Rc4 fixes (#879)

* minor fixes for rc4 dependency

Signed-off-by: Sachidanand Alle <[email protected]>

* add integration tests for pathology app/models

Signed-off-by: Sachidanand Alle <[email protected]>

monailabel/interfaces/tasks/infer.py

@@ -378,7 +378,7 @@ def _get_network(self, device):
                     model_state_dict = checkpoint.get(self.model_state_dict, checkpoint)
                     network.load_state_dict(model_state_dict, strict=self.load_strict)
             else:
-                network = torch.jit.load(path, map_location=torch.device(device)).to(torch.device)
+                network = torch.jit.load(path, map_location=torch.device(device))
 
             network.eval()
             self._networks[device] = (network, statbuf.st_mtime if statbuf else 0)

monailabel/scribbles/infer.py

@@ -9,15 +9,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from monai.transforms import (
-    Compose,
-    EnsureChannelFirstd,
-    FromMetaTensord,
-    LoadImaged,
-    ScaleIntensityRanged,
-    Spacingd,
-    ToMetaTensord,
-)
+from monai.transforms import Compose, EnsureChannelFirstd, LoadImaged, ScaleIntensityRanged, Spacingd
 
 from monailabel.interfaces.tasks.infer import InferTask, InferType
 from monailabel.scribbles.transforms import (
@@ -72,13 +64,11 @@ def pre_transforms(self, data):
         return [
             LoadImaged(keys=["image", "label"]),
             EnsureChannelFirstd(keys=["image", "label"]),
-            FromMetaTensord(keys=["image", "label"]),
             AddBackgroundScribblesFromROId(
                 scribbles="label",
                 scribbles_bg_label=self.scribbles_bg_label,
                 scribbles_fg_label=self.scribbles_fg_label,
             ),
-            ToMetaTensord(keys=["image", "label"]),
             Spacingd(keys=["image", "label"], pixdim=self.pix_dim, mode=["bilinear", "nearest"]),
             ScaleIntensityRanged(
                 keys="image",
@@ -112,7 +102,6 @@ def post_transforms(self, data):
                 lamda=self.lamda,
                 sigma=self.sigma,
             ),
-            FromMetaTensord(keys=["image"]),
             Restored(keys="pred", ref_image="image"),
             BoundingBoxd(keys="pred", result="result", bbox="bbox"),
         ]

monailabel/scribbles/transforms.py

@@ -14,6 +14,7 @@
 
 import numpy as np
 import torch
+from monai.data import MetaTensor
 from monai.networks.blocks import CRF
 from monai.transforms import Transform
 from scipy.special import softmax
@@ -39,7 +40,7 @@ def _fetch_data(self, data, key):
         if key not in data.keys():
             raise ValueError(f"Key {key} not found, present keys {data.keys()}")
 
-        return data[key].numpy() if isinstance(data[key], torch.Tensor) else data[key]
+        return data[key].array if isinstance(data[key], MetaTensor) else data[key]
 
     def _normalise_logits(self, data, axis=0):
         # check if logits is a true prob, if not then apply softmax

monailabel/transform/post.py

@@ -15,6 +15,7 @@
 import numpy as np
 import skimage.measure as measure
 from monai.config import KeysCollection
+from monai.data import MetaTensor
 from monai.transforms import MapTransform, Resize, generate_spatial_bounding_box, get_extreme_points
 from monai.utils import InterpolateMode, ensure_tuple_rep
 
@@ -102,11 +103,16 @@ def __init__(
 
     def __call__(self, data):
         d = dict(data)
-        meta_dict = d[f"{self.ref_image}_{self.meta_key_postfix}"]
+        meta_dict = (
+            d[self.ref_image].meta
+            if d.get(self.ref_image) is not None and isinstance(d[self.ref_image], MetaTensor)
+            else d.get(f"{self.ref_image}_{self.meta_key_postfix}", {})
+        )
+
         for idx, key in enumerate(self.keys):
             result = d[key]
             current_size = result.shape[1:] if self.has_channel else result.shape
-            spatial_shape = meta_dict["spatial_shape"]
+            spatial_shape = meta_dict.get("spatial_shape", current_size)
             spatial_size = spatial_shape[-len(current_size) :]
 
             # Undo Spacing

monailabel/transform/writer.py

@@ -16,7 +16,7 @@
 import nrrd
 import numpy as np
 import torch
-from monai.data import write_nifti
+from monai.data import MetaTensor, write_nifti
 
 from monailabel.utils.others.generic import file_ext
 from monailabel.utils.others.pathology import create_asap_annotations_xml, create_dsa_annotations_json
@@ -185,13 +185,17 @@ def __call__(self, data):
         ext = ext if ext else ".nii.gz"
         logger.info(f"Result ext: {ext}; write_to_file: {write_to_file}; dtype: {dtype}")
 
-        image_np = data[self.label]
-        if isinstance(image_np, torch.Tensor):
-            image_np = image_np.numpy()
+        if isinstance(data[self.label], MetaTensor):
+            image_np = data[self.label].array
+        else:
+            image_np = data[self.label]
+
+        # Always using Restored as the last transform before writing
         meta_dict = data.get(f"{self.ref_image}_{self.meta_key_postfix}")
         affine = meta_dict.get("affine") if meta_dict else None
-        if isinstance(affine, torch.Tensor):
-            affine = affine.numpy()
+        if affine is None and isinstance(data[self.ref_image], MetaTensor):
+            affine = data[self.ref_image].affine
+
         logger.debug(f"Image: {image_np.shape}; Data Image: {data[self.label].shape}")
 
         output_file = None

requirements.txt

@@ -1,5 +1,5 @@
 torch>=1.7
-monai[nibabel, skimage, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, openslide]>=0.9.1rc3
+monai[nibabel, skimage, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, openslide]>=0.9.1rc4
 uvicorn==0.17.6
 pydantic==1.9.1
 python-dotenv==0.20.0

runtests.sh

@@ -426,20 +426,20 @@ function check_server_running() {
   echo ${code}
 }
 
-# network training/inference/eval integration tests
-if [ $doNetTests = true ]; then
+
+function run_integration_tests() {
   echo "${separator}${blue}integration${noColor}"
   torch_validate
 
   ${cmdPrefix}${PY_EXE} tests/setup.py
-  echo "Starting MONAILabel server..."
+  echo "$1 - Starting MONAILabel server..."
   rm -rf tests/data/apps
-  monailabel apps -n radiology -o tests/data/apps -d
-  monailabel start_server -a tests/data/apps/radiology -c models all -s tests/data/dataset/local/spleen -p ${MONAILABEL_SERVER_PORT:-8000} &
+  monailabel apps -n $1 -o tests/data/apps -d
+  monailabel start_server -a tests/data/apps/$1 -c models all -s $2 -p ${MONAILABEL_SERVER_PORT:-8000} &
 
   wait_time=0
   server_is_up=0
-  start_time_out=180
+  start_time_out=240
 
   while [[ $wait_time -le ${start_time_out} ]]; do
     if [ "$(check_server_running)" == "200" ]; then
@@ -448,18 +448,24 @@ if [ $doNetTests = true ]; then
     fi
     sleep 5
     wait_time=$((wait_time + 5))
-    echo "Waiting for MONAILabel to be up and running..."
+    echo "$1 - Waiting for MONAILabel to be up and running..."
   done
   echo ""
 
   if [ "$server_is_up" == "1" ]; then
-    echo "MONAILabel server is up and running."
+    echo "$1 - MONAILabel server is up and running."
   else
-    echo "Failed to start MONAILabel server. Exiting..."
+    echo "$1 - Failed to start MONAILabel server. Exiting..."
     exit 1
   fi
 
-  ${cmdPrefix}${cmd} -m pytest -v tests/integration --no-summary -x
-  echo "Finished All Integration Tests;  Stop/Kill MONAILabel Server..."
+  ${cmdPrefix}${cmd} -m pytest -v tests/integration/$1 --no-summary -x
+  echo "$1 - Finished All Integration Tests;  Stop/Kill MONAILabel Server..."
   kill -9 $(ps -ef | grep monailabel | grep start_server | grep -v grep | awk '{print $2}')
+}
+
+# network training/inference/eval integration tests
+if [ $doNetTests = true ]; then
+  run_integration_tests "radiology" "tests/data/dataset/local/spleen"
+  run_integration_tests "pathology" "tests/data/pathology"
 fi

sample-apps/endoscopy/lib/net/__init__.py

@@ -0,0 +1,10 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.

sample-apps/radiology/lib/activelearning/__init__.py

@@ -0,0 +1,12 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from .first import First

sample-apps/radiology/lib/infers/deepedit.py

@@ -22,13 +22,11 @@
     AsDiscreted,
     EnsureChannelFirstd,
     EnsureTyped,
-    FromMetaTensord,
     LoadImaged,
     Orientationd,
     Resized,
     ScaleIntensityRanged,
     SqueezeDimd,
-    ToMetaTensord,
     ToNumpyd,
 )
 
@@ -85,22 +83,18 @@ def pre_transforms(self, data=None):
                     AddGuidanceFromPointsDeepEditd(ref_image="image", guidance="guidance", label_names=self.labels),
                     Resized(keys="image", spatial_size=self.spatial_size, mode="area"),
                     ResizeGuidanceMultipleLabelDeepEditd(guidance="guidance", ref_image="image"),
-                    FromMetaTensord(keys="image"),
                     AddGuidanceSignalDeepEditd(
                         keys="image", guidance="guidance", number_intensity_ch=self.number_intensity_ch
                     ),
-                    ToMetaTensord(keys="image"),
                 ]
             )
         else:
             t.extend(
                 [
                     Resized(keys="image", spatial_size=self.spatial_size, mode="area"),
-                    FromMetaTensord(keys="image"),
                     DiscardAddGuidanced(
                         keys="image", label_names=self.labels, number_intensity_ch=self.number_intensity_ch
                     ),
-                    ToMetaTensord(keys="image"),
                 ]
             )
 
@@ -120,6 +114,5 @@ def post_transforms(self, data=None) -> Sequence[Callable]:
             AsDiscreted(keys="pred", argmax=True),
             SqueezeDimd(keys="pred", dim=0),
             ToNumpyd(keys="pred"),
-            FromMetaTensord(keys="image"),
             Restored(keys="pred", ref_image="image"),
         ]

sample-apps/radiology/main.py

@@ -15,7 +15,7 @@
 from typing import Dict
 
 import lib.configs
-from lib.activelearning.first import First
+from lib.activelearning import First
 
 from monailabel.interfaces.app import MONAILabelApp
 from monailabel.interfaces.config import TaskConfig

setup.cfg

@@ -25,7 +25,7 @@ setup_requires =
     ninja
 install_requires =
     torch>=1.7
-    monai[nibabel, skimage, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, openslide]>=0.9.1rc3
+    monai[nibabel, skimage, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, openslide]>=0.9.1rc4
     uvicorn==0.17.6
     pydantic==1.9.1
     python-dotenv==0.20.0

tests/integration/pathology/__init__.py

@@ -0,0 +1,10 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.

tests/integration/pathology/test_infer.py

@@ -0,0 +1,79 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import unittest
+
+import requests
+import torch
+
+from tests.integration import SERVER_URI
+
+
+class EndPointInfer(unittest.TestCase):
+    def test_deepedit_nuclei(self):
+        if not torch.cuda.is_available():
+            return
+
+        model = "deepedit_nuclei"
+        image = "JP2K-33003-1"
+        body = {
+            "level": 0,
+            "location": [2206, 4925],
+            "size": [360, 292],
+            "tile_size": [2048, 2048],
+            "min_poly_area": 30,
+            "params": {"foreground": [], "background": []},
+        }
+
+        response = requests.post(f"{SERVER_URI}/infer/wsi/{model}?image={image}&output=dsa", json=body)
+        assert response.status_code == 200
+
+    def test_segmentation_nuclei(self):
+        if not torch.cuda.is_available():
+            return
+
+        model = "segmentation_nuclei"
+        image = "JP2K-33003-1"
+        body = {
+            "level": 0,
+            "location": [2206, 4925],
+            "size": [360, 292],
+            "tile_size": [2048, 2048],
+            "min_poly_area": 30,
+            "params": {"foreground": [], "background": []},
+        }
+
+        response = requests.post(f"{SERVER_URI}/infer/wsi/{model}?image={image}&output=asap", json=body)
+        assert response.status_code == 200
+
+    def test_nuclick(self):
+        if not torch.cuda.is_available():
+            return
+
+        model = "nuclick"
+        image = "JP2K-33003-1"
+        body = {
+            "level": 0,
+            "location": [2206, 4925],
+            "size": [360, 292],
+            "tile_size": [2048, 2048],
+            "min_poly_area": 30,
+            "params": {
+                "foreground": [[2427, 4976], [2341, 5033], [2322, 5207], [2305, 5212], [2268, 5182]],
+                "background": [],
+            },
+        }
+
+        response = requests.post(f"{SERVER_URI}/infer/wsi/{model}?image={image}&output=asap", json=body)
+        assert response.status_code == 200
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/integration/test_info.py → tests/integration/pathology/test_info.py

@@ -13,7 +13,7 @@
 
 import requests
 
-from . import SERVER_URI
+from tests.integration import SERVER_URI
 
 
 class EndPointInfo(unittest.TestCase):

tests/integration/radiology/__init__.py

@@ -0,0 +1,10 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.

tests/integration/test_infer.py → tests/integration/radiology/test_infer.py

@@ -14,7 +14,7 @@
 import requests
 import torch
 
-from . import SERVER_URI
+from tests.integration import SERVER_URI
 
 
 class EndPointInfer(unittest.TestCase):