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_0.0001.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: 0.11166968196630478
Inter Cos: 0.12977290153503418
Norm Quadratic Average: 35.17087173461914
Nearest Class Center Accuracy: 0.8346

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18642547726631165
Inter Cos: 0.1653691828250885
Norm Quadratic Average: 30.385469436645508
Nearest Class Center Accuracy: 0.89175

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2168845534324646
Inter Cos: 0.1691678911447525
Norm Quadratic Average: 31.785663604736328
Nearest Class Center Accuracy: 0.9204333333333333

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2573126256465912
Inter Cos: 0.1544758528470993
Norm Quadratic Average: 16.349468231201172
Nearest Class Center Accuracy: 0.9670333333333333

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.374613493680954
Inter Cos: 0.1844424605369568
Norm Quadratic Average: 12.624363899230957
Nearest Class Center Accuracy: 0.9821166666666666

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5179076194763184
Inter Cos: 0.18238189816474915
Norm Quadratic Average: 6.783215045928955
Nearest Class Center Accuracy: 0.99535

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7757007479667664
Inter Cos: 0.2321232557296753
Norm Quadratic Average: 5.782423973083496
Nearest Class Center Accuracy: 0.9993166666666666

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.80147552490234
Linear Weight Rank: 4031
Intra Cos: 0.9105234742164612
Inter Cos: 0.18025502562522888
Norm Quadratic Average: 32.61086654663086
Nearest Class Center Accuracy: 0.9997833333333334

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.920215606689453
Linear Weight Rank: 3670
Intra Cos: 0.939597487449646
Inter Cos: 0.20278576016426086
Norm Quadratic Average: 27.727161407470703
Nearest Class Center Accuracy: 0.9998666666666667

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 3.002310276031494
Linear Weight Rank: 10
Intra Cos: 0.9437503218650818
Inter Cos: 0.19424809515476227
Norm Quadratic Average: 25.544443130493164
Nearest Class Center Accuracy: 0.9999166666666667

Output Layer:
Intra Cos: 0.9680720567703247
Inter Cos: 0.28882747888565063
Norm Quadratic Average: 25.475147247314453
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.021244205021432573
Accuracy: 0.9942
NC1 Within Class Collapse: 0.4770733714103699
NC2 Equinorm: Features: 0.09417259693145752, Weights: 0.01955725997686386
NC2 Equiangle: Features: 0.18358739217122397, Weights: 0.10345492892795138
NC3 Self-Duality: 0.24136637151241302
NC4 NCC Mismatch: 0.0032999999999999696

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: 0.12317639589309692
Inter Cos: 0.1288943588733673
Norm Quadratic Average: 35.06249237060547
Nearest Class Center Accuracy: 0.8473

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19980555772781372
Inter Cos: 0.16130846738815308
Norm Quadratic Average: 30.260269165039062
Nearest Class Center Accuracy: 0.9036

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22951537370681763
Inter Cos: 0.17580367624759674
Norm Quadratic Average: 31.701278686523438
Nearest Class Center Accuracy: 0.9286

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.26807719469070435
Inter Cos: 0.1669103056192398
Norm Quadratic Average: 16.334976196289062
Nearest Class Center Accuracy: 0.9703

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.38698792457580566
Inter Cos: 0.1992746889591217
Norm Quadratic Average: 12.64409351348877
Nearest Class Center Accuracy: 0.9819

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5247299671173096
Inter Cos: 0.19749407470226288
Norm Quadratic Average: 6.821849822998047
Nearest Class Center Accuracy: 0.9901

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.7766646146774292
Inter Cos: 0.24509954452514648
Norm Quadratic Average: 5.830577373504639
Nearest Class Center Accuracy: 0.9925

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 68.80147552490234
Linear Weight Rank: 4031
Intra Cos: 0.9059026837348938
Inter Cos: 0.19001945853233337
Norm Quadratic Average: 32.8983268737793
Nearest Class Center Accuracy: 0.993

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 27.920215606689453
Linear Weight Rank: 3670
Intra Cos: 0.9337587356567383
Inter Cos: 0.20063172280788422
Norm Quadratic Average: 27.966379165649414
Nearest Class Center Accuracy: 0.9924

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 3.002310276031494
Linear Weight Rank: 10
Intra Cos: 0.9372797608375549
Inter Cos: 0.19279277324676514
Norm Quadratic Average: 25.760705947875977
Nearest Class Center Accuracy: 0.9924

Output Layer:
Intra Cos: 0.9587347507476807
Inter Cos: 0.2752094268798828
Norm Quadratic Average: 25.67949676513672
Nearest Class Center Accuracy: 0.9933

