-
Notifications
You must be signed in to change notification settings - Fork 0
/
training_resnet18_endo.sh
84 lines (43 loc) · 4.88 KB
/
training_resnet18_endo.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
#########################################################################################################################################
# RESNET-18
# FOLD1
# CE/DICE alone
python train_losses.py --save_path endotect/resnet18_only_ce_f1 --csv_train data_endotect/train_f1.csv --loss1 ce --mixture only_loss1 --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_only_dice_f1 --csv_train data_endotect/train_f1.csv --loss1 dice --mixture only_loss1 --model_name fpnet_resnet18_W
#CE + DICE
python train_losses.py --save_path endotect/resnet18_ce_combo_dice_f1 --csv_train data_endotect/train_f1.csv --loss1 ce --loss2 dice --mixture combo --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_linear_dice_f1 --csv_train data_endotect/train_f1.csv --loss1 ce --loss2 dice --mixture linear --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_finetune_dice_f1 --csv_train data_endotect/train_f1.csv --loss1 ce --loss2 dice --mixture fine_tune_loss2 --model_name fpnet_resnet18_W
# FOLD2
# CE/DICE alone
python train_losses.py --save_path endotect/resnet18_only_ce_f2 --csv_train data_endotect/train_f2.csv --loss1 ce --mixture only_loss1 --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_only_dice_f2 --csv_train data_endotect/train_f2.csv --loss1 dice --mixture only_loss1 --model_name fpnet_resnet18_W
#CE + DICE
python train_losses.py --save_path endotect/resnet18_ce_combo_dice_f2 --csv_train data_endotect/train_f2.csv --loss1 ce --loss2 dice --mixture combo --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_linear_dice_f2 --csv_train data_endotect/train_f2.csv --loss1 ce --loss2 dice --mixture linear --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_finetune_dice_f2 --csv_train data_endotect/train_f2.csv --loss1 ce --loss2 dice --mixture fine_tune_loss2 --model_name fpnet_resnet18_W
# FOLD3
# CE/DICE alone
python train_losses.py --save_path endotect/resnet18_only_ce_f3 --csv_train data_endotect/train_f3.csv --loss1 ce --mixture only_loss1 --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_only_dice_f3 --csv_train data_endotect/train_f3.csv --loss1 dice --mixture only_loss1 --model_name fpnet_resnet18_W
#CE + DICE
python train_losses.py --save_path endotect/resnet18_ce_combo_dice_f3 --csv_train data_endotect/train_f3.csv --loss1 ce --loss2 dice --mixture combo --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_linear_dice_f3 --csv_train data_endotect/train_f3.csv --loss1 ce --loss2 dice --mixture linear --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_finetune_dice_f3 --csv_train data_endotect/train_f3.csv --loss1 ce --loss2 dice --mixture fine_tune_loss2 --model_name fpnet_resnet18_W
# FOLD4
# CE/DICE alone
python train_losses.py --save_path endotect/resnet18_only_ce_f4 --csv_train data_endotect/train_f4.csv --loss1 ce --mixture only_loss1 --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_only_dice_f4 --csv_train data_endotect/train_f4.csv --loss1 dice --mixture only_loss1 --model_name fpnet_resnet18_W
# CE + DICE
python train_losses.py --save_path endotect/resnet18_ce_combo_dice_f4 --csv_train data_endotect/train_f4.csv --loss1 ce --loss2 dice --mixture combo --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_linear_dice_f4 --csv_train data_endotect/train_f4.csv --loss1 ce --loss2 dice --mixture linear --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_finetune_dice_f4 --csv_train data_endotect/train_f4.csv --loss1 ce --loss2 dice --mixture fine_tune_loss2 --model_name fpnet_resnet18_W
# FOLD5
# CE/DICE alone
python train_losses.py --save_path endotect/resnet18_only_ce_f5 --csv_train data_endotect/train_f5.csv --loss1 ce --mixture only_loss1 --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_only_dice_f5 --csv_train data_endotect/train_f5.csv --loss1 dice --mixture only_loss1 --model_name fpnet_resnet18_W
# CE + DICE
python train_losses.py --save_path endotect/resnet18_ce_combo_dice_f5 --csv_train data_endotect/train_f5.csv --loss1 ce --loss2 dice --mixture combo --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_linear_dice_f5 --csv_train data_endotect/train_f5.csv --loss1 ce --loss2 dice --mixture linear --model_name fpnet_resnet18_W
python train_losses.py --save_path endotect/resnet18_ce_finetune_dice_f5 --csv_train data_endotect/train_f5.csv --loss1 ce --loss2 dice --mixture fine_tune_loss2 --model_name fpnet_resnet18_W
#########################################################################################################################################