Merge branch 'main' into sw_fastedit

.github/workflows/blossom-ci.yml

@@ -39,14 +39,16 @@ jobs:
       args: ${{ env.args }}
 
     # This job only runs for pull request comments
-    if: contains('\
-      Nic-Ma,\
-      SachidanandAlle,\
-      diazandr3s,\
-      tangy5,\
-      wyli,\
-      YanxuanLiu,\
-      ', format('{0},', github.actor)) && github.event.comment.body == '/build'
+    if: |
+      github.event.comment.body == '/build' &&
+      (
+        github.actor == 'Nic-Ma' ||
+        github.actor == 'SachidanandAlle' ||
+        github.actor == 'diazandr3s' ||
+        github.actor == 'tangy5' ||
+        github.actor == 'wyli' ||
+        github.actor == 'YanxuanLiu'
+      )
     steps:
       - name: Check if comment is issued by authorized person
         run: blossom-ci

Dockerfile

@@ -13,7 +13,7 @@
 # please run `./runtests.sh --clean && DOCKER_BUILDKIT=1 docker build -t projectmonai/monailabel:latest .`
 # to use different version of MONAI pass `--build-arg MONAI_IMAGE=...`
 
-ARG MONAI_IMAGE=projectmonai/monai:1.3.0
+ARG MONAI_IMAGE=projectmonai/monai:1.3.1
 ARG NODE_IMAGE=node:slim
 
 FROM ${NODE_IMAGE} as ohifbuild

monailabel/datastore/dsa.py

@@ -84,7 +84,7 @@ def get_label_by_image_id(self, image_id: str, tag: str) -> str:
     def get_annotations_by_image_id(self, image_id: str) -> Dict[str, Dict[str, List]]:
         image_id, name = self._name_to_id(image_id)
 
-        data = self.gc.get("annotation", parameters={"limit": 0})
+        data = self.gc.get(f"annotation/item/{image_id}", parameters={"limit": 0})
         result: Dict[str, Dict[str, List]] = {}
 
         # TODO(avirodov): probably can request only annotation for a given image_id, need to check how.
@@ -136,26 +136,40 @@ def get_image(self, image_id: str, params=None) -> Any:
     def _name_to_id(self, name):
         folders = self.folders if self.folders else self._get_all_folders()
         for folder in folders:
+            # First check if the name is directly present
+            data = self.gc.get("item", parameters={"folderId": folder, "name": name, "limit": 0})
+            for d in data:
+                if d.get("largeImage"):
+                    return d["_id"], d["name"]
+            # next check if the name is present in a stem form
             data = self.gc.get("item", parameters={"folderId": folder, "limit": 0})
             for d in data:
                 if d.get("largeImage") and d["name"] == name or Path(d["name"]).stem == name:
                     return d["_id"], d["name"]
-        return name
+        # Next check if the name is anywhere in the system
+        data = self.gc.get("item", parameters={"text": f'"{name}"' if '"' not in name else name, "limit": 0})
+        for d in data:
+            if d.get("largeImage") and d["name"] == name or Path(d["name"]).stem == name:
+                return d["_id"], d["name"]
+        # If we fail to find the item, the best we can do is return the name
+        return name, name
 
     def get_image_uri(self, image_id: str) -> str:
         try:
-            name = self.get_image_info(image_id)["name"]
+            info = self.get_image_info(image_id)
+            name = info["name"]
+            file_id = info.get("largeImage", {}).get("fileId")
         except girder_client.HttpError:
             image_id, name = self._name_to_id(image_id)
+            file_id = None
 
         if self.asset_store_path:
-            data = self.gc.get(f"item/{image_id}/files", parameters={"limit": 0})
-            assets = [d["assetstoreId"] for d in data]
-            for asset in assets:
-                files = self.gc.get(f"assetstore/{asset}/files", parameters={"limit": 0})
-                for f in files:
-                    if f["itemId"] == image_id:
-                        return str(os.path.join(self.asset_store_path, f["path"]))
+            if file_id is None:
+                data = self.gc.get(f"item/{image_id}/files", parameters={"limit": 0})
+                file_id = data[0]["_id"]
+            f = self.gc.get(f"resource/{file_id}?type=file")
+            if "path" in f and os.path.exists(os.path.join(self.asset_store_path, f["path"])):
+                return str(os.path.join(self.asset_store_path, f["path"]))
         else:
             cached = os.path.join(self.cache_path, name)
             if os.path.exists(cached):
@@ -243,9 +257,14 @@ def get_dataset_archive(self, limit_cases: Optional[int]) -> str:
         raise NotImplementedError
 
     def status(self) -> Dict[str, Any]:
+        # This is a very costly query, disable it for now
+        # return {
+        #     "total": len(self.list_images()),
+        #     "completed": len(self.get_labeled_images()),
+        # }
         return {
-            "total": len(self.list_images()),
-            "completed": len(self.get_labeled_images()),
+            "total": 0,
+            "completed": 0,
         }
 
     def json(self):

monailabel/endpoints/user/auth.py

@@ -11,11 +11,12 @@
 import logging
 from typing import List, Sequence, Union
 
+import jwt
 import requests
 from cachetools import cached
 from fastapi import Depends, HTTPException, Security, status
 from fastapi.security import OAuth2PasswordBearer
-from jose import JWTError, jwt
+from jwt import InvalidTokenError
 from passlib.context import CryptContext
 from pydantic import BaseModel
 
@@ -114,7 +115,7 @@ async def get_current_user(token: str = Depends(oauth2_scheme) if settings.MONAI
     )
     try:
         return from_token(token)
-    except JWTError as e:
+    except InvalidTokenError as e:
         logger.error(e)
         raise credentials_exception
 

requirements.txt

@@ -9,7 +9,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-monai[nibabel, skimage, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, openslide, fire, mlflow]>=1.3.0
+monai[nibabel, skimage, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, openslide, fire, mlflow]>=1.3.1
 uvicorn==0.29.0
 pydantic==2.7.0
 pydantic-settings==2.2.1
@@ -35,7 +35,7 @@ einops==0.7.0
 pyyaml==6.0.1
 filelock==3.11.0
 passlib==1.7.4
-python-jose[cryptography]==3.3.0
+pyjwt==2.8.0
 bcrypt==4.1.2
 shapely==2.0.4
 requests==2.31.0

sample-apps/endoscopy/lib/configs/inbody.py

@@ -35,7 +35,7 @@ def init(self, name: str, model_dir: str, conf: Dict[str, str], planner: Any, **
         super().init(name, model_dir, conf, planner, **kwargs)
 
         bundle_name = "endoscopic_inbody_classification"
-        version = conf.get("inbody", "0.4.4")
+        version = conf.get("inbody", "0.4.8")
         zoo_source = conf.get("zoo_source", settings.MONAI_ZOO_SOURCE)
 
         self.bundle_path = os.path.join(self.model_dir, bundle_name)

sample-apps/endoscopy/lib/configs/tooltracking.py

@@ -35,7 +35,7 @@ def init(self, name: str, model_dir: str, conf: Dict[str, str], planner: Any, **
         super().init(name, model_dir, conf, planner, **kwargs)
 
         bundle_name = "endoscopic_tool_segmentation"
-        version = conf.get("tooltracking", "0.5.5")
+        version = conf.get("tooltracking", "0.5.9")
         zoo_source = conf.get("zoo_source", settings.MONAI_ZOO_SOURCE)
 
         self.bundle_path = os.path.join(self.model_dir, bundle_name)

sample-apps/endoscopy/update_cvat_model.sh

@@ -38,13 +38,13 @@ fi
 if [ $FUNC_NAME == "deepedit" ];then
     MODEL_PATH="$APP_ROOT/model/$FUNC_NAME.pt"
     # Replace prior pretrained model with lastest model as current pre-trained model
-    MODEL_CONTAINER="/opt/conda/monailabel/sample-apps/endoscopy/model/pretrained_$FUNC_NAME.pt" # default model path at function container
+    MODEL_CONTAINER="/usr/local/monailabel/sample-apps/endoscopy/model/pretrained_$FUNC_NAME.pt" # default model path at function container
 else
     # if bundle is used, get bundle name and fetch the model
     BUNDLE_NAME=${BUNDLENAMES[$FUNC_NAME]}
     MODEL_PATH="$APP_ROOT/model/$BUNDLE_NAME/models/model.pt"
     # Update to bundle nuclio container
-    MODEL_CONTAINER="/opt/conda/monailabel/sample-apps/endoscopy/model/$BUNDLE_NAME/models/model.pt" # default model path at function container
+    MODEL_CONTAINER="/usr/local/monailabel/sample-apps/endoscopy/model/$BUNDLE_NAME/models/model.pt" # default model path at function container
 fi
 
 # Check if latest  model checkpoint is done and saved.

sample-apps/pathology/lib/configs/classification_nuclei.py

@@ -31,7 +31,7 @@ def init(self, name: str, model_dir: str, conf: Dict[str, str], planner: Any, **
 
         bundle_name = "pathology_nuclei_classification"
         zoo_source = conf.get("zoo_source", settings.MONAI_ZOO_SOURCE)
-        version = conf.get("classification_nuclei", "0.1.4")
+        version = conf.get("classification_nuclei", "0.1.7")
 
         self.bundle_path = os.path.join(self.model_dir, bundle_name)
         if not os.path.exists(self.bundle_path):

sample-apps/pathology/lib/configs/hovernet_nuclei.py

@@ -31,7 +31,7 @@ def init(self, name: str, model_dir: str, conf: Dict[str, str], planner: Any, **
 
         bundle_name = "pathology_nuclei_segmentation_classification"
         zoo_source = conf.get("zoo_source", settings.MONAI_ZOO_SOURCE)
-        version = conf.get("hovernet_nuclei", "0.2.1")
+        version = conf.get("hovernet_nuclei", "0.2.4")
 
         self.bundle_path = os.path.join(self.model_dir, bundle_name)
         if not os.path.exists(self.bundle_path):

sample-apps/pathology/lib/configs/nuclick.py

@@ -31,7 +31,7 @@ def init(self, name: str, model_dir: str, conf: Dict[str, str], planner: Any, **
 
         bundle_name = "pathology_nuclick_annotation"
         zoo_source = conf.get("zoo_source", settings.MONAI_ZOO_SOURCE)
-        version = conf.get("nuclick", "0.1.4")
+        version = conf.get("nuclick", "0.1.9")
 
         self.bundle_path = os.path.join(self.model_dir, bundle_name)
         if not os.path.exists(self.bundle_path):

setup.cfg

@@ -35,7 +35,7 @@ setup_requires =
     torch
     ninja
 install_requires =
-    monai[nibabel, skimage, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, openslide, fire, mlflow]>=1.3.0
+    monai[nibabel, skimage, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, openslide, fire, mlflow]>=1.3.1
     uvicorn==0.29.0
     pydantic==2.7.0
     pydantic-settings==2.2.1
@@ -61,7 +61,7 @@ install_requires =
     pyyaml==6.0.1
     filelock==3.11.0
     passlib==1.7.4
-    python-jose[cryptography]==3.3.0
+    pyjwt==2.8.0
     bcrypt==4.1.2
     shapely==2.0.4
     requests==2.31.0