Model save path: ./New_Models/bn_True_dataset_CIFAR100_epochs_200_lr_0.01_model_type_vgg19_rand_seed_265358_test_samples_None_train_samples_None_weight_decay_0.0005.pth.tar
Training Set:
Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.022893192246556282
Inter Cos: 0.07011394202709198
Norm Quadratic Average: 29.326313018798828
Nearest Class Center Accuracy: 0.03258

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.01454498153179884
Inter Cos: 0.028051791712641716
Norm Quadratic Average: 43.966373443603516
Nearest Class Center Accuracy: 0.04068

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.019342029467225075
Inter Cos: 0.03735949099063873
Norm Quadratic Average: 28.99106788635254
Nearest Class Center Accuracy: 0.0578

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.015572982840240002
Inter Cos: 0.027040569111704826
Norm Quadratic Average: 23.414199829101562
Nearest Class Center Accuracy: 0.06364

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02119072526693344
Inter Cos: 0.021320125088095665
Norm Quadratic Average: 16.371532440185547
Nearest Class Center Accuracy: 0.07064

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.021453237161040306
Inter Cos: 0.021021481603384018
Norm Quadratic Average: 13.474222183227539
Nearest Class Center Accuracy: 0.07464

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02245129458606243
Inter Cos: 0.021561017259955406
Norm Quadratic Average: 13.157085418701172
Nearest Class Center Accuracy: 0.07714

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.022778665646910667
Inter Cos: 0.021731261163949966
Norm Quadratic Average: 13.639869689941406
Nearest Class Center Accuracy: 0.07974

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.03308027610182762
Inter Cos: 0.030184851959347725
Norm Quadratic Average: 8.581067085266113
Nearest Class Center Accuracy: 0.08548

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.04491358622908592
Inter Cos: 0.03517427295446396
Norm Quadratic Average: 6.852084159851074
Nearest Class Center Accuracy: 0.08954

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.062411654740571976
Inter Cos: 0.044323086738586426
Norm Quadratic Average: 6.27793550491333
Nearest Class Center Accuracy: 0.09346

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0901658833026886
Inter Cos: 0.06355642527341843
Norm Quadratic Average: 6.380245685577393
Nearest Class Center Accuracy: 0.09694

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21278759837150574
Inter Cos: 0.12902413308620453
Norm Quadratic Average: 5.378062725067139
Nearest Class Center Accuracy: 0.09866

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5272836089134216
Inter Cos: 0.24557720124721527
Norm Quadratic Average: 4.833096504211426
Nearest Class Center Accuracy: 0.1

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7080875039100647
Inter Cos: 0.23102572560310364
Norm Quadratic Average: 5.197561740875244
Nearest Class Center Accuracy: 0.1

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8101635575294495
Inter Cos: 0.26757827401161194
Norm Quadratic Average: 5.423315525054932
Nearest Class Center Accuracy: 0.1

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 21.067270278930664
Linear Weight Rank: 4030
Intra Cos: 0.9112316370010376
Inter Cos: 0.33309608697891235
Norm Quadratic Average: 40.47107696533203
Nearest Class Center Accuracy: 0.1

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 11.333070755004883
Linear Weight Rank: 3648
Intra Cos: 0.9370195269584656
Inter Cos: 0.3414168953895569
Norm Quadratic Average: 32.18444061279297
Nearest Class Center Accuracy: 0.1

Layer 18: Linear(in_features=4096, out_features=100, bias=True)
Linear Weight Norm: 7.600142002105713
Linear Weight Rank: 98
Intra Cos: 0.9423956274986267
Inter Cos: 0.33172738552093506
Norm Quadratic Average: 30.035964965820312
Nearest Class Center Accuracy: 0.1

Output Layer:
Intra Cos: 0.9574094414710999
Inter Cos: 0.44031697511672974
Norm Quadratic Average: 40.51712417602539
Nearest Class Center Accuracy: 0.1

Test Set:
Average Loss: 2.540149129486084
Accuracy: 0.5786
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.20435312390327454, Weights: 0.03042699582874775
NC2 Equiangle: Features: 0.18912681502525253, Weights: 0.10817848899147728
NC3 Self-Duality: 0.3868385851383209
NC4 NCC Mismatch: 0.07369999999999999

Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.006621268577873707
Inter Cos: 0.4067547917366028
Norm Quadratic Average: 29.422183990478516
Nearest Class Center Accuracy: 0.1136

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.006999407894909382
Inter Cos: 0.18625874817371368
Norm Quadratic Average: 44.24633026123047
Nearest Class Center Accuracy: 0.2284

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.011288159526884556
Inter Cos: 0.24233928322792053
Norm Quadratic Average: 29.1818904876709
Nearest Class Center Accuracy: 0.3643

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.010887586511671543
Inter Cos: 0.1470932960510254
Norm Quadratic Average: 23.521026611328125
Nearest Class Center Accuracy: 0.4713

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.015112725086510181
Inter Cos: 0.18078722059726715
Norm Quadratic Average: 16.43482780456543
Nearest Class Center Accuracy: 0.5208

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.015195184387266636
Inter Cos: 0.1511354297399521
Norm Quadratic Average: 13.510069847106934
Nearest Class Center Accuracy: 0.562

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.015229561366140842
Inter Cos: 0.1348492056131363
Norm Quadratic Average: 13.172204971313477
Nearest Class Center Accuracy: 0.5895

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.01461332943290472
Inter Cos: 0.11612888425588608
Norm Quadratic Average: 13.636645317077637
Nearest Class Center Accuracy: 0.6031

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.01654611900448799
Inter Cos: 0.14673076570034027
Norm Quadratic Average: 8.562634468078613
Nearest Class Center Accuracy: 0.6359

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.019140440970659256
Inter Cos: 0.16237938404083252
Norm Quadratic Average: 6.7906599044799805
Nearest Class Center Accuracy: 0.6545

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.023872606456279755
Inter Cos: 0.20252376794815063
Norm Quadratic Average: 6.143660068511963
Nearest Class Center Accuracy: 0.6433

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.030531171709299088
Inter Cos: 0.26577046513557434
Norm Quadratic Average: 6.152510643005371
Nearest Class Center Accuracy: 0.634

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.06707550585269928
Inter Cos: 0.38632431626319885
Norm Quadratic Average: 5.032843112945557
Nearest Class Center Accuracy: 0.5982

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13691914081573486
Inter Cos: 0.5154980421066284
Norm Quadratic Average: 4.303289890289307
Nearest Class Center Accuracy: 0.5857

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18017050623893738
Inter Cos: 0.5001141428947449
Norm Quadratic Average: 4.483129024505615
Nearest Class Center Accuracy: 0.5833

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18932124972343445
Inter Cos: 0.4681190252304077
Norm Quadratic Average: 4.604529857635498
Nearest Class Center Accuracy: 0.5846

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 21.067270278930664
Linear Weight Rank: 4030
Intra Cos: 0.2316696047782898
Inter Cos: 0.4617159366607666
Norm Quadratic Average: 33.15735626220703
Nearest Class Center Accuracy: 0.581

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 11.333070755004883
Linear Weight Rank: 3648
Intra Cos: 0.2373812049627304
Inter Cos: 0.4718998968601227
Norm Quadratic Average: 26.70684051513672
Nearest Class Center Accuracy: 0.5833

Layer 18: Linear(in_features=4096, out_features=100, bias=True)
Linear Weight Norm: 7.600142002105713
Linear Weight Rank: 98
Intra Cos: 0.24412275850772858
Inter Cos: 0.5171253681182861
Norm Quadratic Average: 25.479761123657227
Nearest Class Center Accuracy: 0.5827

Output Layer:
Intra Cos: 0.26214468479156494
Inter Cos: 0.6152822375297546
Norm Quadratic Average: 34.72915267944336
Nearest Class Center Accuracy: 0.5801

