Remove stale kfold/ (pre-v2 weights, will be regenerated)

kfold/README.md

@@ -1,18 +0,0 @@
-# 5-Fold Cross-Validation Results
-
-These results come from an earlier sanity-check run (30 epochs/fold, **Original Dataset only**,
-no group-aware splitting, no v2 protocol). They are kept for completeness and serve as the
-v1 baseline. The headline results in the model card use the v2 protocol on the augmented
-dataset with pHash-grouped 5-fold splits.
-
-## Files
-- `kfold_results.json` — full per-fold metrics
-- `kfold_fold_metrics.csv` — flat per-fold table
-- `kfold_summary.csv` — mean ± std across folds
-- `kfold_thesis_text.md` — pre-formatted thesis paragraph
-- `run.log` — training log
-
-## v1 K-fold accuracy (Original Dataset, 30 epochs/fold, mean ± std)
-- VGG19              : 79.31 ± 1.89%
-- ResNet101          : 79.27 ± 1.07%
-- CLIP Transformer   : 63.92 ± 1.79%

kfold/kfold_fold_metrics.csv

@@ -1,16 +0,0 @@
-model,fold,best_epoch,train_size,validation_size,train_loss,loss,accuracy,precision_macro,recall_macro,f1_macro,elapsed_minutes,checkpoint
-vgg19,1,28,4268,1067,0.49587756247305936,0.5705084835040368,0.8050609184629803,0.8281820928991002,0.7959482444359403,0.810311103586162,26.761109602451324,
-vgg19,2,22,4268,1067,0.5561496519886006,0.6453155194528771,0.7731958762886598,0.7772852115079077,0.7690913231781914,0.7718414723120325,26.143115723133086,
-vgg19,3,14,4268,1067,0.7227823485418693,0.6405022616071352,0.7722586691658857,0.7788600686877383,0.7612916487206677,0.7662782481081638,19.190628111362457,
-vgg19,4,27,4268,1067,0.5004883545408096,0.6042639139861623,0.8022492970946579,0.7894039797418415,0.8065218052758446,0.7959661454741711,26.171603671709697,
-vgg19,5,29,4268,1067,0.48269938182473293,0.5745639920918988,0.8125585754451734,0.8369509065813088,0.804951903720035,0.8184411860665891,26.15475913286209,
-resnet101,1,9,4268,1067,0.5743873137788674,0.5456624888826332,0.7966260543580131,0.8137835750038453,0.812096496356399,0.8116951351387464,13.780445738633473,
-resnet101,2,15,4268,1067,0.46592906167714976,0.6816925578231329,0.7788191190253045,0.7870394272232406,0.7938511208231884,0.7854041450944791,18.33250972032547,
-resnet101,3,12,4268,1067,0.5078951980649811,0.6246329910473614,0.7891283973758201,0.8231509013967143,0.7847325361243083,0.8002388933430273,15.842269384860993,
-resnet101,4,11,4268,1067,0.5430463672578055,0.5688460089943626,0.7910028116213683,0.7833999797476783,0.8079012850934125,0.7928454500340572,15.19050563176473,
-resnet101,5,9,4268,1067,0.5719610963788341,0.5545566146207,0.8078725398313027,0.8350563418654229,0.7937864083217823,0.80982314301139,13.609436937173207,
-transformer,1,25,4268,1067,0.994727843480347,1.0549040075876832,0.6410496719775071,0.5963772969988077,0.5496712723042669,0.5527651120388246,28.380551489194236,
-transformer,2,28,4268,1067,0.9593697755048514,1.0812005294520048,0.6223055295220243,0.5411757987956644,0.5295209396126774,0.5238912478907942,28.372538256645203,
-transformer,3,26,4268,1067,1.057130372066194,1.1125167294317244,0.626991565135895,0.544307583735951,0.4968785226960196,0.5047793108912983,28.2229678829511,
-transformer,4,19,4268,1067,1.0797306158437612,1.01943102213637,0.6682286785379569,0.6215538138446444,0.5467448399905391,0.5735351310730014,25.49445507923762,
-transformer,5,26,4268,1067,0.9861715776739326,1.0478629528526513,0.6373008434864105,0.670058379661995,0.5608641108767098,0.5734993325379196,28.32114017009735,

kfold/kfold_results.json

@@ -1,291 +0,0 @@
-{
-  "folds": [
-    {
-      "loss": 0.5705084835040368,
-      "accuracy": 0.8050609184629803,
-      "precision_macro": 0.8281820928991002,
-      "recall_macro": 0.7959482444359403,
-      "f1_macro": 0.810311103586162,
-      "train_loss": 0.49587756247305936,
-      "epoch": 28,
-      "model": "vgg19",
-      "fold": 1,
-      "best_epoch": 28,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 26.761109602451324
-    },
-    {
-      "loss": 0.6453155194528771,
-      "accuracy": 0.7731958762886598,
-      "precision_macro": 0.7772852115079077,
-      "recall_macro": 0.7690913231781914,
-      "f1_macro": 0.7718414723120325,
-      "train_loss": 0.5561496519886006,
-      "epoch": 22,
-      "model": "vgg19",
-      "fold": 2,
-      "best_epoch": 22,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 26.143115723133086
-    },
-    {
-      "loss": 0.6405022616071352,
-      "accuracy": 0.7722586691658857,
-      "precision_macro": 0.7788600686877383,
-      "recall_macro": 0.7612916487206677,
-      "f1_macro": 0.7662782481081638,
-      "train_loss": 0.7227823485418693,
-      "epoch": 14,
-      "model": "vgg19",
-      "fold": 3,
-      "best_epoch": 14,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 19.190628111362457
-    },
-    {
-      "loss": 0.6042639139861623,
-      "accuracy": 0.8022492970946579,
-      "precision_macro": 0.7894039797418415,
-      "recall_macro": 0.8065218052758446,
-      "f1_macro": 0.7959661454741711,
-      "train_loss": 0.5004883545408096,
-      "epoch": 27,
-      "model": "vgg19",
-      "fold": 4,
-      "best_epoch": 27,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 26.171603671709697
-    },
-    {
-      "loss": 0.5745639920918988,
-      "accuracy": 0.8125585754451734,
-      "precision_macro": 0.8369509065813088,
-      "recall_macro": 0.804951903720035,
-      "f1_macro": 0.8184411860665891,
-      "train_loss": 0.48269938182473293,
-      "epoch": 29,
-      "model": "vgg19",
-      "fold": 5,
-      "best_epoch": 29,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 26.15475913286209
-    },
-    {
-      "loss": 0.5456624888826332,
-      "accuracy": 0.7966260543580131,
-      "precision_macro": 0.8137835750038453,
-      "recall_macro": 0.812096496356399,
-      "f1_macro": 0.8116951351387464,
-      "train_loss": 0.5743873137788674,
-      "epoch": 9,
-      "model": "resnet101",
-      "fold": 1,
-      "best_epoch": 9,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 13.780445738633473
-    },
-    {
-      "loss": 0.6816925578231329,
-      "accuracy": 0.7788191190253045,
-      "precision_macro": 0.7870394272232406,
-      "recall_macro": 0.7938511208231884,
-      "f1_macro": 0.7854041450944791,
-      "train_loss": 0.46592906167714976,
-      "epoch": 15,
-      "model": "resnet101",
-      "fold": 2,
-      "best_epoch": 15,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 18.33250972032547
-    },
-    {
-      "loss": 0.6246329910473614,
-      "accuracy": 0.7891283973758201,
-      "precision_macro": 0.8231509013967143,
-      "recall_macro": 0.7847325361243083,
-      "f1_macro": 0.8002388933430273,
-      "train_loss": 0.5078951980649811,
-      "epoch": 12,
-      "model": "resnet101",
-      "fold": 3,
-      "best_epoch": 12,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 15.842269384860993
-    },
-    {
-      "loss": 0.5688460089943626,
-      "accuracy": 0.7910028116213683,
-      "precision_macro": 0.7833999797476783,
-      "recall_macro": 0.8079012850934125,
-      "f1_macro": 0.7928454500340572,
-      "train_loss": 0.5430463672578055,
-      "epoch": 11,
-      "model": "resnet101",
-      "fold": 4,
-      "best_epoch": 11,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 15.19050563176473
-    },
-    {
-      "loss": 0.5545566146207,
-      "accuracy": 0.8078725398313027,
-      "precision_macro": 0.8350563418654229,
-      "recall_macro": 0.7937864083217823,
-      "f1_macro": 0.80982314301139,
-      "train_loss": 0.5719610963788341,
-      "epoch": 9,
-      "model": "resnet101",
-      "fold": 5,
-      "best_epoch": 9,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 13.609436937173207
-    },
-    {
-      "loss": 1.0549040075876832,
-      "accuracy": 0.6410496719775071,
-      "precision_macro": 0.5963772969988077,
-      "recall_macro": 0.5496712723042669,
-      "f1_macro": 0.5527651120388246,
-      "train_loss": 0.994727843480347,
-      "epoch": 25,
-      "model": "transformer",
-      "fold": 1,
-      "best_epoch": 25,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 28.380551489194236
-    },
-    {
-      "loss": 1.0812005294520048,
-      "accuracy": 0.6223055295220243,
-      "precision_macro": 0.5411757987956644,
-      "recall_macro": 0.5295209396126774,
-      "f1_macro": 0.5238912478907942,
-      "train_loss": 0.9593697755048514,
-      "epoch": 28,
-      "model": "transformer",
-      "fold": 2,
-      "best_epoch": 28,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 28.372538256645203
-    },
-    {
-      "loss": 1.1125167294317244,
-      "accuracy": 0.626991565135895,
-      "precision_macro": 0.544307583735951,
-      "recall_macro": 0.4968785226960196,
-      "f1_macro": 0.5047793108912983,
-      "train_loss": 1.057130372066194,
-      "epoch": 26,
-      "model": "transformer",
-      "fold": 3,
-      "best_epoch": 26,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 28.2229678829511
-    },
-    {
-      "loss": 1.01943102213637,
-      "accuracy": 0.6682286785379569,
-      "precision_macro": 0.6215538138446444,
-      "recall_macro": 0.5467448399905391,
-      "f1_macro": 0.5735351310730014,
-      "train_loss": 1.0797306158437612,
-      "epoch": 19,
-      "model": "transformer",
-      "fold": 4,
-      "best_epoch": 19,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 25.49445507923762
-    },
-    {
-      "loss": 1.0478629528526513,
-      "accuracy": 0.6373008434864105,
-      "precision_macro": 0.670058379661995,
-      "recall_macro": 0.5608641108767098,
-      "f1_macro": 0.5734993325379196,
-      "train_loss": 0.9861715776739326,
-      "epoch": 26,
-      "model": "transformer",
-      "fold": 5,
-      "best_epoch": 26,
-      "train_size": 4268,
-      "validation_size": 1067,
-      "elapsed_minutes": 28.32114017009735
-    }
-  ],
-  "summary": [
-    {
-      "model": "resnet101",
-      "folds": 5,
-      "accuracy_mean": 0.7926897844423617,
-      "accuracy_std": 0.010652875155301094,
-      "accuracy_ci95_low": 0.7794646134150562,
-      "accuracy_ci95_high": 0.8059149554696672,
-      "precision_macro_mean": 0.8084860450473803,
-      "precision_macro_std": 0.022574158880790833,
-      "precision_macro_ci95_low": 0.7804610182218793,
-      "precision_macro_ci95_high": 0.8365110718728813,
-      "recall_macro_mean": 0.7984735693438181,
-      "recall_macro_std": 0.011254093337062896,
-      "recall_macro_ci95_low": 0.7845020070218987,
-      "recall_macro_ci95_high": 0.8124451316657375,
-      "f1_macro_mean": 0.8000013533243401,
-      "f1_macro_std": 0.011152958699890591,
-      "f1_macro_ci95_low": 0.7861553461088018,
-      "f1_macro_ci95_high": 0.8138473605398784
-    },
-    {
-      "model": "transformer",
-      "folds": 5,
-      "accuracy_mean": 0.6391752577319588,
-      "accuracy_std": 0.017917576658205397,
-      "accuracy_ci95_low": 0.6169312144811839,
-      "accuracy_ci95_high": 0.6614193009827337,
-      "precision_macro_mean": 0.5946945746074125,
-      "precision_macro_std": 0.05433015187909789,
-      "precision_macro_ci95_low": 0.5272455958044459,
-      "precision_macro_ci95_high": 0.6621435534103791,
-      "recall_macro_mean": 0.5367359370960425,
-      "recall_macro_std": 0.024951650623910596,
-      "recall_macro_ci95_low": 0.5057593376236849,
-      "recall_macro_ci95_high": 0.5677125365684002,
-      "f1_macro_mean": 0.5456940268863676,
-      "f1_macro_std": 0.03060897542726445,
-      "f1_macro_ci95_low": 0.5076940570101861,
-      "f1_macro_ci95_high": 0.5836939967625491
-    },
-    {
-      "model": "vgg19",
-      "folds": 5,
-      "accuracy_mean": 0.7930646672914714,
-      "accuracy_std": 0.018946888503206548,
-      "accuracy_ci95_low": 0.7695427694716604,
-      "accuracy_ci95_high": 0.8165865651112824,
-      "precision_macro_mean": 0.8021364518835792,
-      "precision_macro_std": 0.028336865041579704,
-      "precision_macro_ci95_low": 0.7669572273935524,
-      "precision_macro_ci95_high": 0.8373156763736059,
-      "recall_macro_mean": 0.7875609850661359,
-      "recall_macro_std": 0.020997079062777587,
-      "recall_macro_ci95_low": 0.7614938475440272,
-      "recall_macro_ci95_high": 0.8136281225882447,
-      "f1_macro_mean": 0.7925676311094236,
-      "f1_macro_std": 0.023002863292551388,
-      "f1_macro_ci95_low": 0.7640103827866236,
-      "f1_macro_ci95_high": 0.8211248794322237
-    }
-  ]
-}

kfold/kfold_summary.csv

@@ -1,4 +0,0 @@
-model,folds,accuracy_mean,accuracy_std,accuracy_ci95_low,accuracy_ci95_high,precision_macro_mean,precision_macro_std,precision_macro_ci95_low,precision_macro_ci95_high,recall_macro_mean,recall_macro_std,recall_macro_ci95_low,recall_macro_ci95_high,f1_macro_mean,f1_macro_std,f1_macro_ci95_low,f1_macro_ci95_high
-resnet101,5,0.7926897844423617,0.010652875155301094,0.7794646134150562,0.8059149554696672,0.8084860450473803,0.022574158880790833,0.7804610182218793,0.8365110718728813,0.7984735693438181,0.011254093337062896,0.7845020070218987,0.8124451316657375,0.8000013533243401,0.011152958699890591,0.7861553461088018,0.8138473605398784
-transformer,5,0.6391752577319588,0.017917576658205397,0.6169312144811839,0.6614193009827337,0.5946945746074125,0.05433015187909789,0.5272455958044459,0.6621435534103791,0.5367359370960425,0.024951650623910596,0.5057593376236849,0.5677125365684002,0.5456940268863676,0.03060897542726445,0.5076940570101861,0.5836939967625491
-vgg19,5,0.7930646672914714,0.018946888503206548,0.7695427694716604,0.8165865651112824,0.8021364518835792,0.028336865041579704,0.7669572273935524,0.8373156763736059,0.7875609850661359,0.020997079062777587,0.7614938475440272,0.8136281225882447,0.7925676311094236,0.023002863292551388,0.7640103827866236,0.8211248794322237

kfold/kfold_thesis_text.md

@@ -1,11 +0,0 @@
-# Thesis-Ready K-Fold Validation Text
-
-A stratified 5-fold cross-validation experiment was conducted on the fundus image dataset. The folds preserved disease-class proportions, and the same preprocessing and augmentation strategy was used across all models. The experiment used a fine-tuned training setting with a maximum of 30 epochs per fold and early stopping. The original single-split validation accuracies are retained as the primary hold-out results, while the K-fold values are reported as an additional robustness check.
-
-| Model | Original hold-out accuracy | K-fold accuracy Mean ± SD | 95% CI | Macro Precision | Macro Recall | Macro F1-score |
-|---|---:|---:|---:|---:|---:|---:|
-| resnet101 | 90.80% | 79.27% ± 1.07% | 77.95%-80.59% | 80.85% | 79.85% | 80.00% |
-| transformer | 91.50% | 63.92% ± 1.79% | 61.69%-66.14% | 59.47% | 53.67% | 54.57% |
-| vgg19 | 88.00% | 79.31% ± 1.89% | 76.95%-81.66% | 80.21% | 78.76% | 79.26% |
-
-These fold-level results provide a more reliable estimate of model generalization than a single train-validation split, because every image is used for validation exactly once while maintaining class balance across folds.

kfold/run.log

@@ -1,411 +0,0 @@
-Dataset: Database/Original_Dataset
-Images: 5335
-Classes: 10
-{
-  "Central Serous Chorioretinopathy [Color Fundus]": 101,
-  "Diabetic Retinopathy": 1509,
-  "Disc Edema": 127,
-  "Glaucoma": 1349,
-  "Healthy": 1024,
-  "Macular Scar": 444,
-  "Myopia": 500,
-  "Pterygium": 17,
-  "Retinal Detachment": 125,
-  "Retinitis Pigmentosa": 139
-}
-Device: cuda
-GPU: Tesla T4
-vgg19 fold 1 epoch 001: train_loss=1.3849 val_acc=0.6542 val_f1=0.3077
-vgg19 fold 1 epoch 002: train_loss=1.1426 val_acc=0.7076 val_f1=0.5494
-vgg19 fold 1 epoch 003: train_loss=1.0747 val_acc=0.6776 val_f1=0.5187
-vgg19 fold 1 epoch 004: train_loss=1.0201 val_acc=0.7104 val_f1=0.5874
-vgg19 fold 1 epoch 005: train_loss=0.9692 val_acc=0.6673 val_f1=0.5931
-vgg19 fold 1 epoch 006: train_loss=0.9579 val_acc=0.7254 val_f1=0.6254
-vgg19 fold 1 epoch 007: train_loss=0.9145 val_acc=0.7188 val_f1=0.6185
-vgg19 fold 1 epoch 008: train_loss=0.8651 val_acc=0.6898 val_f1=0.5762
-vgg19 fold 1 epoch 009: train_loss=0.8400 val_acc=0.7470 val_f1=0.6513
-vgg19 fold 1 epoch 010: train_loss=0.8301 val_acc=0.7723 val_f1=0.6934
-vgg19 fold 1 epoch 011: train_loss=0.8011 val_acc=0.7441 val_f1=0.7135
-vgg19 fold 1 epoch 012: train_loss=0.7709 val_acc=0.7563 val_f1=0.6833
-vgg19 fold 1 epoch 013: train_loss=0.7475 val_acc=0.7385 val_f1=0.6803
-vgg19 fold 1 epoch 014: train_loss=0.7224 val_acc=0.7751 val_f1=0.7321
-vgg19 fold 1 epoch 015: train_loss=0.6872 val_acc=0.7573 val_f1=0.7295
-vgg19 fold 1 epoch 016: train_loss=0.6682 val_acc=0.7713 val_f1=0.7524
-vgg19 fold 1 epoch 017: train_loss=0.6700 val_acc=0.7854 val_f1=0.7356
-vgg19 fold 1 epoch 018: train_loss=0.6442 val_acc=0.7760 val_f1=0.7361
-vgg19 fold 1 epoch 019: train_loss=0.6147 val_acc=0.7816 val_f1=0.7628
-vgg19 fold 1 epoch 020: train_loss=0.6146 val_acc=0.7788 val_f1=0.7664
-vgg19 fold 1 epoch 021: train_loss=0.5922 val_acc=0.7929 val_f1=0.7838
-vgg19 fold 1 epoch 022: train_loss=0.5630 val_acc=0.7816 val_f1=0.7516
-vgg19 fold 1 epoch 023: train_loss=0.5747 val_acc=0.8013 val_f1=0.7953
-vgg19 fold 1 epoch 024: train_loss=0.5389 val_acc=0.7891 val_f1=0.7702
-vgg19 fold 1 epoch 025: train_loss=0.5217 val_acc=0.7919 val_f1=0.7801
-vgg19 fold 1 epoch 026: train_loss=0.5245 val_acc=0.7994 val_f1=0.7963
-vgg19 fold 1 epoch 027: train_loss=0.5187 val_acc=0.7957 val_f1=0.8002
-vgg19 fold 1 epoch 028: train_loss=0.4959 val_acc=0.8051 val_f1=0.8103
-vgg19 fold 1 epoch 029: train_loss=0.4973 val_acc=0.8041 val_f1=0.8042
-vgg19 fold 1 epoch 030: train_loss=0.4874 val_acc=0.8051 val_f1=0.8015
-vgg19 fold 2 epoch 001: train_loss=1.4041 val_acc=0.6157 val_f1=0.2996
-vgg19 fold 2 epoch 002: train_loss=1.1676 val_acc=0.6317 val_f1=0.4760
-vgg19 fold 2 epoch 003: train_loss=1.0477 val_acc=0.6842 val_f1=0.5870
-vgg19 fold 2 epoch 004: train_loss=1.0113 val_acc=0.7010 val_f1=0.6140
-vgg19 fold 2 epoch 005: train_loss=0.9329 val_acc=0.7095 val_f1=0.6137
-vgg19 fold 2 epoch 006: train_loss=0.8749 val_acc=0.7301 val_f1=0.6572
-vgg19 fold 2 epoch 007: train_loss=0.8980 val_acc=0.7142 val_f1=0.6902
-vgg19 fold 2 epoch 008: train_loss=0.8330 val_acc=0.7085 val_f1=0.6101
-vgg19 fold 2 epoch 009: train_loss=0.8039 val_acc=0.7376 val_f1=0.7133
-vgg19 fold 2 epoch 010: train_loss=0.7867 val_acc=0.7291 val_f1=0.6969
-vgg19 fold 2 epoch 011: train_loss=0.7708 val_acc=0.7573 val_f1=0.7284
-vgg19 fold 2 epoch 012: train_loss=0.7417 val_acc=0.7460 val_f1=0.6828
-vgg19 fold 2 epoch 013: train_loss=0.7103 val_acc=0.7554 val_f1=0.7408
-vgg19 fold 2 epoch 014: train_loss=0.7114 val_acc=0.7516 val_f1=0.7398
-vgg19 fold 2 epoch 015: train_loss=0.6656 val_acc=0.7535 val_f1=0.7413
-vgg19 fold 2 epoch 016: train_loss=0.6353 val_acc=0.7554 val_f1=0.7292
-vgg19 fold 2 epoch 017: train_loss=0.6374 val_acc=0.7563 val_f1=0.7215
-vgg19 fold 2 epoch 018: train_loss=0.6023 val_acc=0.7573 val_f1=0.7508
-vgg19 fold 2 epoch 019: train_loss=0.6103 val_acc=0.7657 val_f1=0.7621
-vgg19 fold 2 epoch 020: train_loss=0.5736 val_acc=0.7666 val_f1=0.7564
-vgg19 fold 2 epoch 021: train_loss=0.5614 val_acc=0.7676 val_f1=0.7433
-vgg19 fold 2 epoch 022: train_loss=0.5561 val_acc=0.7732 val_f1=0.7718
-vgg19 fold 2 epoch 023: train_loss=0.5438 val_acc=0.7723 val_f1=0.7819
-vgg19 fold 2 epoch 024: train_loss=0.5217 val_acc=0.7648 val_f1=0.7699
-vgg19 fold 2 epoch 025: train_loss=0.4876 val_acc=0.7676 val_f1=0.7571
-vgg19 fold 2 epoch 026: train_loss=0.5075 val_acc=0.7732 val_f1=0.7734
-vgg19 fold 2 epoch 027: train_loss=0.5069 val_acc=0.7648 val_f1=0.7611
-vgg19 fold 2 epoch 028: train_loss=0.4827 val_acc=0.7713 val_f1=0.7705
-vgg19 fold 2 epoch 029: train_loss=0.4812 val_acc=0.7732 val_f1=0.7706
-vgg19 fold 2 epoch 030: train_loss=0.4743 val_acc=0.7732 val_f1=0.7705
-Early stopping vgg19 fold 2 at epoch 30.
-vgg19 fold 3 epoch 001: train_loss=1.4027 val_acc=0.6307 val_f1=0.3678
-vgg19 fold 3 epoch 002: train_loss=1.2012 val_acc=0.6129 val_f1=0.3808
-vgg19 fold 3 epoch 003: train_loss=1.0840 val_acc=0.6898 val_f1=0.5008
-vgg19 fold 3 epoch 004: train_loss=1.0103 val_acc=0.7001 val_f1=0.5863
-vgg19 fold 3 epoch 005: train_loss=0.9576 val_acc=0.7132 val_f1=0.6255
-vgg19 fold 3 epoch 006: train_loss=0.9050 val_acc=0.7132 val_f1=0.6533
-vgg19 fold 3 epoch 007: train_loss=0.8791 val_acc=0.7338 val_f1=0.6565
-vgg19 fold 3 epoch 008: train_loss=0.8814 val_acc=0.7413 val_f1=0.6817
-vgg19 fold 3 epoch 009: train_loss=0.8465 val_acc=0.6795 val_f1=0.6214
-vgg19 fold 3 epoch 010: train_loss=0.7899 val_acc=0.7582 val_f1=0.7142
-vgg19 fold 3 epoch 011: train_loss=0.7432 val_acc=0.7488 val_f1=0.7072
-vgg19 fold 3 epoch 012: train_loss=0.7024 val_acc=0.7676 val_f1=0.7112
-vgg19 fold 3 epoch 013: train_loss=0.7174 val_acc=0.7488 val_f1=0.7188
-vgg19 fold 3 epoch 014: train_loss=0.7228 val_acc=0.7723 val_f1=0.7663
-vgg19 fold 3 epoch 015: train_loss=0.6752 val_acc=0.7666 val_f1=0.7429
-vgg19 fold 3 epoch 016: train_loss=0.6575 val_acc=0.7685 val_f1=0.7279
-vgg19 fold 3 epoch 017: train_loss=0.6322 val_acc=0.7301 val_f1=0.7180
-vgg19 fold 3 epoch 018: train_loss=0.6002 val_acc=0.7648 val_f1=0.7404
-vgg19 fold 3 epoch 019: train_loss=0.5943 val_acc=0.7601 val_f1=0.7348
-vgg19 fold 3 epoch 020: train_loss=0.5845 val_acc=0.7601 val_f1=0.7482
-vgg19 fold 3 epoch 021: train_loss=0.5573 val_acc=0.7694 val_f1=0.7559
-vgg19 fold 3 epoch 022: train_loss=0.5544 val_acc=0.7713 val_f1=0.7727
-Early stopping vgg19 fold 3 at epoch 22.
-vgg19 fold 4 epoch 001: train_loss=1.3574 val_acc=0.6504 val_f1=0.3608
-vgg19 fold 4 epoch 002: train_loss=1.1289 val_acc=0.6242 val_f1=0.4749
-vgg19 fold 4 epoch 003: train_loss=1.0650 val_acc=0.7226 val_f1=0.5900
-vgg19 fold 4 epoch 004: train_loss=0.9947 val_acc=0.7291 val_f1=0.6058
-vgg19 fold 4 epoch 005: train_loss=0.9427 val_acc=0.6935 val_f1=0.5481
-vgg19 fold 4 epoch 006: train_loss=0.9316 val_acc=0.7385 val_f1=0.6555
-vgg19 fold 4 epoch 007: train_loss=0.9099 val_acc=0.7545 val_f1=0.6545
-vgg19 fold 4 epoch 008: train_loss=0.8717 val_acc=0.7310 val_f1=0.6731
-vgg19 fold 4 epoch 009: train_loss=0.8427 val_acc=0.7704 val_f1=0.7127
-vgg19 fold 4 epoch 010: train_loss=0.8234 val_acc=0.7760 val_f1=0.7351
-vgg19 fold 4 epoch 011: train_loss=0.7697 val_acc=0.7807 val_f1=0.7430
-vgg19 fold 4 epoch 012: train_loss=0.7628 val_acc=0.7788 val_f1=0.7324
-vgg19 fold 4 epoch 013: train_loss=0.7542 val_acc=0.7741 val_f1=0.6906
-vgg19 fold 4 epoch 014: train_loss=0.7118 val_acc=0.7694 val_f1=0.7365
-vgg19 fold 4 epoch 015: train_loss=0.7078 val_acc=0.7582 val_f1=0.7396
-vgg19 fold 4 epoch 016: train_loss=0.6875 val_acc=0.7760 val_f1=0.7333
-vgg19 fold 4 epoch 017: train_loss=0.6633 val_acc=0.7741 val_f1=0.7495
-vgg19 fold 4 epoch 018: train_loss=0.6338 val_acc=0.7760 val_f1=0.7400
-vgg19 fold 4 epoch 019: train_loss=0.6250 val_acc=0.7835 val_f1=0.7604
-vgg19 fold 4 epoch 020: train_loss=0.5929 val_acc=0.7919 val_f1=0.7736
-vgg19 fold 4 epoch 021: train_loss=0.5818 val_acc=0.7835 val_f1=0.7702
-vgg19 fold 4 epoch 022: train_loss=0.5619 val_acc=0.7948 val_f1=0.7773
-vgg19 fold 4 epoch 023: train_loss=0.5426 val_acc=0.7985 val_f1=0.7888
-vgg19 fold 4 epoch 024: train_loss=0.5374 val_acc=0.7948 val_f1=0.7963
-vgg19 fold 4 epoch 025: train_loss=0.5284 val_acc=0.8013 val_f1=0.7923
-vgg19 fold 4 epoch 026: train_loss=0.5080 val_acc=0.7929 val_f1=0.7913
-vgg19 fold 4 epoch 027: train_loss=0.5005 val_acc=0.8022 val_f1=0.7960
-vgg19 fold 4 epoch 028: train_loss=0.5095 val_acc=0.7966 val_f1=0.7910
-vgg19 fold 4 epoch 029: train_loss=0.5004 val_acc=0.7976 val_f1=0.7918
-vgg19 fold 4 epoch 030: train_loss=0.4940 val_acc=0.7994 val_f1=0.7933
-vgg19 fold 5 epoch 001: train_loss=1.3859 val_acc=0.6120 val_f1=0.2877
-vgg19 fold 5 epoch 002: train_loss=1.1860 val_acc=0.6261 val_f1=0.3973
-vgg19 fold 5 epoch 003: train_loss=1.0995 val_acc=0.7132 val_f1=0.6240
-vgg19 fold 5 epoch 004: train_loss=0.9793 val_acc=0.7226 val_f1=0.6173
-vgg19 fold 5 epoch 005: train_loss=0.9266 val_acc=0.7001 val_f1=0.6597
-vgg19 fold 5 epoch 006: train_loss=0.9041 val_acc=0.7198 val_f1=0.6371
-vgg19 fold 5 epoch 007: train_loss=0.8692 val_acc=0.7329 val_f1=0.6866
-vgg19 fold 5 epoch 008: train_loss=0.8728 val_acc=0.7666 val_f1=0.7235
-vgg19 fold 5 epoch 009: train_loss=0.8126 val_acc=0.7619 val_f1=0.7061
-vgg19 fold 5 epoch 010: train_loss=0.8016 val_acc=0.7470 val_f1=0.6821
-vgg19 fold 5 epoch 011: train_loss=0.7634 val_acc=0.7648 val_f1=0.7183
-vgg19 fold 5 epoch 012: train_loss=0.7601 val_acc=0.7751 val_f1=0.7667
-vgg19 fold 5 epoch 013: train_loss=0.7400 val_acc=0.7798 val_f1=0.7754
-vgg19 fold 5 epoch 014: train_loss=0.7030 val_acc=0.7751 val_f1=0.7463
-vgg19 fold 5 epoch 015: train_loss=0.6620 val_acc=0.7723 val_f1=0.7437
-vgg19 fold 5 epoch 016: train_loss=0.6675 val_acc=0.7985 val_f1=0.7804
-vgg19 fold 5 epoch 017: train_loss=0.6546 val_acc=0.7919 val_f1=0.7761
-vgg19 fold 5 epoch 018: train_loss=0.6199 val_acc=0.7948 val_f1=0.7833
-vgg19 fold 5 epoch 019: train_loss=0.5927 val_acc=0.7873 val_f1=0.7481
-vgg19 fold 5 epoch 020: train_loss=0.6023 val_acc=0.7882 val_f1=0.7889
-vgg19 fold 5 epoch 021: train_loss=0.5643 val_acc=0.7994 val_f1=0.8000
-vgg19 fold 5 epoch 022: train_loss=0.5623 val_acc=0.8022 val_f1=0.7846
-vgg19 fold 5 epoch 023: train_loss=0.5477 val_acc=0.7985 val_f1=0.7994
-vgg19 fold 5 epoch 024: train_loss=0.5213 val_acc=0.8069 val_f1=0.8163
-vgg19 fold 5 epoch 025: train_loss=0.5247 val_acc=0.8088 val_f1=0.8176
-vgg19 fold 5 epoch 026: train_loss=0.5067 val_acc=0.8088 val_f1=0.8133
-vgg19 fold 5 epoch 027: train_loss=0.5104 val_acc=0.8107 val_f1=0.8164
-vgg19 fold 5 epoch 028: train_loss=0.4965 val_acc=0.8041 val_f1=0.8170
-vgg19 fold 5 epoch 029: train_loss=0.4827 val_acc=0.8126 val_f1=0.8184
-vgg19 fold 5 epoch 030: train_loss=0.4868 val_acc=0.8088 val_f1=0.8112
-resnet101 fold 1 epoch 001: train_loss=1.2464 val_acc=0.7160 val_f1=0.6060
-resnet101 fold 1 epoch 002: train_loss=0.8628 val_acc=0.7545 val_f1=0.7583
-resnet101 fold 1 epoch 003: train_loss=0.7692 val_acc=0.7573 val_f1=0.7534
-resnet101 fold 1 epoch 004: train_loss=0.7473 val_acc=0.7629 val_f1=0.7403
-resnet101 fold 1 epoch 005: train_loss=0.7024 val_acc=0.7713 val_f1=0.7870
-resnet101 fold 1 epoch 006: train_loss=0.6554 val_acc=0.7638 val_f1=0.7774
-resnet101 fold 1 epoch 007: train_loss=0.6192 val_acc=0.7816 val_f1=0.7820
-resnet101 fold 1 epoch 008: train_loss=0.5997 val_acc=0.7610 val_f1=0.7795
-resnet101 fold 1 epoch 009: train_loss=0.5744 val_acc=0.7966 val_f1=0.8117
-resnet101 fold 1 epoch 010: train_loss=0.5464 val_acc=0.7798 val_f1=0.7764
-resnet101 fold 1 epoch 011: train_loss=0.5364 val_acc=0.7807 val_f1=0.7929
-resnet101 fold 1 epoch 012: train_loss=0.4979 val_acc=0.7788 val_f1=0.7763
-resnet101 fold 1 epoch 013: train_loss=0.4991 val_acc=0.7741 val_f1=0.7750
-resnet101 fold 1 epoch 014: train_loss=0.4714 val_acc=0.7751 val_f1=0.8008
-resnet101 fold 1 epoch 015: train_loss=0.4581 val_acc=0.7732 val_f1=0.7938
-resnet101 fold 1 epoch 016: train_loss=0.4489 val_acc=0.7723 val_f1=0.7852
-resnet101 fold 1 epoch 017: train_loss=0.4407 val_acc=0.7648 val_f1=0.7927
-Early stopping resnet101 fold 1 at epoch 17.
-resnet101 fold 2 epoch 001: train_loss=1.2601 val_acc=0.6963 val_f1=0.5997
-resnet101 fold 2 epoch 002: train_loss=0.8513 val_acc=0.7498 val_f1=0.7341
-resnet101 fold 2 epoch 003: train_loss=0.7722 val_acc=0.7535 val_f1=0.7221
-resnet101 fold 2 epoch 004: train_loss=0.7189 val_acc=0.7348 val_f1=0.7185
-resnet101 fold 2 epoch 005: train_loss=0.6603 val_acc=0.7554 val_f1=0.7542
-resnet101 fold 2 epoch 006: train_loss=0.6438 val_acc=0.7685 val_f1=0.7729
-resnet101 fold 2 epoch 007: train_loss=0.6089 val_acc=0.7601 val_f1=0.7629
-resnet101 fold 2 epoch 008: train_loss=0.6051 val_acc=0.7441 val_f1=0.7414
-resnet101 fold 2 epoch 009: train_loss=0.5789 val_acc=0.7676 val_f1=0.7673
-resnet101 fold 2 epoch 010: train_loss=0.5386 val_acc=0.7676 val_f1=0.7757
-resnet101 fold 2 epoch 011: train_loss=0.5167 val_acc=0.7769 val_f1=0.7657
-resnet101 fold 2 epoch 012: train_loss=0.5209 val_acc=0.7657 val_f1=0.7766
-resnet101 fold 2 epoch 013: train_loss=0.5105 val_acc=0.7545 val_f1=0.7463
-resnet101 fold 2 epoch 014: train_loss=0.4765 val_acc=0.7648 val_f1=0.7653
-resnet101 fold 2 epoch 015: train_loss=0.4659 val_acc=0.7788 val_f1=0.7854
-resnet101 fold 2 epoch 016: train_loss=0.4304 val_acc=0.7573 val_f1=0.7730
-resnet101 fold 2 epoch 017: train_loss=0.4273 val_acc=0.7582 val_f1=0.7648
-resnet101 fold 2 epoch 018: train_loss=0.4260 val_acc=0.7554 val_f1=0.7723
-resnet101 fold 2 epoch 019: train_loss=0.3868 val_acc=0.7526 val_f1=0.7661
-resnet101 fold 2 epoch 020: train_loss=0.3849 val_acc=0.7563 val_f1=0.7755
-resnet101 fold 2 epoch 021: train_loss=0.3834 val_acc=0.7591 val_f1=0.7748
-resnet101 fold 2 epoch 022: train_loss=0.3641 val_acc=0.7610 val_f1=0.7739
-resnet101 fold 2 epoch 023: train_loss=0.3567 val_acc=0.7573 val_f1=0.7670
-Early stopping resnet101 fold 2 at epoch 23.
-resnet101 fold 3 epoch 001: train_loss=1.2696 val_acc=0.7085 val_f1=0.6328
-resnet101 fold 3 epoch 002: train_loss=0.8836 val_acc=0.7282 val_f1=0.6988
-resnet101 fold 3 epoch 003: train_loss=0.7623 val_acc=0.7516 val_f1=0.7538
-resnet101 fold 3 epoch 004: train_loss=0.7316 val_acc=0.7685 val_f1=0.7690
-resnet101 fold 3 epoch 005: train_loss=0.6781 val_acc=0.7516 val_f1=0.7767
-resnet101 fold 3 epoch 006: train_loss=0.6471 val_acc=0.7432 val_f1=0.7552
-resnet101 fold 3 epoch 007: train_loss=0.6075 val_acc=0.7638 val_f1=0.7812
-resnet101 fold 3 epoch 008: train_loss=0.5822 val_acc=0.7470 val_f1=0.7741
-resnet101 fold 3 epoch 009: train_loss=0.5710 val_acc=0.7760 val_f1=0.7941
-resnet101 fold 3 epoch 010: train_loss=0.5624 val_acc=0.7516 val_f1=0.7684
-resnet101 fold 3 epoch 011: train_loss=0.5447 val_acc=0.7591 val_f1=0.7676
-resnet101 fold 3 epoch 012: train_loss=0.5079 val_acc=0.7891 val_f1=0.8002
-resnet101 fold 3 epoch 013: train_loss=0.4843 val_acc=0.7582 val_f1=0.7776
-resnet101 fold 3 epoch 014: train_loss=0.4691 val_acc=0.7648 val_f1=0.7855
-resnet101 fold 3 epoch 015: train_loss=0.4443 val_acc=0.7591 val_f1=0.7890
-resnet101 fold 3 epoch 016: train_loss=0.4442 val_acc=0.7516 val_f1=0.7597
-resnet101 fold 3 epoch 017: train_loss=0.4355 val_acc=0.7666 val_f1=0.7934
-resnet101 fold 3 epoch 018: train_loss=0.3951 val_acc=0.7535 val_f1=0.7635
-resnet101 fold 3 epoch 019: train_loss=0.3905 val_acc=0.7535 val_f1=0.7719
-resnet101 fold 3 epoch 020: train_loss=0.3941 val_acc=0.7545 val_f1=0.7877
-Early stopping resnet101 fold 3 at epoch 20.
-resnet101 fold 4 epoch 001: train_loss=1.2787 val_acc=0.7348 val_f1=0.6684
-resnet101 fold 4 epoch 002: train_loss=0.8956 val_acc=0.7573 val_f1=0.7293
-resnet101 fold 4 epoch 003: train_loss=0.7928 val_acc=0.7648 val_f1=0.7513
-resnet101 fold 4 epoch 004: train_loss=0.7305 val_acc=0.7769 val_f1=0.7549
-resnet101 fold 4 epoch 005: train_loss=0.6968 val_acc=0.7282 val_f1=0.7370
-resnet101 fold 4 epoch 006: train_loss=0.6480 val_acc=0.7901 val_f1=0.7617
-resnet101 fold 4 epoch 007: train_loss=0.6174 val_acc=0.7891 val_f1=0.7732
-resnet101 fold 4 epoch 008: train_loss=0.5667 val_acc=0.7816 val_f1=0.7801
-resnet101 fold 4 epoch 009: train_loss=0.5692 val_acc=0.7798 val_f1=0.7670
-resnet101 fold 4 epoch 010: train_loss=0.5377 val_acc=0.7835 val_f1=0.7795
-resnet101 fold 4 epoch 011: train_loss=0.5430 val_acc=0.7910 val_f1=0.7928
-resnet101 fold 4 epoch 012: train_loss=0.5070 val_acc=0.7713 val_f1=0.7788
-resnet101 fold 4 epoch 013: train_loss=0.4884 val_acc=0.7648 val_f1=0.7732
-resnet101 fold 4 epoch 014: train_loss=0.4618 val_acc=0.7676 val_f1=0.7766
-resnet101 fold 4 epoch 015: train_loss=0.4641 val_acc=0.7769 val_f1=0.7875
-resnet101 fold 4 epoch 016: train_loss=0.4474 val_acc=0.7535 val_f1=0.7681
-resnet101 fold 4 epoch 017: train_loss=0.4245 val_acc=0.7554 val_f1=0.7595
-resnet101 fold 4 epoch 018: train_loss=0.4179 val_acc=0.7648 val_f1=0.7604
-resnet101 fold 4 epoch 019: train_loss=0.4066 val_acc=0.7601 val_f1=0.7627
-Early stopping resnet101 fold 4 at epoch 19.
-resnet101 fold 5 epoch 001: train_loss=1.2793 val_acc=0.7188 val_f1=0.6415
-resnet101 fold 5 epoch 002: train_loss=0.8835 val_acc=0.7685 val_f1=0.7286
-resnet101 fold 5 epoch 003: train_loss=0.7872 val_acc=0.7760 val_f1=0.7876
-resnet101 fold 5 epoch 004: train_loss=0.7318 val_acc=0.7844 val_f1=0.7712
-resnet101 fold 5 epoch 005: train_loss=0.6766 val_acc=0.7798 val_f1=0.7861
-resnet101 fold 5 epoch 006: train_loss=0.6492 val_acc=0.7882 val_f1=0.8110
-resnet101 fold 5 epoch 007: train_loss=0.6413 val_acc=0.7844 val_f1=0.8058
-resnet101 fold 5 epoch 008: train_loss=0.5910 val_acc=0.7816 val_f1=0.8041
-resnet101 fold 5 epoch 009: train_loss=0.5720 val_acc=0.8079 val_f1=0.8098
-resnet101 fold 5 epoch 010: train_loss=0.5494 val_acc=0.7854 val_f1=0.8072
-resnet101 fold 5 epoch 011: train_loss=0.5358 val_acc=0.7919 val_f1=0.8137
-resnet101 fold 5 epoch 012: train_loss=0.5337 val_acc=0.7882 val_f1=0.7707
-resnet101 fold 5 epoch 013: train_loss=0.4927 val_acc=0.7976 val_f1=0.8154
-resnet101 fold 5 epoch 014: train_loss=0.4817 val_acc=0.7966 val_f1=0.8248
-resnet101 fold 5 epoch 015: train_loss=0.4442 val_acc=0.7976 val_f1=0.8172
-resnet101 fold 5 epoch 016: train_loss=0.4479 val_acc=0.7873 val_f1=0.8154
-resnet101 fold 5 epoch 017: train_loss=0.4399 val_acc=0.7723 val_f1=0.8023
-Early stopping resnet101 fold 5 at epoch 17.
-transformer fold 1 epoch 001: train_loss=1.7890 val_acc=0.4545 val_f1=0.1538
-transformer fold 1 epoch 002: train_loss=1.6913 val_acc=0.3618 val_f1=0.1224
-transformer fold 1 epoch 003: train_loss=1.6651 val_acc=0.4386 val_f1=0.1267
-transformer fold 1 epoch 004: train_loss=1.5711 val_acc=0.4470 val_f1=0.1386
-transformer fold 1 epoch 005: train_loss=1.4982 val_acc=0.5033 val_f1=0.2069
-transformer fold 1 epoch 006: train_loss=1.4373 val_acc=0.5155 val_f1=0.2948
-transformer fold 1 epoch 007: train_loss=1.3697 val_acc=0.5080 val_f1=0.2548
-transformer fold 1 epoch 008: train_loss=1.3298 val_acc=0.5558 val_f1=0.3806
-transformer fold 1 epoch 009: train_loss=1.2885 val_acc=0.5436 val_f1=0.3501
-transformer fold 1 epoch 010: train_loss=1.2746 val_acc=0.5623 val_f1=0.4145
-transformer fold 1 epoch 011: train_loss=1.2500 val_acc=0.5764 val_f1=0.4555
-transformer fold 1 epoch 012: train_loss=1.2101 val_acc=0.5511 val_f1=0.3780
-transformer fold 1 epoch 013: train_loss=1.2042 val_acc=0.5708 val_f1=0.4295
-transformer fold 1 epoch 014: train_loss=1.1624 val_acc=0.5801 val_f1=0.4341
-transformer fold 1 epoch 015: train_loss=1.1697 val_acc=0.5820 val_f1=0.4762
-transformer fold 1 epoch 016: train_loss=1.1371 val_acc=0.5886 val_f1=0.4580
-transformer fold 1 epoch 017: train_loss=1.1139 val_acc=0.5614 val_f1=0.4915
-transformer fold 1 epoch 018: train_loss=1.1123 val_acc=0.5904 val_f1=0.4993
-transformer fold 1 epoch 019: train_loss=1.0791 val_acc=0.5848 val_f1=0.5075
-transformer fold 1 epoch 020: train_loss=1.0824 val_acc=0.5783 val_f1=0.4949
-transformer fold 1 epoch 021: train_loss=1.0439 val_acc=0.6204 val_f1=0.5489
-transformer fold 1 epoch 022: train_loss=1.0530 val_acc=0.6270 val_f1=0.5409
-transformer fold 1 epoch 023: train_loss=1.0331 val_acc=0.6336 val_f1=0.5405
-transformer fold 1 epoch 024: train_loss=1.0042 val_acc=0.6336 val_f1=0.5429
-transformer fold 1 epoch 025: train_loss=0.9947 val_acc=0.6410 val_f1=0.5528
-transformer fold 1 epoch 026: train_loss=0.9898 val_acc=0.6354 val_f1=0.5564
-transformer fold 1 epoch 027: train_loss=0.9719 val_acc=0.6345 val_f1=0.5647
-transformer fold 1 epoch 028: train_loss=0.9587 val_acc=0.6336 val_f1=0.5644
-transformer fold 1 epoch 029: train_loss=0.9607 val_acc=0.6307 val_f1=0.5612
-transformer fold 1 epoch 030: train_loss=0.9711 val_acc=0.6354 val_f1=0.5662
-transformer fold 2 epoch 001: train_loss=1.7973 val_acc=0.4264 val_f1=0.1412
-transformer fold 2 epoch 002: train_loss=1.7250 val_acc=0.4292 val_f1=0.1370
-transformer fold 2 epoch 003: train_loss=1.6780 val_acc=0.3871 val_f1=0.1190
-transformer fold 2 epoch 004: train_loss=1.6300 val_acc=0.4620 val_f1=0.1556
-transformer fold 2 epoch 005: train_loss=1.5385 val_acc=0.4452 val_f1=0.1827
-transformer fold 2 epoch 006: train_loss=1.5021 val_acc=0.4799 val_f1=0.1990
-transformer fold 2 epoch 007: train_loss=1.4328 val_acc=0.4939 val_f1=0.2402
-transformer fold 2 epoch 008: train_loss=1.3746 val_acc=0.5239 val_f1=0.2762
-transformer fold 2 epoch 009: train_loss=1.3283 val_acc=0.5145 val_f1=0.2741
-transformer fold 2 epoch 010: train_loss=1.2757 val_acc=0.5595 val_f1=0.4045
-transformer fold 2 epoch 011: train_loss=1.2435 val_acc=0.5455 val_f1=0.3783
-transformer fold 2 epoch 012: train_loss=1.2287 val_acc=0.5417 val_f1=0.4062
-transformer fold 2 epoch 013: train_loss=1.1795 val_acc=0.5764 val_f1=0.4130
-transformer fold 2 epoch 014: train_loss=1.1663 val_acc=0.5633 val_f1=0.4707
-transformer fold 2 epoch 015: train_loss=1.1543 val_acc=0.5520 val_f1=0.4112
-transformer fold 2 epoch 016: train_loss=1.1343 val_acc=0.5717 val_f1=0.4670
-transformer fold 2 epoch 017: train_loss=1.1155 val_acc=0.5754 val_f1=0.4609
-transformer fold 2 epoch 018: train_loss=1.0962 val_acc=0.5839 val_f1=0.4639
-transformer fold 2 epoch 019: train_loss=1.0594 val_acc=0.6186 val_f1=0.5140
-transformer fold 2 epoch 020: train_loss=1.0589 val_acc=0.6204 val_f1=0.5181
-transformer fold 2 epoch 021: train_loss=1.0357 val_acc=0.6148 val_f1=0.4768
-transformer fold 2 epoch 022: train_loss=1.0379 val_acc=0.6111 val_f1=0.5108
-transformer fold 2 epoch 023: train_loss=1.0107 val_acc=0.6007 val_f1=0.4882
-transformer fold 2 epoch 024: train_loss=0.9895 val_acc=0.6167 val_f1=0.4868
-transformer fold 2 epoch 025: train_loss=1.0005 val_acc=0.6167 val_f1=0.4956
-transformer fold 2 epoch 026: train_loss=0.9839 val_acc=0.6064 val_f1=0.4909
-transformer fold 2 epoch 027: train_loss=0.9782 val_acc=0.6176 val_f1=0.5157
-transformer fold 2 epoch 028: train_loss=0.9594 val_acc=0.6223 val_f1=0.5239
-transformer fold 2 epoch 029: train_loss=0.9429 val_acc=0.6214 val_f1=0.5231
-transformer fold 2 epoch 030: train_loss=0.9438 val_acc=0.6186 val_f1=0.5163
-transformer fold 3 epoch 001: train_loss=1.8194 val_acc=0.4030 val_f1=0.1121
-transformer fold 3 epoch 002: train_loss=1.7179 val_acc=0.4311 val_f1=0.1237
-transformer fold 3 epoch 003: train_loss=1.6587 val_acc=0.4442 val_f1=0.1447
-transformer fold 3 epoch 004: train_loss=1.6063 val_acc=0.4592 val_f1=0.1461
-transformer fold 3 epoch 005: train_loss=1.5399 val_acc=0.4967 val_f1=0.2444
-transformer fold 3 epoch 006: train_loss=1.4701 val_acc=0.4930 val_f1=0.1992
-transformer fold 3 epoch 007: train_loss=1.4348 val_acc=0.5248 val_f1=0.2917
-transformer fold 3 epoch 008: train_loss=1.3977 val_acc=0.5530 val_f1=0.2901
-transformer fold 3 epoch 009: train_loss=1.3754 val_acc=0.5464 val_f1=0.2959
-transformer fold 3 epoch 010: train_loss=1.3446 val_acc=0.5473 val_f1=0.3072
-transformer fold 3 epoch 011: train_loss=1.3294 val_acc=0.5342 val_f1=0.3095
-transformer fold 3 epoch 012: train_loss=1.3069 val_acc=0.5464 val_f1=0.3902
-transformer fold 3 epoch 013: train_loss=1.2667 val_acc=0.5604 val_f1=0.3484
-transformer fold 3 epoch 014: train_loss=1.2513 val_acc=0.5773 val_f1=0.4435
-transformer fold 3 epoch 015: train_loss=1.2289 val_acc=0.5520 val_f1=0.4769
-transformer fold 3 epoch 016: train_loss=1.2169 val_acc=0.5483 val_f1=0.4324
-transformer fold 3 epoch 017: train_loss=1.2074 val_acc=0.5933 val_f1=0.4728
-transformer fold 3 epoch 018: train_loss=1.1674 val_acc=0.5829 val_f1=0.4776
-transformer fold 3 epoch 019: train_loss=1.1602 val_acc=0.5979 val_f1=0.4623
-transformer fold 3 epoch 020: train_loss=1.1451 val_acc=0.6082 val_f1=0.5212
-transformer fold 3 epoch 021: train_loss=1.1217 val_acc=0.6073 val_f1=0.4718
-transformer fold 3 epoch 022: train_loss=1.1301 val_acc=0.5998 val_f1=0.4842
-transformer fold 3 epoch 023: train_loss=1.0847 val_acc=0.6101 val_f1=0.4863
-transformer fold 3 epoch 024: train_loss=1.0849 val_acc=0.6186 val_f1=0.4956
-transformer fold 3 epoch 025: train_loss=1.0778 val_acc=0.6214 val_f1=0.4912
-transformer fold 3 epoch 026: train_loss=1.0571 val_acc=0.6270 val_f1=0.5048
-transformer fold 3 epoch 027: train_loss=1.0570 val_acc=0.6167 val_f1=0.4825
-transformer fold 3 epoch 028: train_loss=1.0447 val_acc=0.6195 val_f1=0.4920
-transformer fold 3 epoch 029: train_loss=1.0606 val_acc=0.6214 val_f1=0.4925
-transformer fold 3 epoch 030: train_loss=1.0339 val_acc=0.6214 val_f1=0.4939
-transformer fold 4 epoch 001: train_loss=1.7875 val_acc=0.4105 val_f1=0.1297
-transformer fold 4 epoch 002: train_loss=1.7089 val_acc=0.4302 val_f1=0.1406
-transformer fold 4 epoch 003: train_loss=1.6493 val_acc=0.4302 val_f1=0.1726
-transformer fold 4 epoch 004: train_loss=1.5695 val_acc=0.3918 val_f1=0.1488
-transformer fold 4 epoch 005: train_loss=1.5149 val_acc=0.4911 val_f1=0.2165
-transformer fold 4 epoch 006: train_loss=1.4381 val_acc=0.5370 val_f1=0.3032
-transformer fold 4 epoch 007: train_loss=1.3845 val_acc=0.5689 val_f1=0.3141
-transformer fold 4 epoch 008: train_loss=1.3289 val_acc=0.5098 val_f1=0.2904
-transformer fold 4 epoch 009: train_loss=1.2909 val_acc=0.5811 val_f1=0.3882
-transformer fold 4 epoch 010: train_loss=1.2593 val_acc=0.5848 val_f1=0.3814
-transformer fold 4 epoch 011: train_loss=1.2390 val_acc=0.5586 val_f1=0.4543
-transformer fold 4 epoch 012: train_loss=1.2001 val_acc=0.5820 val_f1=0.4300
-transformer fold 4 epoch 013: train_loss=1.1804 val_acc=0.5867 val_f1=0.4860
-transformer fold 4 epoch 014: train_loss=1.1612 val_acc=0.6186 val_f1=0.5195
-transformer fold 4 epoch 015: train_loss=1.1549 val_acc=0.6129 val_f1=0.5202
-transformer fold 4 epoch 016: train_loss=1.1292 val_acc=0.6232 val_f1=0.5332
-transformer fold 4 epoch 017: train_loss=1.1110 val_acc=0.6261 val_f1=0.5095
-transformer fold 4 epoch 018: train_loss=1.0963 val_acc=0.6336 val_f1=0.5333
-transformer fold 4 epoch 019: train_loss=1.0797 val_acc=0.6682 val_f1=0.5735
-transformer fold 4 epoch 020: train_loss=1.0481 val_acc=0.6298 val_f1=0.5320
-transformer fold 4 epoch 021: train_loss=1.0657 val_acc=0.6289 val_f1=0.5371
-transformer fold 4 epoch 022: train_loss=1.0421 val_acc=0.6467 val_f1=0.5815
-transformer fold 4 epoch 023: train_loss=1.0175 val_acc=0.6664 val_f1=0.5736
-transformer fold 4 epoch 024: train_loss=1.0025 val_acc=0.6626 val_f1=0.5565
-transformer fold 4 epoch 025: train_loss=0.9965 val_acc=0.6607 val_f1=0.5808
-transformer fold 4 epoch 026: train_loss=0.9770 val_acc=0.6570 val_f1=0.5570
-transformer fold 4 epoch 027: train_loss=0.9740 val_acc=0.6664 val_f1=0.5727
-Early stopping transformer fold 4 at epoch 27.
-transformer fold 5 epoch 001: train_loss=1.8162 val_acc=0.4049 val_f1=0.1079
-transformer fold 5 epoch 002: train_loss=1.7148 val_acc=0.4246 val_f1=0.1348
-transformer fold 5 epoch 003: train_loss=1.6742 val_acc=0.4442 val_f1=0.1462
-transformer fold 5 epoch 004: train_loss=1.6135 val_acc=0.4667 val_f1=0.1560
-transformer fold 5 epoch 005: train_loss=1.5457 val_acc=0.4817 val_f1=0.2032
-transformer fold 5 epoch 006: train_loss=1.4678 val_acc=0.4920 val_f1=0.2522
-transformer fold 5 epoch 007: train_loss=1.3934 val_acc=0.4995 val_f1=0.2501
-transformer fold 5 epoch 008: train_loss=1.3623 val_acc=0.5201 val_f1=0.2886
-transformer fold 5 epoch 009: train_loss=1.3225 val_acc=0.5492 val_f1=0.3439
-transformer fold 5 epoch 010: train_loss=1.2764 val_acc=0.5501 val_f1=0.3092
-transformer fold 5 epoch 011: train_loss=1.2534 val_acc=0.5426 val_f1=0.4440
-transformer fold 5 epoch 012: train_loss=1.2217 val_acc=0.5773 val_f1=0.3614
-transformer fold 5 epoch 013: train_loss=1.1942 val_acc=0.5839 val_f1=0.5061
-transformer fold 5 epoch 014: train_loss=1.1847 val_acc=0.5679 val_f1=0.5256
-transformer fold 5 epoch 015: train_loss=1.1453 val_acc=0.6036 val_f1=0.5105
-transformer fold 5 epoch 016: train_loss=1.1344 val_acc=0.6082 val_f1=0.4969
-transformer fold 5 epoch 017: train_loss=1.1182 val_acc=0.6111 val_f1=0.5293
-transformer fold 5 epoch 018: train_loss=1.0982 val_acc=0.5876 val_f1=0.4895
-transformer fold 5 epoch 019: train_loss=1.0747 val_acc=0.6139 val_f1=0.5021
-transformer fold 5 epoch 020: train_loss=1.0591 val_acc=0.5886 val_f1=0.4897
-transformer fold 5 epoch 021: train_loss=1.0570 val_acc=0.5961 val_f1=0.5433
-transformer fold 5 epoch 022: train_loss=1.0309 val_acc=0.6129 val_f1=0.5539
-transformer fold 5 epoch 023: train_loss=1.0180 val_acc=0.6129 val_f1=0.5394
-transformer fold 5 epoch 024: train_loss=1.0082 val_acc=0.6326 val_f1=0.5833
-transformer fold 5 epoch 025: train_loss=0.9965 val_acc=0.6336 val_f1=0.5681
-transformer fold 5 epoch 026: train_loss=0.9862 val_acc=0.6373 val_f1=0.5735
-transformer fold 5 epoch 027: train_loss=0.9789 val_acc=0.6373 val_f1=0.5754
-transformer fold 5 epoch 028: train_loss=0.9827 val_acc=0.6289 val_f1=0.5616
-transformer fold 5 epoch 029: train_loss=0.9827 val_acc=0.6326 val_f1=0.5656
-transformer fold 5 epoch 030: train_loss=0.9547 val_acc=0.6326 val_f1=0.5653
-Saved K-fold results to kfold_results_full