update ensuretype (Project-MONAI#261)

Signed-off-by: Yiheng Wang <vennw@nvidia.com>

2d_classification/mednist_tutorial.ipynb

@@ -117,7 +117,7 @@
     "    RandRotate,\n",
     "    RandZoom,\n",
     "    ScaleIntensity,\n",
-    "    ToTensor,\n",
+    "    EnsureType,\n",
     ")\n",
     "from monai.utils import set_determinism\n",
     "\n",
@@ -367,15 +367,15 @@
     "        RandRotate(range_x=np.pi / 12, prob=0.5, keep_size=True),\n",
     "        RandFlip(spatial_axis=0, prob=0.5),\n",
     "        RandZoom(min_zoom=0.9, max_zoom=1.1, prob=0.5),\n",
-    "        ToTensor(),\n",
+    "        EnsureType(),\n",
     "    ]\n",
     ")\n",
     "\n",
     "val_transforms = Compose(\n",
-    "    [LoadImage(image_only=True), AddChannel(), ScaleIntensity(), ToTensor()])\n",
+    "    [LoadImage(image_only=True), AddChannel(), ScaleIntensity(), EnsureType()])\n",
     "\n",
-    "y_pred_trans = Compose([ToTensor(), Activations(softmax=True)])\n",
-    "y_trans = Compose([ToTensor(), AsDiscrete(to_onehot=True, n_classes=num_class)])"
+    "y_pred_trans = Compose([EnsureType(), Activations(softmax=True)])\n",
+    "y_trans = Compose([EnsureType(), AsDiscrete(to_onehot=True, n_classes=num_class)])"
    ]
   },
   {

2d_registration/registration_mednist.ipynb

@@ -101,7 +101,7 @@
     "    RandRotateD,\n",
     "    RandZoomD,\n",
     "    ScaleIntensityRanged,\n",
-    "    ToTensorD,\n",
+    "    EnsureTypeD,\n",
     ")\n",
     "from monai.data import DataLoader, Dataset, CacheDataset\n",
     "from monai.config import print_config, USE_COMPILED\n",
@@ -206,7 +206,7 @@
     "                             a_min=0., a_max=255., b_min=0.0, b_max=1.0, clip=True,),\n",
     "        RandRotateD(keys=[\"moving_hand\"], range_x=np.pi/4, prob=1.0, keep_size=True, mode=\"bicubic\"),\n",
     "        RandZoomD(keys=[\"moving_hand\"], min_zoom=0.9, max_zoom=1.1, prob=1.0, mode=\"bicubic\", align_corners=False),\n",
-    "        ToTensorD(keys=[\"fixed_hand\", \"moving_hand\"]),\n",
+    "        EnsureTypeD(keys=[\"fixed_hand\", \"moving_hand\"]),\n",
     "    ]\n",
     ")"
    ]

2d_segmentation/torch/unet_evaluation_array.py

@@ -24,7 +24,7 @@
 from monai.inferers import sliding_window_inference
 from monai.metrics import DiceMetric
 from monai.networks.nets import UNet
-from monai.transforms import Activations, AddChannel, AsDiscrete, Compose, LoadImage, SaveImage, ScaleIntensity, ToTensor
+from monai.transforms import Activations, AddChannel, AsDiscrete, Compose, LoadImage, SaveImage, ScaleIntensity, EnsureType
 
 
 def main(tempdir):
@@ -41,13 +41,13 @@ def main(tempdir):
     segs = sorted(glob(os.path.join(tempdir, "seg*.png")))
 
     # define transforms for image and segmentation
-    imtrans = Compose([LoadImage(image_only=True), ScaleIntensity(), AddChannel(), ToTensor()])
-    segtrans = Compose([LoadImage(image_only=True), AddChannel(), ToTensor()])
+    imtrans = Compose([LoadImage(image_only=True), ScaleIntensity(), AddChannel(), EnsureType()])
+    segtrans = Compose([LoadImage(image_only=True), AddChannel(), EnsureType()])
     val_ds = ArrayDataset(images, imtrans, segs, segtrans)
     # sliding window inference for one image at every iteration
     val_loader = DataLoader(val_ds, batch_size=1, num_workers=1, pin_memory=torch.cuda.is_available())
     dice_metric = DiceMetric(include_background=True, reduction="mean", get_not_nans=False)
-    post_trans = Compose([ToTensor(), Activations(sigmoid=True), AsDiscrete(threshold_values=True)])
+    post_trans = Compose([EnsureType(), Activations(sigmoid=True), AsDiscrete(threshold_values=True)])
     saver = SaveImage(output_dir="./output", output_ext=".png", output_postfix="seg")
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model = UNet(

2d_segmentation/torch/unet_evaluation_dict.py

@@ -24,7 +24,7 @@
 from monai.inferers import sliding_window_inference
 from monai.metrics import DiceMetric
 from monai.networks.nets import UNet
-from monai.transforms import Activations, AddChanneld, AsDiscrete, Compose, LoadImaged, SaveImage, ScaleIntensityd, ToTensord, ToTensor
+from monai.transforms import Activations, AddChanneld, AsDiscrete, Compose, LoadImaged, SaveImage, ScaleIntensityd, EnsureTyped, EnsureType
 
 
 def main(tempdir):
@@ -47,14 +47,14 @@ def main(tempdir):
             LoadImaged(keys=["img", "seg"]),
             AddChanneld(keys=["img", "seg"]),
             ScaleIntensityd(keys="img"),
-            ToTensord(keys=["img", "seg"]),
+            EnsureTyped(keys=["img", "seg"]),
         ]
     )
     val_ds = monai.data.Dataset(data=val_files, transform=val_transforms)
     # sliding window inference need to input 1 image in every iteration
     val_loader = DataLoader(val_ds, batch_size=1, num_workers=4, collate_fn=list_data_collate)
     dice_metric = DiceMetric(include_background=True, reduction="mean", get_not_nans=False)
-    post_trans = Compose([ToTensor(), Activations(sigmoid=True), AsDiscrete(threshold_values=True)])
+    post_trans = Compose([EnsureType(), Activations(sigmoid=True), AsDiscrete(threshold_values=True)])
     saver = SaveImage(output_dir="./output", output_ext=".png", output_postfix="seg")
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model = UNet(

2d_segmentation/torch/unet_training_array.py

@@ -33,7 +33,7 @@
     RandRotate90,
     RandSpatialCrop,
     ScaleIntensity,
-    ToTensor,
+    EnsureType,
 )
 from monai.visualize import plot_2d_or_3d_image
 
@@ -60,7 +60,7 @@ def main(tempdir):
             AddChannel(),
             RandSpatialCrop((96, 96), random_size=False),
             RandRotate90(prob=0.5, spatial_axes=(0, 1)),
-            ToTensor(),
+            EnsureType(),
         ]
     )
     train_segtrans = Compose(
@@ -69,11 +69,11 @@ def main(tempdir):
             AddChannel(),
             RandSpatialCrop((96, 96), random_size=False),
             RandRotate90(prob=0.5, spatial_axes=(0, 1)),
-            ToTensor(),
+            EnsureType(),
         ]
     )
-    val_imtrans = Compose([LoadImage(image_only=True), ScaleIntensity(), AddChannel(), ToTensor()])
-    val_segtrans = Compose([LoadImage(image_only=True), AddChannel(), ToTensor()])
+    val_imtrans = Compose([LoadImage(image_only=True), ScaleIntensity(), AddChannel(), EnsureType()])
+    val_segtrans = Compose([LoadImage(image_only=True), AddChannel(), EnsureType()])
 
     # define array dataset, data loader
     check_ds = ArrayDataset(images, train_imtrans, segs, train_segtrans)
@@ -88,7 +88,7 @@ def main(tempdir):
     val_ds = ArrayDataset(images[-20:], val_imtrans, segs[-20:], val_segtrans)
     val_loader = DataLoader(val_ds, batch_size=1, num_workers=4, pin_memory=torch.cuda.is_available())
     dice_metric = DiceMetric(include_background=True, reduction="mean", get_not_nans=False)
-    post_trans = Compose([ToTensor(), Activations(sigmoid=True), AsDiscrete(threshold_values=True)])
+    post_trans = Compose([EnsureType(), Activations(sigmoid=True), AsDiscrete(threshold_values=True)])
     # create UNet, DiceLoss and Adam optimizer
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model = monai.networks.nets.UNet(

2d_segmentation/torch/unet_training_dict.py

@@ -33,8 +33,8 @@
     RandCropByPosNegLabeld,
     RandRotate90d,
     ScaleIntensityd,
-    ToTensord,
-    ToTensor,
+    EnsureTyped,
+    EnsureType,
 )
 from monai.visualize import plot_2d_or_3d_image
 
@@ -65,15 +65,15 @@ def main(tempdir):
                 keys=["img", "seg"], label_key="seg", spatial_size=[96, 96], pos=1, neg=1, num_samples=4
             ),
             RandRotate90d(keys=["img", "seg"], prob=0.5, spatial_axes=[0, 1]),
-            ToTensord(keys=["img", "seg"]),
+            EnsureTyped(keys=["img", "seg"]),
         ]
     )
     val_transforms = Compose(
         [
             LoadImaged(keys=["img", "seg"]),
             AddChanneld(keys=["img", "seg"]),
             ScaleIntensityd(keys="img"),
-            ToTensord(keys=["img", "seg"]),
+            EnsureTyped(keys=["img", "seg"]),
         ]
     )
 
@@ -99,7 +99,7 @@ def main(tempdir):
     val_ds = monai.data.Dataset(data=val_files, transform=val_transforms)
     val_loader = DataLoader(val_ds, batch_size=1, num_workers=4, collate_fn=list_data_collate)
     dice_metric = DiceMetric(include_background=True, reduction="mean", get_not_nans=False)
-    post_trans = Compose([ToTensor(), Activations(sigmoid=True), AsDiscrete(threshold_values=True)])
+    post_trans = Compose([EnsureType(), Activations(sigmoid=True), AsDiscrete(threshold_values=True)])
     # create UNet, DiceLoss and Adam optimizer
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model = monai.networks.nets.UNet(

3d_classification/densenet_training_array.ipynb

@@ -103,10 +103,10 @@
     "    RandRotate90,\n",
     "    Resize,\n",
     "    ScaleIntensity,\n",
-    "    ToTensor,\n",
+    "    EnsureType,\n",
     "    Randomizable,\n",
     "    LoadImaged,\n",
-    "    ToTensord,\n",
+    "    EnsureTyped,\n",
     ")\n",
     "\n",
     "logging.basicConfig(stream=sys.stdout, level=logging.INFO)\n",
@@ -249,7 +249,7 @@
     "train_ds = IXIDataset(\n",
     "    root_dir=root_dir,\n",
     "    section=\"training\",\n",
-    "    transform=Compose([LoadImaged(\"image\"), ToTensord(\"image\")]),\n",
+    "    transform=Compose([LoadImaged(\"image\"), EnsureTyped(\"image\")]),\n",
     "    download=True,\n",
     ")"
    ]
@@ -342,9 +342,9 @@
     "\n",
     "# Define transforms\n",
     "train_transforms = Compose([ScaleIntensity(), AddChannel(), Resize(\n",
-    "    (96, 96, 96)), RandRotate90(), ToTensor()])\n",
+    "    (96, 96, 96)), RandRotate90(), EnsureType()])\n",
     "val_transforms = Compose(\n",
-    "    [ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), ToTensor()])\n",
+    "    [ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), EnsureType()])\n",
     "\n",
     "# Define nifti dataset, data loader\n",
     "check_ds = ImageDataset(image_files=images, labels=labels,\n",

3d_classification/ignite/densenet_evaluation_array.py

@@ -22,7 +22,7 @@
 import monai
 from monai.data import ImageDataset
 from monai.handlers import CheckpointLoader, ClassificationSaver, StatsHandler
-from monai.transforms import AddChannel, Compose, Resize, ScaleIntensity, ToTensor
+from monai.transforms import AddChannel, Compose, Resize, ScaleIntensity, EnsureType
 
 
 def main():
@@ -50,7 +50,7 @@ def main():
     labels = np.array([0, 0, 1, 0, 1, 0, 1, 0, 1, 0], dtype=np.int64)
 
     # define transforms for image
-    val_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), ToTensor()])
+    val_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), EnsureType()])
     # define image dataset
     val_ds = ImageDataset(image_files=images, labels=labels, transform=val_transforms, image_only=False)
     # create DenseNet121

3d_classification/ignite/densenet_evaluation_dict.py

@@ -21,7 +21,7 @@
 
 import monai
 from monai.handlers import CheckpointLoader, ClassificationSaver, StatsHandler
-from monai.transforms import AddChanneld, Compose, LoadImaged, Resized, ScaleIntensityd, ToTensord
+from monai.transforms import AddChanneld, Compose, LoadImaged, Resized, ScaleIntensityd, EnsureTyped
 
 
 def main():
@@ -56,7 +56,7 @@ def main():
             AddChanneld(keys=["img"]),
             ScaleIntensityd(keys=["img"]),
             Resized(keys=["img"], spatial_size=(96, 96, 96)),
-            ToTensord(keys=["img"]),
+            EnsureTyped(keys=["img"]),
         ]
     )
 

3d_classification/ignite/densenet_training_array.py

@@ -23,7 +23,7 @@
 import monai
 from monai.data import ImageDataset, decollate_batch
 from monai.handlers import StatsHandler, TensorBoardStatsHandler, stopping_fn_from_metric
-from monai.transforms import AddChannel, Compose, RandRotate90, Resize, ScaleIntensity, ToTensor
+from monai.transforms import AddChannel, Compose, RandRotate90, Resize, ScaleIntensity, EnsureType
 
 
 def main():
@@ -61,8 +61,8 @@ def main():
     labels = np.array([0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 1, 0, 1, 0, 1, 0, 1, 0], dtype=np.int64)
 
     # define transforms
-    train_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), RandRotate90(), ToTensor()])
-    val_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), ToTensor()])
+    train_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), RandRotate90(), EnsureType()])
+    val_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), EnsureType()])
 
     # define image dataset, data loader
     check_ds = ImageDataset(image_files=images, labels=labels, transform=train_transforms)

3d_classification/ignite/densenet_training_dict.py

@@ -22,7 +22,7 @@
 import monai
 from monai.data import decollate_batch
 from monai.handlers import ROCAUC, StatsHandler, TensorBoardStatsHandler, stopping_fn_from_metric
-from monai.transforms import Activations, AddChanneld, AsDiscrete, Compose, LoadImaged, RandRotate90d, Resized, ScaleIntensityd, ToTensord, ToTensor
+from monai.transforms import Activations, AddChanneld, AsDiscrete, Compose, LoadImaged, RandRotate90d, Resized, ScaleIntensityd, EnsureTyped, EnsureType
 
 
 def main():
@@ -69,7 +69,7 @@ def main():
             ScaleIntensityd(keys=["img"]),
             Resized(keys=["img"], spatial_size=(96, 96, 96)),
             RandRotate90d(keys=["img"], prob=0.8, spatial_axes=[0, 2]),
-            ToTensord(keys=["img"]),
+            EnsureTyped(keys=["img"]),
         ]
     )
     val_transforms = Compose(
@@ -78,7 +78,7 @@ def main():
             AddChanneld(keys=["img"]),
             ScaleIntensityd(keys=["img"]),
             Resized(keys=["img"], spatial_size=(96, 96, 96)),
-            ToTensord(keys=["img"]),
+            EnsureTyped(keys=["img"]),
         ]
     )
 
@@ -126,8 +126,8 @@ def prepare_batch(batch, device=None, non_blocking=False):
     # add evaluation metric to the evaluator engine
     val_metrics = {metric_name: ROCAUC()}
 
-    post_label = Compose([ToTensor(), AsDiscrete(to_onehot=True, n_classes=2)])
-    post_pred = Compose([ToTensor(), Activations(softmax=True)])
+    post_label = Compose([EnsureType(), AsDiscrete(to_onehot=True, n_classes=2)])
+    post_pred = Compose([EnsureType(), Activations(softmax=True)])
     # Ignite evaluator expects batch=(img, label) and returns output=(y_pred, y) at every iteration,
     # user can add output_transform to return other values
     evaluator = create_supervised_evaluator(net, val_metrics, device, True, prepare_batch=prepare_batch, output_transform=lambda x, y, y_pred: ([post_pred(i) for i in decollate_batch(y_pred)], [post_label(i) for i in decollate_batch(y)]))

3d_classification/torch/densenet_evaluation_array.py

@@ -19,7 +19,7 @@
 
 import monai
 from monai.data import CSVSaver, ImageDataset
-from monai.transforms import AddChannel, Compose, Resize, ScaleIntensity, ToTensor
+from monai.transforms import AddChannel, Compose, Resize, ScaleIntensity, EnsureType
 
 
 def main():
@@ -47,7 +47,7 @@ def main():
     labels = np.array([0, 0, 1, 0, 1, 0, 1, 0, 1, 0], dtype=np.int64)
 
     # Define transforms for image
-    val_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), ToTensor()])
+    val_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), EnsureType()])
 
     # Define image dataset
     val_ds = ImageDataset(image_files=images, labels=labels, transform=val_transforms, image_only=False)

3d_classification/torch/densenet_evaluation_dict.py

@@ -19,7 +19,7 @@
 
 import monai
 from monai.data import CSVSaver
-from monai.transforms import AddChanneld, Compose, LoadImaged, Resized, ScaleIntensityd, ToTensord
+from monai.transforms import AddChanneld, Compose, LoadImaged, Resized, ScaleIntensityd, EnsureTyped
 
 
 def main():
@@ -54,7 +54,7 @@ def main():
             AddChanneld(keys=["img"]),
             ScaleIntensityd(keys=["img"]),
             Resized(keys=["img"], spatial_size=(96, 96, 96)),
-            ToTensord(keys=["img"]),
+            EnsureTyped(keys=["img"]),
         ]
     )
 

3d_classification/torch/densenet_training_array.py

@@ -20,7 +20,7 @@
 
 import monai
 from monai.data import ImageDataset
-from monai.transforms import AddChannel, Compose, RandRotate90, Resize, ScaleIntensity, ToTensor
+from monai.transforms import AddChannel, Compose, RandRotate90, Resize, ScaleIntensity, EnsureType
 
 
 def main():
@@ -58,8 +58,8 @@ def main():
     labels = np.array([0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 1, 0, 1, 0, 1, 0, 1, 0], dtype=np.int64)
 
     # Define transforms
-    train_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), RandRotate90(), ToTensor()])
-    val_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), ToTensor()])
+    train_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), RandRotate90(), EnsureType()])
+    val_transforms = Compose([ScaleIntensity(), AddChannel(), Resize((96, 96, 96)), EnsureType()])
 
     # Define image dataset, data loader
     check_ds = ImageDataset(image_files=images, labels=labels, transform=train_transforms)

3d_classification/torch/densenet_training_dict.py

@@ -21,7 +21,7 @@
 import monai
 from monai.data import decollate_batch
 from monai.metrics import ROCAUCMetric
-from monai.transforms import Activations, AddChanneld, AsDiscrete, Compose, LoadImaged, RandRotate90d, Resized, ScaleIntensityd, ToTensord, ToTensor
+from monai.transforms import Activations, AddChanneld, AsDiscrete, Compose, LoadImaged, RandRotate90d, Resized, ScaleIntensityd, EnsureTyped, EnsureType
 
 
 def main():
@@ -68,7 +68,7 @@ def main():
             ScaleIntensityd(keys=["img"]),
             Resized(keys=["img"], spatial_size=(96, 96, 96)),
             RandRotate90d(keys=["img"], prob=0.8, spatial_axes=[0, 2]),
-            ToTensord(keys=["img"]),
+            EnsureTyped(keys=["img"]),
         ]
     )
     val_transforms = Compose(
@@ -77,11 +77,11 @@ def main():
             AddChanneld(keys=["img"]),
             ScaleIntensityd(keys=["img"]),
             Resized(keys=["img"], spatial_size=(96, 96, 96)),
-            ToTensord(keys=["img"]),
+            EnsureTyped(keys=["img"]),
         ]
     )
-    post_pred = Compose([ToTensor(), Activations(softmax=True)])
-    post_label = Compose([ToTensor(), AsDiscrete(to_onehot=True, n_classes=2)])
+    post_pred = Compose([EnsureType(), Activations(softmax=True)])
+    post_label = Compose([EnsureType(), AsDiscrete(to_onehot=True, n_classes=2)])
 
     # Define dataset, data loader
     check_ds = monai.data.Dataset(data=train_files, transform=train_transforms)