Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_314159_test_samples_None_train_samples_None_weight_decay_0.0003.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09116754680871964
Inter Cos: 0.10967153310775757
Norm Quadratic Average: 23.567678451538086
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.11383040249347687
Inter Cos: 0.13401708006858826
Norm Quadratic Average: 37.811798095703125
Nearest Class Center Accuracy: 0.8236333333333333

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19094426929950714
Inter Cos: 0.16701705753803253
Norm Quadratic Average: 36.028018951416016
Nearest Class Center Accuracy: 0.87455

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2216026782989502
Inter Cos: 0.18122075498104095
Norm Quadratic Average: 35.68181610107422
Nearest Class Center Accuracy: 0.90685

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2577511668205261
Inter Cos: 0.1630961298942566
Norm Quadratic Average: 16.475831985473633
Nearest Class Center Accuracy: 0.9565833333333333

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3818392753601074
Inter Cos: 0.22342032194137573
Norm Quadratic Average: 10.905954360961914
Nearest Class Center Accuracy: 0.9773166666666666

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5204988121986389
Inter Cos: 0.28172817826271057
Norm Quadratic Average: 5.509700775146484
Nearest Class Center Accuracy: 0.9932

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8083550930023193
Inter Cos: 0.3344404399394989
Norm Quadratic Average: 4.587822914123535
Nearest Class Center Accuracy: 0.9981833333333333

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.72406768798828
Linear Weight Rank: 4031
Intra Cos: 0.8916999697685242
Inter Cos: 0.26937398314476013
Norm Quadratic Average: 26.001352310180664
Nearest Class Center Accuracy: 0.9991166666666667

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 13.069181442260742
Linear Weight Rank: 3668
Intra Cos: 0.9228048920631409
Inter Cos: 0.23574332892894745
Norm Quadratic Average: 23.311019897460938
Nearest Class Center Accuracy: 0.9995833333333334

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.774962902069092
Linear Weight Rank: 10
Intra Cos: 0.929141104221344
Inter Cos: 0.22650381922721863
Norm Quadratic Average: 21.513887405395508
Nearest Class Center Accuracy: 0.9997

Output Layer:
Intra Cos: 0.9501944780349731
Inter Cos: 0.31486430764198303
Norm Quadratic Average: 21.78328514099121
Nearest Class Center Accuracy: 0.99995

Test Set:
Average Loss: 0.02000261303615989
Accuracy: 0.9938
NC1 Within Class Collapse: 0.6493574380874634
NC2 Equinorm: Features: 0.11521485447883606, Weights: 0.03642136976122856
NC2 Equiangle: Features: 0.2036860360039605, Weights: 0.11136913299560547
NC3 Self-Duality: 0.13186167180538177
NC4 NCC Mismatch: 0.0028000000000000247

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582915306091
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: 0.12674546241760254
Inter Cos: 0.14354881644248962
Norm Quadratic Average: 37.743648529052734
Nearest Class Center Accuracy: 0.8377

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20525303483009338
Inter Cos: 0.17297245562076569
Norm Quadratic Average: 35.88336944580078
Nearest Class Center Accuracy: 0.8868

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23527704179286957
Inter Cos: 0.18994852900505066
Norm Quadratic Average: 35.57237243652344
Nearest Class Center Accuracy: 0.9182

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26971232891082764
Inter Cos: 0.17593608796596527
Norm Quadratic Average: 16.445941925048828
Nearest Class Center Accuracy: 0.9626

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3960139751434326
Inter Cos: 0.23966914415359497
Norm Quadratic Average: 10.907432556152344
Nearest Class Center Accuracy: 0.9788

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5299924612045288
Inter Cos: 0.29646816849708557
Norm Quadratic Average: 5.534312725067139
Nearest Class Center Accuracy: 0.9878

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8168989419937134
Inter Cos: 0.3494962155818939
Norm Quadratic Average: 4.6239519119262695
Nearest Class Center Accuracy: 0.9915

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 31.72406768798828
Linear Weight Rank: 4031
Intra Cos: 0.8939698338508606
Inter Cos: 0.28237494826316833
Norm Quadratic Average: 26.214019775390625
Nearest Class Center Accuracy: 0.9924

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 13.069181442260742
Linear Weight Rank: 3668
Intra Cos: 0.9232731461524963
Inter Cos: 0.24907247722148895
Norm Quadratic Average: 23.49694061279297
Nearest Class Center Accuracy: 0.9925

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.774962902069092
Linear Weight Rank: 10
Intra Cos: 0.9290356040000916
Inter Cos: 0.21994268894195557
Norm Quadratic Average: 21.681682586669922
Nearest Class Center Accuracy: 0.993

Output Layer:
Intra Cos: 0.9455760717391968
Inter Cos: 0.30777668952941895
Norm Quadratic Average: 21.946861267089844
Nearest Class Center Accuracy: 0.9936

