Support Python 3.12 via newer scipy and nmslib-metabrainz (#523)

---------

Co-authored-by: Daniel King <43149077+dakinggg@users.noreply.github.com>

.github/workflows/main.yml

@@ -20,3 +20,4 @@ jobs:
         docker run --rm scispacy black scispacy --check --line-length 88
         docker run --rm scispacy bash scripts/mypy.sh
         docker run --rm scispacy pytest tests/ --cov scispacy --cov-fail-under=20
+

.github/workflows/old_scipy.yml

@@ -0,0 +1,17 @@
+name: CI (old scipy)
+
+on:
+  pull_request:
+    branches:
+      - main
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v1
+      - name: Build and test with Docker using an older version of scipy
+        run: |
+          docker build --tag scispacy .
+          docker run --rm scispacy bash -c "pip install 'scipy<1.11' && pytest tests/"

Dockerfile

@@ -18,7 +18,7 @@ WORKDIR /work
 COPY requirements.in .
 
 RUN pip install -r requirements.in
-RUN pip install https://s3-us-west-2.amazonaws.com/ai2-s2-scispacy/releases/v0.5.3/en_core_sci_sm-0.5.3.tar.gz
+RUN pip install https://s3-us-west-2.amazonaws.com/ai2-s2-scispacy/releases/v0.5.4/en_core_sci_sm-0.5.4.tar.gz
 RUN python -m spacy download en_core_web_sm
 RUN python -m spacy download en_core_web_md
 

requirements.in

@@ -1,5 +1,8 @@
 numpy
-scipy<1.11
+# NOTE: scipy<1.11 is required when creating the linkers, so that's currently
+# only supported on Python<3.11
+# https://github.com/allenai/scispacy/issues/519#issuecomment-2229915999
+scipy
 spacy>=3.7.0,<3.8.0
 spacy-lookups-data
 pandas
@@ -8,7 +11,10 @@ conllu
 
 # Candidate generation and entity linking
 joblib
-nmslib>=1.7.3.6
+nmslib>=1.7.3.6; python_version < '3.11'
+# Use the metabrainz fork until nmslib supports installing on Python 3.11+
+# https://github.com/nmslib/nmslib/issues/555
+nmslib-metabrainz==2.1.3; python_version >= '3.11'
 scikit-learn>=0.20.3
 
 # Required for testing.

scispacy/candidate_generation.py

@@ -10,6 +10,7 @@
 import nmslib
 from nmslib.dist import FloatIndex
 
+from scispacy.util import scipy_supports_sparse_float16
 from scispacy.file_cache import cached_path
 from scispacy.linking_utils import (
     KnowledgeBase,
@@ -375,6 +376,11 @@ def create_tfidf_ann_index(
         The kb items to generate the index and vectors for.
 
     """
+    if not scipy_supports_sparse_float16():
+        raise RuntimeError(
+            "This function requires scipy<1.11, which only runs on Python<3.11."
+        )
+
     tfidf_vectorizer_path = f"{out_path}/tfidf_vectorizer.joblib"
     ann_index_path = f"{out_path}/nmslib_index.bin"
     tfidf_vectors_path = f"{out_path}/tfidf_vectors_sparse.npz"

scispacy/util.py

@@ -1,4 +1,6 @@
+from packaging.version import Version
 import spacy
+import scipy
 from spacy.language import Language
 from spacy.tokens import Doc
 
@@ -17,6 +19,11 @@ def create_combined_rule_model() -> Language:
     return nlp
 
 
+def scipy_supports_sparse_float16() -> bool:
+    # https://github.com/scipy/scipy/issues/7408
+    return Version(scipy.__version__) < Version("1.11")
+
+
 class WhitespaceTokenizer:
     """
     Spacy doesn't assume that text is tokenised. Sometimes this

setup.py

@@ -42,12 +42,13 @@
     license="Apache",
     install_requires=[
         "spacy>=3.7.0,<3.8.0",
-        "scipy<1.11",
+        "scipy",
         "requests>=2.0.0,<3.0.0",
         "conllu",
         "numpy",
         "joblib",
-        "nmslib>=1.7.3.6",
+        "nmslib>=1.7.3.6; python_version < '3.11'",
+        "nmslib-metabrainz==2.1.3; python_version >= '3.11'",
         "scikit-learn>=0.20.3",
         "pysbd",
     ],

tests/test_candidate_generation.py

@@ -3,10 +3,17 @@
 
 from scispacy.candidate_generation import CandidateGenerator, create_tfidf_ann_index, MentionCandidate
 from scispacy.umls_utils import UmlsKnowledgeBase
+from scispacy.util import scipy_supports_sparse_float16
 
 
 class TestCandidateGeneration(unittest.TestCase):
 
+    def setUp(self):
+        super().setUp()
+        if not scipy_supports_sparse_float16():
+            # https://github.com/allenai/scispacy/issues/519#issuecomment-2229915999
+            self.skipTest("Candidate generation isn't supported for scipy>=1.11")
+
     def test_create_index(self):
 
         umls_fixture = UmlsKnowledgeBase("tests/fixtures/umls_test_fixture.json")

tests/test_linking.py

@@ -7,11 +7,16 @@
 from scispacy.linking import EntityLinker
 from scispacy.umls_utils import UmlsKnowledgeBase
 from scispacy.abbreviation import AbbreviationDetector
+from scispacy.util import scipy_supports_sparse_float16
 
 
 class TestLinker(unittest.TestCase):
     def setUp(self):
         super().setUp()
+        if not scipy_supports_sparse_float16():
+            # https://github.com/allenai/scispacy/issues/519#issuecomment-2229915999
+            self.skipTest("Candidate generation isn't supported for scipy>=1.11")
+
         self.nlp = spacy.load("en_core_web_sm")
 
         umls_fixture = UmlsKnowledgeBase("tests/fixtures/umls_test_fixture.json", "tests/fixtures/test_umls_tree.tsv")