melanoma_training.txt

Loaded pretrained weights for efficientnet-b1
==================== Fold 1 ====================
Loaded pretrained weights for efficientnet-b1
Epoch 001: | Loss: 70.813 | Train acc: 0.982 | Val acc: 0.984 | Val roc_auc: 0.883 | Training time: 0:50:11
Epoch 002: | Loss: 63.951 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.840 | Training time: 0:50:02
Epoch 003: | Loss: 59.879 | Train acc: 0.982 | Val acc: 0.982 | Val roc_auc: 0.890 | Training time: 0:50:02
Epoch 004: | Loss: 58.594 | Train acc: 0.982 | Val acc: 0.984 | Val roc_auc: 0.846 | Training time: 0:50:02
Epoch 005: | Loss: 57.671 | Train acc: 0.982 | Val acc: 0.984 | Val roc_auc: 0.869 | Training time: 0:50:02
Epoch     5: reducing learning rate of group 0 to 2.0000e-04.
Epoch 006: | Loss: 49.297 | Train acc: 0.982 | Val acc: 0.980 | Val roc_auc: 0.897 | Training time: 0:50:02
Epoch 007: | Loss: 45.973 | Train acc: 0.984 | Val acc: 0.984 | Val roc_auc: 0.879 | Training time: 0:50:01
Epoch 008: | Loss: 43.654 | Train acc: 0.984 | Val acc: 0.982 | Val roc_auc: 0.895 | Training time: 0:50:02
Epoch     8: reducing learning rate of group 0 to 4.0000e-05.
Epoch 009: | Loss: 38.933 | Train acc: 0.984 | Val acc: 0.982 | Val roc_auc: 0.895 | Training time: 0:50:02
Early stopping. Best Val roc_auc: 0.897
==================== Fold 2 ====================
Loaded pretrained weights for efficientnet-b1
Epoch 001: | Loss: 71.593 | Train acc: 0.981 | Val acc: 0.984 | Val roc_auc: 0.848 | Training time: 1:13:30
Epoch 002: | Loss: 63.099 | Train acc: 0.982 | Val acc: 0.984 | Val roc_auc: 0.852 | Training time: 0:56:37
Epoch 003: | Loss: 60.964 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.889 | Training time: 1:02:47
Epoch 004: | Loss: 59.252 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.886 | Training time: 1:25:43
Epoch 005: | Loss: 57.501 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.909 | Training time: 1:06:38
Epoch 006: | Loss: 56.646 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.875 | Training time: 0:56:35
Epoch 007: | Loss: 48.586 | Train acc: 0.982 | Val acc: 0.982 | Val roc_auc: 0.905 | Training time: 0:52:27
Epoch     7: reducing learning rate of group 0 to 9.5619e-02.
Epoch 008: | Loss: 44.188 | Train acc: 0.983 | Val acc: 0.983 | Val roc_auc: 0.908 | Training time: 0:55:38
Early stopping. Best Val roc_auc: 0.909
==================== Fold 3 ====================
Loaded pretrained weights for efficientnet-b1
Epoch 001: | Loss: 69.758 | Train acc: 0.982 | Val acc: 0.980 | Val roc_auc: 0.852 | Training time: 0:56:10
Epoch 002: | Loss: 60.650 | Train acc: 0.983 | Val acc: 0.980 | Val roc_auc: 0.863 | Training time: 0:55:42
Epoch 003: | Loss: 59.665 | Train acc: 0.983 | Val acc: 0.980 | Val roc_auc: 0.856 | Training time: 1:15:41
Epoch 004: | Loss: 57.620 | Train acc: 0.983 | Val acc: 0.980 | Val roc_auc: 0.889 | Training time: 0:50:02
Epoch 005: | Loss: 55.557 | Train acc: 0.983 | Val acc: 0.980 | Val roc_auc: 0.895 | Training time: 0:50:03
Epoch 006: | Loss: 54.531 | Train acc: 0.983 | Val acc: 0.980 | Val roc_auc: 0.904 | Training time: 0:50:03
Epoch 007: | Loss: 52.382 | Train acc: 0.983 | Val acc: 0.980 | Val roc_auc: 0.885 | Training time: 0:50:03
Epoch 008: | Loss: 51.075 | Train acc: 0.983 | Val acc: 0.981 | Val roc_auc: 0.854 | Training time: 0:50:03
Epoch     8: reducing learning rate of group 0 to 2.0000e-04.
Epoch 009: | Loss: 43.227 | Train acc: 0.984 | Val acc: 0.980 | Val roc_auc: 0.912 | Training time: 0:50:02
Epoch 010: | Loss: 40.071 | Train acc: 0.984 | Val acc: 0.980 | Val roc_auc: 0.915 | Training time: 0:50:02
Epoch 011: | Loss: 36.932 | Train acc: 0.985 | Val acc: 0.980 | Val roc_auc: 0.926 | Training time: 0:52:05
Epoch 012: | Loss: 33.440 | Train acc: 0.986 | Val acc: 0.981 | Val roc_auc: 0.913 | Training time: 1:11:32
==================== Fold 4 ====================
Loaded pretrained weights for efficientnet-b1
Epoch 001: | Loss: 71.114 | Train acc: 0.981 | Val acc: 0.983 | Val roc_auc: 0.806 | Training time: 0:50:03
Epoch 002: | Loss: 62.462 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.857 | Training time: 0:50:03
Epoch 003: | Loss: 61.782 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.848 | Training time: 0:50:03
Epoch 004: | Loss: 59.294 | Train acc: 0.982 | Val acc: 0.982 | Val roc_auc: 0.853 | Training time: 0:50:03
Epoch     4: reducing learning rate of group 0 to 2.0000e-04.
Epoch 005: | Loss: 53.923 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.889 | Training time: 0:50:03
Epoch 006: | Loss: 50.626 | Train acc: 0.982 | Val acc: 0.983 | Val roc_auc: 0.878 | Training time: 0:50:03
Epoch 007: | Loss: 49.103 | Train acc: 0.982 | Val acc: 0.982 | Val roc_auc: 0.897 | Training time: 0:59:27
Epoch 008: | Loss: 46.988 | Train acc: 0.983 | Val acc: 0.982 | Val roc_auc: 0.893 | Training time: 1:10:52
Epoch 009: | Loss: 45.461 | Train acc: 0.983 | Val acc: 0.983 | Val roc_auc: 0.902 | Training time: 1:35:53
Epoch 010: | Loss: 41.838 | Train acc: 0.984 | Val acc: 0.982 | Val roc_auc: 0.897 | Training time: 1:00:26
Epoch 011: | Loss: 40.375 | Train acc: 0.984 | Val acc: 0.983 | Val roc_auc: 0.900 | Training time: 0:59:20
Epoch    11: reducing learning rate of group 0 to 4.0000e-05.
Epoch 012: | Loss: 35.203 | Train acc: 0.986 | Val acc: 0.983 | Val roc_auc: 0.904 | Training time: 1:18:08