update configure files for examples

taigw · taigw · commit 35613584facd · 2023-02-16T12:53:14.000+08:00
diff --git a/classification/AntBee/README.md b/classification/AntBee/README.md
@@ -13,14 +13,14 @@ In this example, we finetune a pretrained resnet18 for classification of images
 [data_link]:https://download.pytorch.org/tutorial/hymenoptera_data.zip
 
 ## Finetuning all layers of resnet18
-1. Here we use resnet18 for finetuning, and update all the layers. Open the configure file `config/train_test_ce1.cfg`. In  the `network` section we can find details for the network. Here `update_layers = 0` means updating all the layers.
+1. Here we use resnet18 for finetuning, and update all the layers. Open the configure file `config/train_test_ce1.cfg`. In  the `network` section we can find details for the network. Here `update_mode = all` means updating all the layers.
 ```bash
 # type of network
 net_type   = resnet18
 pretrain   = True
 input_chns = 3
 # finetune all the layers
-update_layers = 0  
+update_mode = all 
 ```
 
 Then start to train by running:
@@ -48,20 +48,20 @@ pymic_run test config/train_test_ce1.cfg
 pymic_eval_cls config/evaluation.cfg
 ```
 
-The obtained accuracy by default setting should be around 0.9412, and the AUC will be around 0.976.
+The obtained accuracy by default setting should be around 0.9477, and the AUC will be around 0.9745.
 
 3. Run `python show_roc.py` to show the receiver operating characteristic curve. 
 
 ![roc](./picture/roc.png)
 
 ## Finetuning the last layer of resnet18
-Similarly to the above example, we further try to only finetune the last layer of resnet18 for the same classification task. Use a different configure file `config/train_test_ce2.cfg` for training and testing, where `update_layers = -1` in the `network` section means updating the last layer only:
+Similarly to the above example, we further try to only finetune the last layer of resnet18 for the same classification task. Use a different configure file `config/train_test_ce2.cfg` for training and testing, where `update_mode = last` in the `network` section means updating the last layer only:
 ```bash
 net_type   = resnet18
 pretrain   = True
 input_chns = 3
 # finetune the last layer only
-update_layers = -1
+update_mode = last
 ```
 
-Edit `config/evaluation.cfg` accordinly for evaluation. 
+Edit `config/evaluation.cfg` accordinly for evaluation. The corresponding accuracy and AUC would be around 0.9477 and 0.9778, respectively.
diff --git a/classification/AntBee/config/train_test_ce1.cfg b/classification/AntBee/config/train_test_ce1.cfg
@@ -14,8 +14,8 @@ train_batch_size = 4
 modal_num = 1
 
 # data transforms
-train_transform = [Rescale, RandomCrop, RandomFlip, NormalizeWithMeanStd]
-valid_transform = [Rescale, CenterCrop, NormalizeWithMeanStd]
+train_transform = [Rescale, RandomCrop, RandomFlip, NormalizeWithMeanStd, LabelToProbability]
+valid_transform = [Rescale, CenterCrop, NormalizeWithMeanStd, LabelToProbability]
 test_transform  = [Rescale, CenterCrop, NormalizeWithMeanStd]
 
 Rescale_output_size    = [256, 256]
@@ -39,7 +39,7 @@ net_type   = resnet18
 pretrain   = True
 input_chns = 3
 # finetune all the layers
-update_layers = 0  
+update_mode = all
 
 # number of classes
 class_num = 2
@@ -56,19 +56,20 @@ learning_rate = 1e-3
 momentum      = 0.9
 weight_decay  = 1e-5
 
-# for lr schedular (MultiStepLR)
-lr_scheduler  = MultiStepLR
-lr_gamma      = 0.1
-lr_milestones = [500, 1000]
+# for lr schedular (StepLR)
+lr_scheduler = StepLR
+lr_gamma     = 0.5
+lr_step      = 500
 
-ckpt_save_dir    = model/resnet18_ce1
-ckpt_prefix = resnet18
+ckpt_save_dir = model/resnet18_ce1
+ckpt_prefix   = resnet18
 
 # iteration
 iter_start = 0
-iter_max   = 1500
+iter_max   = 2000
 iter_valid = 100
-iter_save  = 1500
+iter_save  = 2000
+early_stop_patience = 1000
 
 [testing]
 # list of gpus
diff --git a/classification/AntBee/config/train_test_ce2.cfg b/classification/AntBee/config/train_test_ce2.cfg
@@ -14,8 +14,8 @@ train_batch_size = 4
 modal_num = 1
 
 # data transforms
-train_transform = [Rescale, RandomCrop, RandomFlip, NormalizeWithMeanStd]
-valid_transform = [Rescale, CenterCrop, NormalizeWithMeanStd]
+train_transform = [Rescale, RandomCrop, RandomFlip, NormalizeWithMeanStd, LabelToProbability]
+valid_transform = [Rescale, CenterCrop, NormalizeWithMeanStd, LabelToProbability]
 test_transform  = [Rescale, CenterCrop, NormalizeWithMeanStd]
 
 Rescale_output_size    = [256, 256]
@@ -39,8 +39,7 @@ net_type   = resnet18
 pretrain   = True
 input_chns = 3
 # finetune the last layer only
-update_layers = -1
-
+update_mode = last
 
 # number of classes
 class_num = 2
@@ -57,19 +56,20 @@ learning_rate = 1e-3
 momentum      = 0.9
 weight_decay  = 1e-5
 
-# for lr schedular (MultiStepLR)
-lr_scheduler  = MultiStepLR
-lr_gamma      = 0.1
-lr_milestones = [500, 1000]
+# for lr schedular (StepLR)
+lr_scheduler = StepLR
+lr_gamma     = 0.5
+lr_step      = 500
 
 ckpt_save_dir    = model/resnet18_ce2
 ckpt_prefix = resnet18
 
 # iteration
 iter_start = 0
-iter_max   = 1500
+iter_max   = 2000
 iter_valid = 100
-iter_save  = 1500
+iter_save  = 2000
+early_stop_patience = 1000
 
 [testing]
 # list of gpus
diff --git a/classification/CHNCXR/README.md b/classification/CHNCXR/README.md
@@ -20,7 +20,7 @@ net_type = resnet18
 pretrain = True
 input_chns = 3
 # finetune all the layers
-update_layers = 0
+update_mode = all
 ```
 
 Start to train by running:
@@ -48,12 +48,12 @@ pymic_run test config/net_resnet18.cfg
 pymic_eval_cls config/evaluation.cfg
 ```
 
-The obtained accuracy by default setting should be around 0.8571, and the AUC is 0.94.
+The obtained accuracy by default setting should be around 0.8271, and the AUC is 0.9343.
 
 3. Run `python show_roc.py` to show the receiver operating characteristic curve. 
 
 ![roc](./picture/roc.png)
 
 
 ## Finetuning vgg16
-Similarly to the above example, we further try to finetune vgg16 for the same classification task. Use a different configure file `config/net_vg16.cfg` for training and testing. Edit `config/evaluation.cfg` accordinly for evaluation. The iteration number for the highest accuracy on the validation set was 2300, and the accuracy will be around 0.8797. 
+Similarly to the above example, we further try to finetune vgg16 for the same classification task. Use a different configure file `config/net_vg16.cfg` for training and testing. Edit `config/evaluation.cfg` accordinly for evaluation. The accuracy and AUC would be around 0.8571 and 0.9271, respectively. 
diff --git a/classification/CHNCXR/config/net_resnet18.cfg b/classification/CHNCXR/config/net_resnet18.cfg
@@ -14,8 +14,8 @@ train_batch_size = 4
 modal_num = 1
 
 # data transforms
-train_transform = [Rescale, RandomCrop, RandomFlip, GrayscaleToRGB, NormalizeWithMeanStd]
-valid_transform = [Rescale, CenterCrop, GrayscaleToRGB, NormalizeWithMeanStd]
+train_transform = [Rescale, RandomCrop, RandomFlip, GrayscaleToRGB, NormalizeWithMeanStd, LabelToProbability]
+valid_transform = [Rescale, CenterCrop, GrayscaleToRGB, NormalizeWithMeanStd, LabelToProbability]
 test_transform  = [Rescale, CenterCrop, GrayscaleToRGB, NormalizeWithMeanStd]
 
 Rescale_output_size    = [256, 256]
@@ -37,7 +37,7 @@ net_type = resnet18
 pretrain = True
 input_chns = 3
 # finetune all the layers
-update_layers = 0
+update_mode = all
 
 # number of classes
 class_num = 2
@@ -54,10 +54,10 @@ learning_rate = 1e-3
 momentum      = 0.9
 weight_decay  = 1e-5
 
-# for lr schedular (MultiStepLR)
-lr_scheduler  = MultiStepLR
-lr_gamma      = 0.1
-lr_milestones = [1500, 3000]
+# for lr schedular (StepLR)
+lr_scheduler  = StepLR
+lr_gamma      = 0.5
+lr_step       = 1000
 
 ckpt_save_dir    = model/resnet18
 ckpt_prefix      = resnet18
@@ -66,7 +66,8 @@ ckpt_prefix      = resnet18
 iter_start = 0
 iter_max   = 5000
 iter_valid = 100
-iter_save  = 1000
+iter_save  = 5000
+early_stop_patience = 2000
 
 [testing]
 # list of gpus
diff --git a/classification/CHNCXR/config/net_vgg16.cfg b/classification/CHNCXR/config/net_vgg16.cfg
@@ -14,8 +14,8 @@ train_batch_size = 4
 modal_num = 1
 
 # data transforms
-train_transform = [Rescale, RandomCrop, RandomFlip, GrayscaleToRGB, NormalizeWithMeanStd]
-valid_transform = [Rescale, CenterCrop, GrayscaleToRGB, NormalizeWithMeanStd]
+train_transform = [Rescale, RandomCrop, RandomFlip, GrayscaleToRGB, NormalizeWithMeanStd, LabelToProbability]
+valid_transform = [Rescale, CenterCrop, GrayscaleToRGB, NormalizeWithMeanStd, LabelToProbability]
 test_transform  = [Rescale, CenterCrop, GrayscaleToRGB, NormalizeWithMeanStd]
 
 Rescale_output_size    = [256, 256]
@@ -37,7 +37,7 @@ net_type   = vgg16
 pretrain   = True
 input_chns = 3
 # finetune all the layers
-update_layers = 0
+update_mode = all
 
 # number of classes
 class_num = 2
@@ -54,10 +54,10 @@ learning_rate = 1e-3
 momentum      = 0.9
 weight_decay  = 1e-5
 
-# for lr schedular (MultiStepLR)
-lr_scheduler  = MultiStepLR
-lr_gamma      = 0.1
-lr_milestones = [1500, 3000]
+# for lr schedular (StepLR)
+lr_scheduler  = StepLR
+lr_gamma      = 0.5
+lr_step       = 1000
 
 ckpt_save_dir = model/vgg16
 ckpt_prefix   = vgg16
@@ -66,7 +66,8 @@ ckpt_prefix   = vgg16
 iter_start = 0
 iter_max   = 5000
 iter_valid = 100
-iter_save  = 1000
+iter_save  = 5000
+early_stop_patience = 2000
 
 [testing]
 # list of gpus
diff --git a/seg_ssl/ACDC/config/unet2d_baseline.cfg b/seg_ssl/ACDC/config/unet2d_baseline.cfg
@@ -40,7 +40,6 @@ GammaCorrection_gamma_max = 1.5
 GaussianNoise_channels = [0]
 GaussianNoise_mean     = 0
 GaussianNoise_std      = 0.05
-GaussianNoise_probability = 0.5
 
 [network]
 # this section gives parameters for network
@@ -55,11 +54,11 @@ in_chns       = 1
 feature_chns  = [16, 32, 64, 128, 256]
 dropout       = [0.0, 0.0, 0.0, 0.5, 0.5]
 bilinear      = True
-deep_supervise= False
+multiscale_pred = False
 
 [training]
 # list of gpus
-gpus       = [0]
+gpus       = [1]
 
 loss_type     = [DiceLoss, CrossEntropyLoss]
 loss_weight   = [0.5, 0.5]
@@ -82,7 +81,7 @@ ckpt_save_dir    = model/unet2d_baseline
 iter_start = 0
 iter_max   = 30000
 iter_valid = 100
-iter_save  = [30000]
+iter_save  = 30000
 
 [testing]
 # list of gpus
diff --git a/seg_ssl/ACDC/config/unet2d_cps.cfg b/seg_ssl/ACDC/config/unet2d_cps.cfg
@@ -58,7 +58,7 @@ in_chns       = 1
 feature_chns  = [16, 32, 64, 128, 256]
 dropout       = [0.0, 0.0, 0.0, 0.5, 0.5]
 bilinear      = True
-deep_supervise= False
+multiscale_pred = False
 
 [training]
 # list of gpus
diff --git a/seg_ssl/ACDC/config/unet2d_em.cfg b/seg_ssl/ACDC/config/unet2d_em.cfg
@@ -58,7 +58,7 @@ in_chns       = 1
 feature_chns  = [16, 32, 64, 128, 256]
 dropout       = [0.0, 0.0, 0.0, 0.5, 0.5]
 bilinear      = True
-deep_supervise= False
+multiscale_pred = False
 
 [training]
 # list of gpus
diff --git a/seg_ssl/ACDC/config/unet2d_mt.cfg b/seg_ssl/ACDC/config/unet2d_mt.cfg
@@ -59,7 +59,7 @@ in_chns       = 1
 feature_chns  = [16, 32, 64, 128, 256]
 dropout       = [0.0, 0.0, 0.0, 0.5, 0.5]
 bilinear      = True
-deep_supervise= False
+multiscale_pred = False
 
 [training]
 # list of gpus
diff --git a/seg_ssl/ACDC/config/unet2d_uamt.cfg b/seg_ssl/ACDC/config/unet2d_uamt.cfg
@@ -58,7 +58,7 @@ in_chns       = 1
 feature_chns  = [16, 32, 64, 128, 256]
 dropout       = [0.0, 0.0, 0.0, 0.5, 0.5]
 bilinear      = True
-deep_supervise= False
+multiscale_pred = False
 
 [training]
 # list of gpus
diff --git a/seg_ssl/ACDC/config/unet2d_urpc.cfg b/seg_ssl/ACDC/config/unet2d_urpc.cfg
@@ -58,12 +58,13 @@ in_chns       = 1
 feature_chns  = [16, 32, 64, 128, 256]
 dropout       = [0.0, 0.0, 0.0, 0.5, 0.5]
 bilinear      = True
-deep_supervise= True
+multiscale_pred = True
 
 [training]
 # list of gpus
 gpus       = [0]
 
+deep_supervise = True
 loss_type     = [DiceLoss, CrossEntropyLoss]
 loss_weight   = [0.5, 0.5]
 
diff --git a/segmentation/JSRT/README.md b/segmentation/JSRT/README.md
@@ -37,4 +37,4 @@ pymic_run test config/unet.cfg
 pymic_eval_seg config/evaluation.cfg
 ```
 
-The obtained average Dice score by default setting should be close to 97.989%. If setting `ckpt_mode` to 1 during testing, the average Dice would be around 97.997%. 
+The obtained average Dice score by default setting should be close to 98.068%. If setting `ckpt_mode` to 1 during testing, the average Dice would be around 98.096%. 
diff --git a/segmentation/JSRT/config/unet.cfg b/segmentation/JSRT/config/unet.cfg
@@ -35,7 +35,7 @@ in_chns       = 1
 feature_chns  = [16, 32, 64, 128, 256]
 dropout       = [0,  0,  0.3, 0.4, 0.5]
 bilinear      = False
-deep_supervise= False
+multiscale_pred = False
 
 [training]
 # list of gpus
@@ -54,7 +54,7 @@ lr_scheduler  = MultiStepLR
 lr_gamma      = 0.5
 lr_milestones = [2000, 4000, 6000]
 
-ckpt_save_dir    = model/unet_dice_loss
+ckpt_save_dir    = model/unet
 ckpt_prefix = unet
 
 # start iter
@@ -69,7 +69,7 @@ gpus       = [0]
 
 # checkpoint mode can be [0-latest, 1-best, 2-specified]
 ckpt_mode         = 0
-output_dir        = result
+output_dir        = result/unet
 
 # convert the label of prediction output
 label_source = [0, 1]
diff --git a/segmentation/JSRT2/config/mynet.cfg b/segmentation/JSRT2/config/mynet.cfg
@@ -49,10 +49,10 @@ learning_rate = 1e-3
 momentum      = 0.9
 weight_decay  = 1e-5
 
-# for lr schedular (MultiStepLR)
-lr_scheduler  = MultiStepLR
+# for lr schedular (StepLR)
+lr_scheduler  = StepLR
 lr_gamma      = 0.5
-lr_milestones = [2000, 4000, 6000]
+lr_step       = 2000
 
 ckpt_save_dir    = model/mynet
 
@@ -61,6 +61,7 @@ iter_start = 0
 iter_max   = 8000
 iter_valid = 200
 iter_save  = 8000
+early_stop_patience = 2000
 
 [testing]
 # list of gpus
diff --git a/segmentation/JSRT2/net_run_jsrt.py b/segmentation/JSRT2/net_run_jsrt.py
diff --git a/segmentation/fetal_hc/config/unet.cfg b/segmentation/fetal_hc/config/unet.cfg
diff --git a/segmentation/prostate/README.md b/segmentation/prostate/README.md
diff --git a/segmentation/prostate/config/unet3d.cfg b/segmentation/prostate/config/unet3d.cfg
