Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_979323_test_samples_None_train_samples_None_weight_decay_0.0005.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09116753935813904
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567676544189453
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1187586858868599
Inter Cos: 0.13971763849258423
Norm Quadratic Average: 39.414634704589844
Nearest Class Center Accuracy: 0.8211166666666667

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1874927133321762
Inter Cos: 0.17531368136405945
Norm Quadratic Average: 38.85490798950195
Nearest Class Center Accuracy: 0.8668833333333333

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21824435889720917
Inter Cos: 0.19147028028964996
Norm Quadratic Average: 37.671566009521484
Nearest Class Center Accuracy: 0.9042166666666667

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.24356307089328766
Inter Cos: 0.1694757491350174
Norm Quadratic Average: 17.33605194091797
Nearest Class Center Accuracy: 0.9510666666666666

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3966430425643921
Inter Cos: 0.2515263855457306
Norm Quadratic Average: 10.434883117675781
Nearest Class Center Accuracy: 0.9757

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.548947811126709
Inter Cos: 0.3322559893131256
Norm Quadratic Average: 5.5319976806640625
Nearest Class Center Accuracy: 0.9917

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8075714707374573
Inter Cos: 0.36379945278167725
Norm Quadratic Average: 4.7108683586120605
Nearest Class Center Accuracy: 0.9979833333333333

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.764206886291504
Linear Weight Rank: 4031
Intra Cos: 0.8825559020042419
Inter Cos: 0.2981448173522949
Norm Quadratic Average: 26.23912811279297
Nearest Class Center Accuracy: 0.9988

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.449947357177734
Linear Weight Rank: 3666
Intra Cos: 0.9144703149795532
Inter Cos: 0.2702604830265045
Norm Quadratic Average: 23.610023498535156
Nearest Class Center Accuracy: 0.9993833333333333

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.61690616607666
Linear Weight Rank: 10
Intra Cos: 0.9209621548652649
Inter Cos: 0.2284391075372696
Norm Quadratic Average: 21.046918869018555
Nearest Class Center Accuracy: 0.9997

Output Layer:
Intra Cos: 0.9408663511276245
Inter Cos: 0.2791685163974762
Norm Quadratic Average: 21.042205810546875
Nearest Class Center Accuracy: 0.9999333333333333

Test Set:
Average Loss: 0.020178263679874364
Accuracy: 0.9936
NC1 Within Class Collapse: 0.6489834189414978
NC2 Equinorm: Features: 0.10998733341693878, Weights: 0.03348829597234726
NC2 Equiangle: Features: 0.21952537960476345, Weights: 0.1316163804796007
NC3 Self-Duality: 0.1127689778804779
NC4 NCC Mismatch: 0.0027000000000000357

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219582170248032
Inter Cos: 0.12048851698637009
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: 0.13172025978565216
Inter Cos: 0.1484423726797104
Norm Quadratic Average: 39.36421203613281
Nearest Class Center Accuracy: 0.8358

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2009761482477188
Inter Cos: 0.17036297917366028
Norm Quadratic Average: 38.717201232910156
Nearest Class Center Accuracy: 0.8811

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.23156388103961945
Inter Cos: 0.18590223789215088
Norm Quadratic Average: 37.576595306396484
Nearest Class Center Accuracy: 0.9154

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2545679211616516
Inter Cos: 0.1834387630224228
Norm Quadratic Average: 17.30547332763672
Nearest Class Center Accuracy: 0.9595

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.41064709424972534
Inter Cos: 0.2698015868663788
Norm Quadratic Average: 10.445467948913574
Nearest Class Center Accuracy: 0.9783

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.556952714920044
Inter Cos: 0.3498796224594116
Norm Quadratic Average: 5.56353235244751
Nearest Class Center Accuracy: 0.9876

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8060806393623352
Inter Cos: 0.38076284527778625
Norm Quadratic Average: 4.750782489776611
Nearest Class Center Accuracy: 0.9908

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 14.764206886291504
Linear Weight Rank: 4031
Intra Cos: 0.8810794353485107
Inter Cos: 0.3135480284690857
Norm Quadratic Average: 26.440078735351562
Nearest Class Center Accuracy: 0.9919

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 6.449947357177734
Linear Weight Rank: 3666
Intra Cos: 0.9112116694450378
Inter Cos: 0.2860250771045685
Norm Quadratic Average: 23.78305435180664
Nearest Class Center Accuracy: 0.9926

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 2.61690616607666
Linear Weight Rank: 10
Intra Cos: 0.9169130921363831
Inter Cos: 0.24406744539737701
Norm Quadratic Average: 21.200204849243164
Nearest Class Center Accuracy: 0.9927

Output Layer:
Intra Cos: 0.9330269694328308
Inter Cos: 0.27119752764701843
Norm Quadratic Average: 21.192584991455078
Nearest Class Center Accuracy: 0.993

