Model save path: /content/drive/MyDrive/Neural Collapse/New_Models/bn_True_dataset_MNIST_epochs_100_lr_0.001_model_type_vgg11_weight_decay_0.02.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.08946067094802856
Inter Cos: 0.11311887204647064
Norm Quadratic Average: 23.53293800354004
Nearest Class Center Accuracy: 0.81175

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09969369322061539
Inter Cos: 0.1224520206451416
Norm Quadratic Average: 36.58124542236328
Nearest Class Center Accuracy: 0.8365

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15141892433166504
Inter Cos: 0.13664405047893524
Norm Quadratic Average: 22.01261329650879
Nearest Class Center Accuracy: 0.8635

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.15548019111156464
Inter Cos: 0.12975719571113586
Norm Quadratic Average: 21.496461868286133
Nearest Class Center Accuracy: 0.883625

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20035681128501892
Inter Cos: 0.1225559264421463
Norm Quadratic Average: 13.332058906555176
Nearest Class Center Accuracy: 0.9315

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21716201305389404
Inter Cos: 0.10883870720863342
Norm Quadratic Average: 13.63369369506836
Nearest Class Center Accuracy: 0.964625

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.287824422121048
Inter Cos: 0.10945389419794083
Norm Quadratic Average: 9.356523513793945
Nearest Class Center Accuracy: 0.995375

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.5267437696456909
Inter Cos: 0.10837765038013458
Norm Quadratic Average: 7.353024482727051
Nearest Class Center Accuracy: 1.0

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 38.79948806762695
Linear Weight Rank: 4031
Intra Cos: 0.852316677570343
Inter Cos: 0.09298475086688995
Norm Quadratic Average: 69.42158508300781
Nearest Class Center Accuracy: 1.0

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 15.724119186401367
Linear Weight Rank: 3670
Intra Cos: 0.9408954977989197
Inter Cos: 0.12750323116779327
Norm Quadratic Average: 32.107521057128906
Nearest Class Center Accuracy: 1.0

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.5449968576431274
Linear Weight Rank: 10
Intra Cos: 0.9580090045928955
Inter Cos: 0.21125856041908264
Norm Quadratic Average: 17.94001579284668
Nearest Class Center Accuracy: 1.0

Output Layer:
Intra Cos: 0.9678802490234375
Inter Cos: 0.3156610429286957
Norm Quadratic Average: 9.852869987487793
Nearest Class Center Accuracy: 1.0

Test Set:
Average Loss: 0.06528513151407242
Accuracy: 0.983
NC1 Within Class Collapse: 0.9715830087661743
NC2 Equinorm: Features: 0.07317441701889038, Weights: 0.01939224824309349
NC2 Equiangle: Features: 0.19887080722384984, Weights: 0.0997817145453559
NC3 Self-Duality: 0.17139089107513428
NC4 NCC Mismatch: 0.0040000000000000036

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10133806616067886
Inter Cos: 0.11957792937755585
Norm Quadratic Average: 22.82430076599121
Nearest Class Center Accuracy: 0.809

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.1254557967185974
Inter Cos: 0.12788242101669312
Norm Quadratic Average: 36.086700439453125
Nearest Class Center Accuracy: 0.8305

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16157151758670807
Inter Cos: 0.15031573176383972
Norm Quadratic Average: 21.8208065032959
Nearest Class Center Accuracy: 0.8545

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16426901519298553
Inter Cos: 0.13293397426605225
Norm Quadratic Average: 21.313631057739258
Nearest Class Center Accuracy: 0.876

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.20376397669315338
Inter Cos: 0.12065014988183975
Norm Quadratic Average: 13.307933807373047
Nearest Class Center Accuracy: 0.926

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.22146116197109222
Inter Cos: 0.11089421063661575
Norm Quadratic Average: 13.629676818847656
Nearest Class Center Accuracy: 0.9475

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2736179828643799
Inter Cos: 0.11174345761537552
Norm Quadratic Average: 9.347108840942383
Nearest Class Center Accuracy: 0.9735

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4705735146999359
Inter Cos: 0.11832761764526367
Norm Quadratic Average: 7.251337051391602
Nearest Class Center Accuracy: 0.982

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 38.79948806762695
Linear Weight Rank: 4031
Intra Cos: 0.7459456324577332
Inter Cos: 0.11726083606481552
Norm Quadratic Average: 66.96288299560547
Nearest Class Center Accuracy: 0.9825

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 15.724119186401367
Linear Weight Rank: 3670
Intra Cos: 0.839162290096283
Inter Cos: 0.1385737508535385
Norm Quadratic Average: 30.870573043823242
Nearest Class Center Accuracy: 0.982

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.5449968576431274
Linear Weight Rank: 10
Intra Cos: 0.8568532466888428
Inter Cos: 0.2157730609178543
Norm Quadratic Average: 17.271806716918945
Nearest Class Center Accuracy: 0.9835

Output Layer:
Intra Cos: 0.8699632883071899
Inter Cos: 0.31198379397392273
Norm Quadratic Average: 9.476142883300781
Nearest Class Center Accuracy: 0.983

