Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_265358_test_samples_None_train_samples_None_weight_decay_0.005.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09116753190755844
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567670822143555
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12295698374509811
Inter Cos: 0.1501161754131317
Norm Quadratic Average: 39.80167007446289
Nearest Class Center Accuracy: 0.80635

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17546817660331726
Inter Cos: 0.17814332246780396
Norm Quadratic Average: 40.67475891113281
Nearest Class Center Accuracy: 0.81715

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20658759772777557
Inter Cos: 0.21388722956180573
Norm Quadratic Average: 39.66675567626953
Nearest Class Center Accuracy: 0.8615

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1940954625606537
Inter Cos: 0.23225639760494232
Norm Quadratic Average: 19.4941349029541
Nearest Class Center Accuracy: 0.9058

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2540404796600342
Inter Cos: 0.27407971024513245
Norm Quadratic Average: 11.837469100952148
Nearest Class Center Accuracy: 0.93545

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4872259199619293
Inter Cos: 0.3490757942199707
Norm Quadratic Average: 6.874314308166504
Nearest Class Center Accuracy: 0.9728333333333333

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7144479155540466
Inter Cos: 0.3832879066467285
Norm Quadratic Average: 7.026821136474609
Nearest Class Center Accuracy: 0.9873166666666666

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.982290506362915
Linear Weight Rank: 40
Intra Cos: 0.8041614890098572
Inter Cos: 0.38238397240638733
Norm Quadratic Average: 34.45301055908203
Nearest Class Center Accuracy: 0.9937

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.984706163406372
Linear Weight Rank: 2753
Intra Cos: 0.8744425177574158
Inter Cos: 0.4082576632499695
Norm Quadratic Average: 27.45507049560547
Nearest Class Center Accuracy: 0.9962166666666666

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.9778801202774048
Linear Weight Rank: 9
Intra Cos: 0.900309145450592
Inter Cos: 0.38770592212677
Norm Quadratic Average: 21.123001098632812
Nearest Class Center Accuracy: 0.99705

Output Layer:
Intra Cos: 0.9288296699523926
Inter Cos: 0.45727574825286865
Norm Quadratic Average: 18.47223663330078
Nearest Class Center Accuracy: 0.9971666666666666

Test Set:
Average Loss: 0.02714179635345936
Accuracy: 0.9903
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.11097188293933868, Weights: 0.048208266496658325
NC2 Equiangle: Features: 0.27096659342447915, Weights: 0.24534481896294488
NC3 Self-Duality: 0.056752827018499374
NC4 NCC Mismatch: 0.0033999999999999586

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219583660364151
Inter Cos: 0.12048852443695068
Norm Quadratic Average: 23.595197677612305
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1364704668521881
Inter Cos: 0.16448234021663666
Norm Quadratic Average: 39.84952926635742
Nearest Class Center Accuracy: 0.8212

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19163648784160614
Inter Cos: 0.1928413361310959
Norm Quadratic Average: 40.61015701293945
Nearest Class Center Accuracy: 0.8336

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22054903209209442
Inter Cos: 0.23280587792396545
Norm Quadratic Average: 39.62950897216797
Nearest Class Center Accuracy: 0.8757

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20412449538707733
Inter Cos: 0.2271333485841751
Norm Quadratic Average: 19.461763381958008
Nearest Class Center Accuracy: 0.9195

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.264064222574234
Inter Cos: 0.29720714688301086
Norm Quadratic Average: 11.836831092834473
Nearest Class Center Accuracy: 0.9413

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.501508891582489
Inter Cos: 0.3774046003818512
Norm Quadratic Average: 6.898225784301758
Nearest Class Center Accuracy: 0.9704

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7233220338821411
Inter Cos: 0.4070667624473572
Norm Quadratic Average: 7.078099727630615
Nearest Class Center Accuracy: 0.983

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.982290506362915
Linear Weight Rank: 40
Intra Cos: 0.8021774888038635
Inter Cos: 0.4023016095161438
Norm Quadratic Average: 34.76894760131836
Nearest Class Center Accuracy: 0.9887

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.984706163406372
Linear Weight Rank: 2753
Intra Cos: 0.8768965601921082
Inter Cos: 0.4274611175060272
Norm Quadratic Average: 27.728696823120117
Nearest Class Center Accuracy: 0.9898

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.9778801202774048
Linear Weight Rank: 9
Intra Cos: 0.900771975517273
Inter Cos: 0.40566644072532654
Norm Quadratic Average: 21.329511642456055
Nearest Class Center Accuracy: 0.9899

Output Layer:
Intra Cos: 0.9176531434059143
Inter Cos: 0.47313860058784485
Norm Quadratic Average: 18.655399322509766
Nearest Class Center Accuracy: 0.99

