True)
Linear Weight Norm: 8.297325134277344
Linear Weight Rank: 179
Intra Cos: 0.011957054026424885
Inter Cos: 0.002580116968601942
Norm Quadratic Average: 14.046919822692871
Nearest Class Center Accuracy: 0.464625

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 3.8542068004608154
Linear Weight Rank: 4
Intra Cos: 0.014966793358325958
Inter Cos: 0.0007030398119240999
Norm Quadratic Average: 14.51003646850586
Nearest Class Center Accuracy: 0.605625

Output Layer:
Intra Cos: 0.5336436629295349
Inter Cos: -0.03038257360458374
Norm Quadratic Average: 4.340514183044434
Nearest Class Center Accuracy: 0.92875

Test Set:
Average Loss: 1.1824491958618164
Accuracy: 0.551
NC1 Within Class Collapse: 15.410301208496094
NC2 Equinorm: Features: 0.018737994134426117, Weights: 0.022540520876646042
NC2 Equiangle: Features: 0.2605423331260681, Weights: 0.038187178472677864
NC3 Self-Duality: 1.0389580726623535
NC4 NCC Mismatch: 0.41300000000000003

Layer 0: Linear(in_features=16, out_features=200, bias=True)
Linear Weight Norm: 8.265368461608887
Linear Weight Rank: 16
Intra Cos: 0.014108894392848015
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: 8.297325134277344
Linear Weight Rank: 179
Intra Cos: 0.012823415920138359
Inter Cos: 0.0027949747163802385
Norm Quadratic Average: 14.017266273498535
Nearest Class Center Accuracy: 0.4755

Layer 2: Linear(in_features=200, out_features=4, bias=True)
Linear Weight Norm: 3.8542068004608154
Linear Weight Rank: 4
Intra Cos: 0.010523132979869843
Inter Cos: 0.0016113596502691507
Norm Quadratic Average: 14.510043144226074
Nearest Class Center Accuracy: 0.523

Output Layer:
Intra Cos: 0.15624992549419403
Inter Cos: 0.019485503435134888
Norm Quadratic Average: 4.1510515213012695
Nearest Class Center Accuracy: 0.5565

