feat: upgrade to MedCAT 2.2 and update dependencies

baixiac · baixiac · commit ea3e32a9ad3e · 2025-10-22T13:24:37.000+01:00
diff --git a/.github/workflows/main.yaml b/.github/workflows/main.yaml
@@ -24,11 +24,11 @@ jobs:
       - name: Install uv and set Python to ${{ matrix.python-version }}
         uses: astral-sh/setup-uv@v6
         with:
-          version: "0.7.20"
+          version: "0.8.10"
           python-version: ${{ matrix.python-version }}
       - name: Install dependencies
         run: |
-          uv sync --group dev --group docs --group vllm
+          uv sync --extra dev --extra docs --extra vllm
           uv run python -m ensurepip
       - name: Check types
         run: |
diff --git a/.github/workflows/release.yaml b/.github/workflows/release.yaml
@@ -17,11 +17,11 @@ jobs:
       - name: Install uv
         uses: astral-sh/setup-uv@v5
         with:
-            version: "0.6.10"
+            version: "0.8.10"
             python-version: "3.10"
       - name: Install dependencies
         run: |
-          uv sync --group dev --group docs --group vllm
+          uv sync --extra dev --extra docs --extra vllm
       - name: Run unit tests
         run: |
           uv run pytest -v tests/app --cov --cov-report=html:coverage_reports #--random-order
diff --git a/app/model_services/medcat_model.py b/app/model_services/medcat_model.py
@@ -192,7 +192,6 @@ def batch_annotate(self, texts: List[str]) -> List[List[Annotation]]:
             texts,
             batch_size_chars=batch_size_chars,
             n_process=max(int(cpu_count() / 2), 1),
-            # addl_info=["cui2icd10", "cui2ontologies", "cui2snomed", "cui2athena_ids"],
         ))}
         docs = dict(sorted(docs.items(), key=lambda x: x[0]))
         annotations_list = []
diff --git a/app/model_services/medcat_model_opcs4.py b/app/model_services/medcat_model_opcs4.py
@@ -58,11 +58,12 @@ def info(self) -> ModelCard:
             ModelCard: A card containing information about the MedCAT OPCS-4 model.
         """
 
+        assert self.model is not None, "Model is not initialised"
         return ModelCard(
             model_description=self.model_name,
             model_type=ModelType.MEDCAT_OPCS4,
             api_version=self.api_version,
-            model_card=self.model.get_model_card(as_dict=True),
+            model_card=dict(self.model.get_model_card(as_dict=True)),
         )
 
     def get_records_from_doc(self, doc: Dict) -> List[Dict]:
diff --git a/docker/Dockerfile b/docker/Dockerfile
@@ -30,11 +30,13 @@ RUN addgroup --gid $CMS_GID cms || true && \
 COPY app /app
 COPY pyproject.toml /pyproject.toml
 COPY uv.lock /uv.lock
+COPY uv-requirements.txt /uv-requirements.txt
 COPY docker/entrypoint/serve.sh /app/entrypoint.sh
 RUN mkdir -p /app/model/model && \
     mkdir -p /app/model/retrained && \
     chown -R $CMS_UID:$CMS_GID /app
-RUN pip install --no-cache-dir uv==0.7.20 && \
+RUN pip install --require-hashes -r /uv-requirements.txt --no-cache-dir && \
+    uv run --no-project python -m ensurepip && \
     uv sync --locked && \
     chown -R $CMS_UID:$CMS_GID /.venv && \
     chmod +x /app/entrypoint.sh
diff --git a/docker/huggingface-ner/requirements.txt b/docker/huggingface-ner/requirements.txt
@@ -1,20 +1,20 @@
-medcat~=1.14.1
-transformers~=4.43.0
-blis<1.0.0
-fastapi~=0.102.0
+medcat[spacy,meta-cat,deid,rel-cat]~=2.2.0
+datasets>=2.21.0
+fastapi~=0.115.0
 uvicorn~=0.29.0
-python-multipart~=0.0.5
+python-multipart~=0.0.7
 ijson~=3.1.4
 python-dotenv~=0.20.0
 mlflow~=2.16.2
 psycopg2-binary~=2.9.4
 boto3~=1.28.84
-typer~=0.7.0
-prometheus-fastapi-instrumentator~=5.11.2
+click<8.2.0
+typer~=0.15.1
+prometheus-fastapi-instrumentator~=7.0.0
 sentencepiece~=0.2.0
-slowapi~=0.1.7
+slowapi~=0.1.
 graypy~=2.1.0
-fastapi-users~=11.0.0
+fastapi-users~=13.0.0
 fastapi-users-db-sqlalchemy~=5.0.0
 asyncpg~=0.27.0
 aiosqlite~=0.19.0
@@ -23,5 +23,6 @@ websockets~=12.0
 pynvml~=11.5.3
 toml~=0.10.2
 peft<0.14.0
+huggingface-hub~=0.33.0
 setuptools
 wheel
diff --git a/docker/medcat-deid/requirements.txt b/docker/medcat-deid/requirements.txt
@@ -1,20 +1,20 @@
-medcat~=1.14.1
-transformers~=4.43.0
-blis<1.0.0
-fastapi~=0.102.0
+medcat[spacy,meta-cat,deid,rel-cat]~=2.2.0
+datasets>=2.21.0
+fastapi~=0.115.0
 uvicorn~=0.29.0
-python-multipart~=0.0.5
+python-multipart~=0.0.7
 ijson~=3.1.4
 python-dotenv~=0.20.0
 mlflow~=2.16.2
 psycopg2-binary~=2.9.4
 boto3~=1.28.84
-typer~=0.7.0
-prometheus-fastapi-instrumentator~=5.11.2
+click<8.2.0
+typer~=0.15.1
+prometheus-fastapi-instrumentator~=7.0.0
 sentencepiece~=0.2.0
-slowapi~=0.1.7
+slowapi~=0.1.
 graypy~=2.1.0
-fastapi-users~=11.0.0
+fastapi-users~=13.0.0
 fastapi-users-db-sqlalchemy~=5.0.0
 asyncpg~=0.27.0
 aiosqlite~=0.19.0
@@ -23,5 +23,6 @@ websockets~=12.0
 pynvml~=11.5.3
 toml~=0.10.2
 peft<0.14.0
+huggingface-hub~=0.33.0
 setuptools
 wheel
diff --git a/docker/medcat-icd10/requirements.txt b/docker/medcat-icd10/requirements.txt
@@ -1,20 +1,20 @@
-medcat~=1.14.1
-transformers~=4.43.0
-blis<1.0.0
-fastapi~=0.102.0
+medcat[spacy,meta-cat,deid,rel-cat]~=2.2.0
+datasets>=2.21.0
+fastapi~=0.115.0
 uvicorn~=0.29.0
-python-multipart~=0.0.5
+python-multipart~=0.0.7
 ijson~=3.1.4
 python-dotenv~=0.20.0
 mlflow~=2.16.2
 psycopg2-binary~=2.9.4
 boto3~=1.28.84
-typer~=0.7.0
-prometheus-fastapi-instrumentator~=5.11.2
+click<8.2.0
+typer~=0.15.1
+prometheus-fastapi-instrumentator~=7.0.0
 sentencepiece~=0.2.0
-slowapi~=0.1.7
+slowapi~=0.1.
 graypy~=2.1.0
-fastapi-users~=11.0.0
+fastapi-users~=13.0.0
 fastapi-users-db-sqlalchemy~=5.0.0
 asyncpg~=0.27.0
 aiosqlite~=0.19.0
@@ -23,5 +23,6 @@ websockets~=12.0
 pynvml~=11.5.3
 toml~=0.10.2
 peft<0.14.0
+huggingface-hub~=0.33.0
 setuptools
 wheel
diff --git a/docker/medcat-opcs4/requirements.txt b/docker/medcat-opcs4/requirements.txt
@@ -1,20 +1,20 @@
-medcat~=1.14.1
-transformers~=4.43.0
-blis<1.0.0
-fastapi~=0.102.0
+medcat[spacy,meta-cat,deid,rel-cat]~=2.2.0
+datasets>=2.21.0
+fastapi~=0.115.0
 uvicorn~=0.29.0
-python-multipart~=0.0.5
+python-multipart~=0.0.7
 ijson~=3.1.4
 python-dotenv~=0.20.0
 mlflow~=2.16.2
 psycopg2-binary~=2.9.4
 boto3~=1.28.84
-typer~=0.7.0
-prometheus-fastapi-instrumentator~=5.11.2
+click<8.2.0
+typer~=0.15.1
+prometheus-fastapi-instrumentator~=7.0.0
 sentencepiece~=0.2.0
-slowapi~=0.1.7
+slowapi~=0.1.
 graypy~=2.1.0
-fastapi-users~=11.0.0
+fastapi-users~=13.0.0
 fastapi-users-db-sqlalchemy~=5.0.0
 asyncpg~=0.27.0
 aiosqlite~=0.19.0
@@ -23,5 +23,6 @@ websockets~=12.0
 pynvml~=11.5.3
 toml~=0.10.2
 peft<0.14.0
+huggingface-hub~=0.33.0
 setuptools
 wheel
diff --git a/docker/medcat-snomed/requirements.txt b/docker/medcat-snomed/requirements.txt
@@ -1,20 +1,20 @@
-medcat~=1.14.1
-transformers~=4.43.0
-blis<1.0.0
-fastapi~=0.102.0
+medcat[spacy,meta-cat,deid,rel-cat]~=2.2.0
+datasets>=2.21.0
+fastapi~=0.115.0
 uvicorn~=0.29.0
-python-multipart~=0.0.5
+python-multipart~=0.0.7
 ijson~=3.1.4
 python-dotenv~=0.20.0
 mlflow~=2.16.2
 psycopg2-binary~=2.9.4
 boto3~=1.28.84
-typer~=0.7.0
-prometheus-fastapi-instrumentator~=5.11.2
+click<8.2.0
+typer~=0.15.1
+prometheus-fastapi-instrumentator~=7.0.0
 sentencepiece~=0.2.0
-slowapi~=0.1.7
+slowapi~=0.1.
 graypy~=2.1.0
-fastapi-users~=11.0.0
+fastapi-users~=13.0.0
 fastapi-users-db-sqlalchemy~=5.0.0
 asyncpg~=0.27.0
 aiosqlite~=0.19.0
@@ -23,5 +23,6 @@ websockets~=12.0
 pynvml~=11.5.3
 toml~=0.10.2
 peft<0.14.0
+huggingface-hub~=0.33.0
 setuptools
 wheel
diff --git a/docker/medcat-umls/requirements.txt b/docker/medcat-umls/requirements.txt
@@ -1,20 +1,20 @@
-medcat~=1.14.1
-transformers~=4.43.0
-blis<1.0.0
-fastapi~=0.102.0
+medcat[spacy,meta-cat,deid,rel-cat]~=2.2.0
+datasets>=2.21.0
+fastapi~=0.115.0
 uvicorn~=0.29.0
-python-multipart~=0.0.5
+python-multipart~=0.0.7
 ijson~=3.1.4
 python-dotenv~=0.20.0
 mlflow~=2.16.2
 psycopg2-binary~=2.9.4
 boto3~=1.28.84
-typer~=0.7.0
-prometheus-fastapi-instrumentator~=5.11.2
+click<8.2.0
+typer~=0.15.1
+prometheus-fastapi-instrumentator~=7.0.0
 sentencepiece~=0.2.0
-slowapi~=0.1.7
+slowapi~=0.1.
 graypy~=2.1.0
-fastapi-users~=11.0.0
+fastapi-users~=13.0.0
 fastapi-users-db-sqlalchemy~=5.0.0
 asyncpg~=0.27.0
 aiosqlite~=0.19.0
@@ -23,5 +23,6 @@ websockets~=12.0
 pynvml~=11.5.3
 toml~=0.10.2
 peft<0.14.0
+huggingface-hub~=0.33.0
 setuptools
 wheel
diff --git a/pyproject.toml b/pyproject.toml
@@ -9,7 +9,8 @@ description = "A model serving and governance system for CogStack NLP solutions"
 requires-python = ">3.9,<3.12"
 dynamic = ["version"]
 dependencies = [
-    "medcat[spacy,meta-cat,deid,rel-cat]~=2.1.0",
+    "medcat[spacy,meta-cat,deid,rel-cat]~=2.2.0; python_version>='3.10'",
+    "medcat[spacy,meta-cat,deid,rel-cat]~=2.1.0; python_version<'3.10'",
     "datasets>=2.21.0",
     "fastapi~=0.115.0",
     "uvicorn~=0.29.0",
@@ -35,7 +36,6 @@ dependencies = [
     "toml~=0.10.2",
     "peft<0.14.0",
     "huggingface-hub~=0.33.0",
-    "spacy>3.8"
 ]
 readme = "README.md"
 keywords = ["natural-language-processing", "electronic-health-records", "clinical-data"]
diff --git a/uv-requirements.in b/uv-requirements.in
@@ -0,0 +1 @@
+uv==0.8.10
diff --git a/uv-requirements.txt b/uv-requirements.txt
@@ -0,0 +1,23 @@
+# This file was autogenerated by uv via the following command:
+#    uv pip compile --generate-hashes --output-file=uv-requirements.txt --strip-extras --python=3.10 uv-requirements.in
+uv==0.8.10 \
+    --hash=sha256:31e4fc37ee94b94c032384a0957ad32ba7dce4ce6c04b4880fd3e31e25e51a82 \
+    --hash=sha256:36a5ce708d52388c37043e7335f9eb3fea5a19a56166a2cc6adb365179a1cd77 \
+    --hash=sha256:38286d230daad82388469c8dc7a1d2f5dc279c11178319c886d1a88d7938e513 \
+    --hash=sha256:3e190cee3bb2b4f574a419eef87ae8e33f713e9cd6f856b83277ece70ad9ca9b \
+    --hash=sha256:3fdf89fc40af9902141c39ed943bcfca15664623363335eb032a44f22001e2b4 \
+    --hash=sha256:4cc190d403a89e46d13cec83b6f8e8d7d07aaf1e5a996eac9a3f0c2a8cd92537 \
+    --hash=sha256:57b71dc79eff25a5419d3fe4a563d3b9397f55d789f685ef27f43f033b31f482 \
+    --hash=sha256:86fe044c2be43977566a0d184a487edd7aace2febb757fd95927684b629ef50b \
+    --hash=sha256:88df34c32555064fae459cce665757619fd1af7deb2dc393352b15d909d2d131 \
+    --hash=sha256:9ad21eeaa4156a1bf5ed85903f80db06e2c02badd3a587ba98d3171517960555 \
+    --hash=sha256:a5495b5a6e3111c03cf5e4dbdd598bc8fd1da887e3920d58cd5a2d4c8bc9a473 \
+    --hash=sha256:ab072cd3bf2f9dc264659a1ff48ad91a910ac4830bcfe965e2d3f89c86646f46 \
+    --hash=sha256:af8a5526b0e331775a264fa0dbccfd53c183cb974f269a208af136d7561f9eb2 \
+    --hash=sha256:b00637c63d5dfc9f879281c5c91db2bb909ab1f9ab275dab015e7fb6cac6be5b \
+    --hash=sha256:b3ff3c451fcd23ea78356d8c18e802d0e423cbe655273601e3ec039a51b33286 \
+    --hash=sha256:c4a493cd4b15b3aef11523531aff96a77a586666a63e842fa437966b7b7ee62d \
+    --hash=sha256:defc50bb319be2d58be74a680710cd4b7697e88d5f79974eacd354df95f0b6b0 \
+    --hash=sha256:e0a02bcec766eb0862b7082ab746b204add7d9fcaa62322502d159b5a7ccc54a \
+    --hash=sha256:eb79a46d8099f563ef58237bf4e9009f876a40145e757ea883a92b24b724d01e
+    # via -r uv-requirements.in
diff --git a/uv.lock b/uv.lock
