Model save path: ./New_Models/bn_False_dataset_CIFAR10_epochs_200_lr_0.01_model_type_vgg19_rand_seed_265358_test_samples_None_train_samples_None_weight_decay_0.0001.pth.tar
Training Set:
Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.01989092119038105
Inter Cos: 0.10477276891469955
Norm Quadratic Average: 27.597177505493164
Nearest Class Center Accuracy: 0.26968

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.017095226794481277
Inter Cos: 0.06900107860565186
Norm Quadratic Average: 48.07049560546875
Nearest Class Center Accuracy: 0.35492

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.017285456880927086
Inter Cos: 0.05692107975482941
Norm Quadratic Average: 56.755245208740234
Nearest Class Center Accuracy: 0.49212

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.02065141126513481
Inter Cos: 0.047984182834625244
Norm Quadratic Average: 82.87820434570312
Nearest Class Center Accuracy: 0.56884

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.023447096347808838
Inter Cos: 0.050043292343616486
Norm Quadratic Average: 60.940242767333984
Nearest Class Center Accuracy: 0.63306

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.026084866374731064
Inter Cos: 0.04467713460326195
Norm Quadratic Average: 62.622955322265625
Nearest Class Center Accuracy: 0.67272

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.03153340145945549
Inter Cos: 0.042220670729875565
Norm Quadratic Average: 54.21736145019531
Nearest Class Center Accuracy: 0.71128

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.0418626070022583
Inter Cos: 0.047929808497428894
Norm Quadratic Average: 49.36967849731445
Nearest Class Center Accuracy: 0.7366

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.07616747170686722
Inter Cos: 0.06899162381887436
Norm Quadratic Average: 25.36836814880371
Nearest Class Center Accuracy: 0.80278

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1337687373161316
Inter Cos: 0.1003982350230217
Norm Quadratic Average: 18.75311279296875
Nearest Class Center Accuracy: 0.83532

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2296750396490097
Inter Cos: 0.15557831525802612
Norm Quadratic Average: 12.857633590698242
Nearest Class Center Accuracy: 0.877

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.3473062217235565
Inter Cos: 0.20417499542236328
Norm Quadratic Average: 9.742527961730957
Nearest Class Center Accuracy: 0.904

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5604883432388306
Inter Cos: 0.35729363560676575
Norm Quadratic Average: 6.142980575561523
Nearest Class Center Accuracy: 0.91292

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7868043184280396
Inter Cos: 0.35642439126968384
Norm Quadratic Average: 5.609342575073242
Nearest Class Center Accuracy: 0.9392

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.8633365035057068
Inter Cos: 0.296516478061676
Norm Quadratic Average: 5.28830623626709
Nearest Class Center Accuracy: 0.95898

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.9027298092842102
Inter Cos: 0.24499452114105225
Norm Quadratic Average: 4.5753397941589355
Nearest Class Center Accuracy: 0.98238

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 73.43518829345703
Linear Weight Rank: 4031
Intra Cos: 0.9290028214454651
Inter Cos: 0.17635071277618408
Norm Quadratic Average: 25.8895263671875
Nearest Class Center Accuracy: 0.99402

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 29.96353530883789
Linear Weight Rank: 3670
Intra Cos: 0.9327285289764404
Inter Cos: 0.16304698586463928
Norm Quadratic Average: 21.603172302246094
Nearest Class Center Accuracy: 0.99716

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 4.024632930755615
Linear Weight Rank: 10
Intra Cos: 0.929174542427063
Inter Cos: 0.12971842288970947
Norm Quadratic Average: 19.664426803588867
Nearest Class Center Accuracy: 0.99814

Output Layer:
Intra Cos: 0.9723761081695557
Inter Cos: 0.33508291840553284
Norm Quadratic Average: 24.25532341003418
Nearest Class Center Accuracy: 0.99958

Test Set:
Average Loss: 1.1197999616622925
Accuracy: 0.8344
NC1 Within Class Collapse: 5.129049301147461
NC2 Equinorm: Features: 0.2847326099872589, Weights: 0.023290643468499184
NC2 Equiangle: Features: 0.1928075154622396, Weights: 0.16845264434814453
NC3 Self-Duality: 0.46811118721961975
NC4 NCC Mismatch: 0.07450000000000001

Layer 0: Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.018526073545217514
Inter Cos: 0.1042705774307251
Norm Quadratic Average: 27.55013656616211
Nearest Class Center Accuracy: 0.284

Layer 1: Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.01503691915422678
Inter Cos: 0.06847459822893143
Norm Quadratic Average: 48.025054931640625
Nearest Class Center Accuracy: 0.3757

Layer 2: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.016165317967534065
Inter Cos: 0.05763956904411316
Norm Quadratic Average: 56.720436096191406
Nearest Class Center Accuracy: 0.5063

Layer 3: Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.01960667595267296
Inter Cos: 0.048022232949733734
Norm Quadratic Average: 82.86656188964844
Nearest Class Center Accuracy: 0.5838

Layer 4: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.021575896069407463
Inter Cos: 0.05056547746062279
Norm Quadratic Average: 60.96892547607422
Nearest Class Center Accuracy: 0.6381

Layer 5: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.023891203105449677
Inter Cos: 0.04507250338792801
Norm Quadratic Average: 62.68454360961914
Nearest Class Center Accuracy: 0.6639

Layer 6: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.028371665626764297
Inter Cos: 0.04464185982942581
Norm Quadratic Average: 54.26823425292969
Nearest Class Center Accuracy: 0.6887

Layer 7: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.037255480885505676
Inter Cos: 0.04936278611421585
Norm Quadratic Average: 49.382869720458984
Nearest Class Center Accuracy: 0.7001

Layer 8: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.06491123884916306
Inter Cos: 0.07077936083078384
Norm Quadratic Average: 25.316661834716797
Nearest Class Center Accuracy: 0.7328

Layer 9: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1203286424279213
Inter Cos: 0.100214883685112
Norm Quadratic Average: 18.634971618652344
Nearest Class Center Accuracy: 0.7389

Layer 10: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19298183917999268
Inter Cos: 0.16367694735527039
Norm Quadratic Average: 12.709357261657715
Nearest Class Center Accuracy: 0.7489

Layer 11: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.27865222096443176
Inter Cos: 0.21100808680057526
Norm Quadratic Average: 9.58460521697998
Nearest Class Center Accuracy: 0.7578

Layer 12: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.42919740080833435
Inter Cos: 0.3818509578704834
Norm Quadratic Average: 6.011037349700928
Nearest Class Center Accuracy: 0.7566

Layer 13: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5836811661720276
Inter Cos: 0.40891891717910767
Norm Quadratic Average: 5.435441017150879
Nearest Class Center Accuracy: 0.7656

Layer 14: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6092895865440369
Inter Cos: 0.3694961667060852
Norm Quadratic Average: 5.0886054039001465
Nearest Class Center Accuracy: 0.7764

Layer 15: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5885356068611145
Inter Cos: 0.2822018265724182
Norm Quadratic Average: 4.3700714111328125
Nearest Class Center Accuracy: 0.797

Layer 16: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 73.43518829345703
Linear Weight Rank: 4031
Intra Cos: 0.5787652134895325
Inter Cos: 0.23250383138656616
Norm Quadratic Average: 24.551555633544922
Nearest Class Center Accuracy: 0.8092

Layer 17: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 29.96353530883789
Linear Weight Rank: 3670
Intra Cos: 0.5546795725822449
Inter Cos: 0.20582924783229828
Norm Quadratic Average: 20.54365348815918
Nearest Class Center Accuracy: 0.8135

Layer 18: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 4.024632930755615
Linear Weight Rank: 10
Intra Cos: 0.5463547110557556
Inter Cos: 0.22665467858314514
Norm Quadratic Average: 18.82482147216797
Nearest Class Center Accuracy: 0.8143

Output Layer:
Intra Cos: 0.5710019469261169
Inter Cos: 0.3963582217693329
Norm Quadratic Average: 22.883516311645508
Nearest Class Center Accuracy: 0.8265

