add device in HoVerNetNuclearTypePostProcessing and `HoVerNetInst…

…anceMapPostProcessing` (#6333)

Fixes # .

### Description

Since some operations in post-processing of HoVerNet will convert data
to numpy. And most of the time we need to calculate the metric from the
model output and label which should both be on CUDA.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: KumoLiu <yunl@nvidia.com>

monai/apps/pathology/transforms/post/array.py

@@ -31,7 +31,7 @@
 from monai.transforms.utils_pytorch_numpy_unification import max, maximum, min, sum, unique
 from monai.utils import TransformBackends, convert_to_numpy, optional_import
 from monai.utils.misc import ensure_tuple_rep
-from monai.utils.type_conversion import convert_to_dst_type
+from monai.utils.type_conversion import convert_to_dst_type, convert_to_tensor
 
 label, _ = optional_import("scipy.ndimage.measurements", name="label")
 disk, _ = optional_import("skimage.morphology", name="disk")
@@ -671,6 +671,7 @@ class HoVerNetInstanceMapPostProcessing(Transform):
         min_num_points: minimum number of points to be considered as a contour. Defaults to 3.
         contour_level: an optional value for `skimage.measure.find_contours` to find contours in the array.
             If not provided, the level is set to `(max(image) + min(image)) / 2`.
+        device: target device to put the output Tensor data.
     """
 
     def __init__(
@@ -686,9 +687,10 @@ def __init__(
         watershed_connectivity: int | None = 1,
         min_num_points: int = 3,
         contour_level: float | None = None,
+        device: str | torch.device | None = None,
     ) -> None:
         super().__init__()
-
+        self.device = device
         self.generate_watershed_mask = GenerateWatershedMask(
             activation=activation, threshold=mask_threshold, min_object_size=min_object_size
         )
@@ -742,7 +744,7 @@ def __call__(  # type: ignore
                     "centroid": instance_centroid,
                     "contour": instance_contour,
                 }
-
+        instance_map = convert_to_tensor(instance_map, device=self.device)
         return instance_info, instance_map
 
 
@@ -758,13 +760,19 @@ class HoVerNetNuclearTypePostProcessing(Transform):
         threshold: an optional float value to threshold to binarize probability map.
             If not provided, defaults to 0.5 when activation is not "softmax", otherwise None.
         return_type_map: whether to calculate and return pixel-level type map.
+        device: target device to put the output Tensor data.
 
     """
 
     def __init__(
-        self, activation: str | Callable = "softmax", threshold: float | None = None, return_type_map: bool = True
+        self,
+        activation: str | Callable = "softmax",
+        threshold: float | None = None,
+        return_type_map: bool = True,
+        device: str | torch.device | None = None,
     ) -> None:
         super().__init__()
+        self.device = device
         self.return_type_map = return_type_map
         self.generate_instance_type = GenerateInstanceType()
 
@@ -824,5 +832,6 @@ def __call__(  # type: ignore
             # update instance type map
             if type_map is not None:
                 type_map[instance_map == inst_id] = instance_type
+                type_map = convert_to_tensor(type_map, device=self.device)
 
         return instance_info, type_map

monai/apps/pathology/transforms/post/dictionary.py

@@ -14,6 +14,7 @@
 from collections.abc import Callable, Hashable, Mapping
 
 import numpy as np
+import torch
 
 from monai.apps.pathology.transforms.post.array import (
     GenerateDistanceMap,
@@ -488,6 +489,7 @@ class HoVerNetInstanceMapPostProcessingd(Transform):
         min_num_points: minimum number of points to be considered as a contour. Defaults to 3.
         contour_level: an optional value for `skimage.measure.find_contours` to find contours in the array.
             If not provided, the level is set to `(max(image) + min(image)) / 2`.
+        device: target device to put the output Tensor data.
     """
 
     def __init__(
@@ -507,6 +509,7 @@ def __init__(
         watershed_connectivity: int | None = 1,
         min_num_points: int = 3,
         contour_level: float | None = None,
+        device: str | torch.device | None = None,
     ) -> None:
         super().__init__()
         self.instance_map_post_process = HoVerNetInstanceMapPostProcessing(
@@ -521,6 +524,7 @@ def __init__(
             watershed_connectivity=watershed_connectivity,
             min_num_points=min_num_points,
             contour_level=contour_level,
+            device=device,
         )
         self.nuclear_prediction_key = nuclear_prediction_key
         self.hover_map_key = hover_map_key
@@ -553,7 +557,7 @@ class HoVerNetNuclearTypePostProcessingd(Transform):
             Defaults to `"instance_info"`.
         instance_map_key: the key where instance map is stored. Defaults to `"instance_map"`.
         type_map_key: the output key where type map is written. Defaults to `"type_map"`.
-
+        device: target device to put the output Tensor data.
 
     """
 
@@ -566,10 +570,11 @@ def __init__(
         activation: str | Callable = "softmax",
         threshold: float | None = None,
         return_type_map: bool = True,
+        device: str | torch.device | None = None,
     ) -> None:
         super().__init__()
         self.type_post_process = HoVerNetNuclearTypePostProcessing(
-            activation=activation, threshold=threshold, return_type_map=return_type_map
+            activation=activation, threshold=threshold, return_type_map=return_type_map, device=device
         )
         self.type_prediction_key = type_prediction_key
         self.instance_info_key = instance_info_key