Update python_bundle_workflow

bundle/python_bundle_workflow/scripts/inference.py

@@ -38,6 +38,7 @@
     ScaleIntensityd,
 )
 from monai.utils import BundleProperty
+from scripts.train import prepare_data
 
 
 class InferenceWorkflow(BundleWorkflow):
@@ -46,28 +47,22 @@ class InferenceWorkflow(BundleWorkflow):
 
     """
 
-    def __init__(self, dataset_dir: str = "."):
+    def __init__(self, dataset_dir: str = "./infer"):
         super().__init__(workflow="inference")
         print_config()
         # set root log level to INFO and init a evaluation logger, will be used in `StatsHandler`
         logging.basicConfig(stream=sys.stdout, level=logging.INFO)
         get_logger("eval_log")
 
         # create a temporary directory and 40 random image, mask pairs
-        print(f"generating synthetic data to {dataset_dir} (this may take a while)")
-        for i in range(5):
-            im, seg = create_test_image_3d(128, 128, 128, num_seg_classes=1, channel_dim=-1)
-            n = nib.Nifti1Image(im, np.eye(4))
-            nib.save(n, os.path.join(dataset_dir, f"img{i:d}.nii.gz"))
-            n = nib.Nifti1Image(seg, np.eye(4))
-            nib.save(n, os.path.join(dataset_dir, f"seg{i:d}.nii.gz"))
+        prepare_data(dataset_dir=dataset_dir)
 
         self._props = {}
         self._set_props = {}
         self.dataset_dir = dataset_dir
 
     def initialize(self):
-        self.props = {}
+        self._props = {}
 
     def run(self):
         self.evaluator.run()
@@ -76,6 +71,7 @@ def finalize(self):
         pass
 
     def _set_property(self, name, property, value):
+        # stores user-reset initialized objects that should not be re-initialized.
         self._set_props[name] = value
 
     def _get_property(self, name, property):
@@ -112,7 +108,7 @@ def get_device(self):
         return torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
     def get_dataset_dir(self):
-        return "."
+        return "./infer"
 
     def get_network_def(self):
         return UNet(

bundle/python_bundle_workflow/scripts/train.py

@@ -12,6 +12,7 @@
 import logging
 import os
 import sys
+from pathlib import Path
 from glob import glob
 
 import nibabel as nib
@@ -48,32 +49,36 @@
 from monai.utils import BundleProperty, set_determinism
 
 
+def prepare_data(dataset_dir):
+    Path(dataset_dir).mkdir(exist_ok=True)
+    print(f"generating synthetic data to {dataset_dir} (this may take a while)")
+    for i in range(40):
+        im, seg = create_test_image_3d(128, 128, 128, num_seg_classes=1, channel_dim=-1)
+        n = nib.Nifti1Image(im, np.eye(4))
+        nib.save(n, os.path.join(dataset_dir, f"img{i:d}.nii.gz"))
+        n = nib.Nifti1Image(seg, np.eye(4))
+        nib.save(n, os.path.join(dataset_dir, f"seg{i:d}.nii.gz"))
+
+
 class TrainWorkflow(BundleWorkflow):
     """
     Test class simulates the bundle training workflow defined by Python script directly.
 
     """
 
-    def __init__(self, dataset_dir: str = "."):
+    def __init__(self, dataset_dir: str = "./train"):
         super().__init__(workflow="train")
         print_config()
         # set root log level to INFO and init a train logger, will be used in `StatsHandler`
         logging.basicConfig(stream=sys.stdout, level=logging.INFO)
         get_logger("train_log")
 
         # create a temporary directory and 40 random image, mask pairs
-        print(f"generating synthetic data to {dataset_dir} (this may take a while)")
-        for i in range(40):
-            im, seg = create_test_image_3d(128, 128, 128, num_seg_classes=1, channel_dim=-1)
-            n = nib.Nifti1Image(im, np.eye(4))
-            nib.save(n, os.path.join(dataset_dir, f"img{i:d}.nii.gz"))
-            n = nib.Nifti1Image(seg, np.eye(4))
-            nib.save(n, os.path.join(dataset_dir, f"seg{i:d}.nii.gz"))
+        prepare_data(dataset_dir=dataset_dir)
 
         # define buckets to store the generated properties and set properties
         self._props = {}
         self._set_props = {}
-        self.dataset_dir = dataset_dir
 
         # besides the predefined properties, this bundle workflow can also provide `network`, `loss`, `optimizer`
         self.add_property(name="network", required=True, desc="network for the training.")
@@ -82,7 +87,7 @@ def __init__(self, dataset_dir: str = "."):
 
     def initialize(self):
         set_determinism(0)
-        self.props = {}
+        self._props = {}
 
     def run(self):
         self.trainer.run()
@@ -91,6 +96,7 @@ def finalize(self):
         set_determinism(None)
 
     def _set_property(self, name, property, value):
+        # stores user-reset initialized objects that should not be re-initialized.
         self._set_props[name] = value
 
     def _get_property(self, name, property):
@@ -127,7 +133,7 @@ def get_device(self):
         return torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
     def get_dataset_dir(self):
-        return "."
+        return "./train"
 
     def get_network(self):
         return UNet(