build(env): raise the shared baseline to PyMC 6.2

environment.yml

@@ -7,35 +7,35 @@ dependencies:
   # These lines must match that spec exactly. Verify with:
   #   dse-check-env environment.yml
   - python=3.14.6
-  - numpy>=2.4.6,<2.5 # pymc -> pytensor 3.x -> numba<=0.65.1 -> numpy<2.5
+  - numpy>=2.4.6,<2.5 # pymc -> pytensor 3.x -> numba -> numpy<2.5; numba 0.66.0 still pins <2.5, so lifting this needs a numba release that supports 2.5
   - scipy>=1.18.0
-  - pandas>=3.0.3
+  - pandas>=3.0.5
   - pyarrow>=24.0.0 # pandas 3.x's default `str` dtype is Arrow-backed; floor held at 24 by jaxlib's libabseil pin — see environment-core.yml
-  - numba>=0.58,<=0.65.1 # every pytensor 3.x caps numba here (verified 3.1.3, 3.2.1, 3.2.2); this is what pymc commits us to
-  - matplotlib>=3.11.0
+  - numba>=0.58,<=0.65.1 # every pytensor 3.x caps numba here (verified 3.1.3, 3.2.1, 3.2.2, 3.2.3); this is what pymc commits us to
+  - matplotlib>=3.11.1
   - scikit-learn>=1.9.0
   - statsmodels>=0.14.6
-  - pymc>=6.1.0 # pulls pytensor + a C toolchain from conda-forge; each pymc 6.x pins one pytensor 3.x minor (6.1 -> 3.1, 6.2 -> 3.2), which caps numba <=0.65.1 -> numpy <2.5. Floor is the minimum supported, not what solves: 6.2.0/pytensor 3.2.2 is current
+  - pymc>=6.2.0 # pulls pytensor + a C toolchain from conda-forge; each pymc 6.x admits one pytensor 3.x minor (6.1 -> 3.1, 6.2 -> >=3.2.2,<3.3), which caps numba <=0.65.1 -> numpy <2.5. Floor is the minimum supported, not what solves: 6.2.0/pytensor 3.2.3 is current
   - nutpie>=0.16.11
   - numpyro>=0.21.0
   - jax>=0.10.2 # CPU build; GPU acceleration via environment-gpu.yml
   - arviz>=1.2.0
   - arviz-base>=1.2.0
   - arviz-stats>=1.2.0
   - arviz-plots>=1.2.0
-  - preliz>=0.27.0
-  - h5py>=3.12.0
-  - h5netcdf>=1.3.0
+  - preliz>=0.27.1
+  - h5py>=3.16.0
+  - h5netcdf>=1.8.1
   - graphviz>=14.1.2
-  - python-graphviz>=0.20
+  - python-graphviz>=0.21
   # ── research add-ons (compiled / columnar data / system binaries) ─────────
-  - lightgbm>=4.6.0
+  - lightgbm>=4.7.0
   - xgboost>=3.3.0
   - shap>=0.52.0
-  - polars>=1.42.1
-  - duckdb>=1.5.4
+  - polars>=1.43.1
+  - duckdb>=1.5.5
   - pyreadstat>=1.3.5
-  - xarray>=2026.4.0
+  - xarray>=2026.7.0
   - zarr
   - networkx
   - pandoc
@@ -58,5 +58,5 @@ dependencies:
       - hatch
       - build
       - pytest>=9.1.1
-      - ruff>=0.15.20
-      - setuptools>=82.0.1
+      - ruff>=0.16.1
+      - setuptools>=83.0.0

src/python/pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["hatchling>=1.30.1"]
+requires = ["hatchling>=1.31.0"]
 build-backend = "hatchling.build"
 
 [project]
@@ -14,13 +14,13 @@ dependencies = [
     "arviz>=1.2.0",
     "azure-identity",
     "azure-storage-blob",
-    "matplotlib>=3.11.0",
-    "numpy>=2.4.6,<2.5",  # numba 0.65.1 requires numpy <2.5; lift cap when numba supports 2.5
-    "pandas>=3.0.3",
-    "preliz>=0.27.0",
+    "matplotlib>=3.11.1",
+    "numpy>=2.4.6,<2.5",  # numba requires numpy <2.5 (still true of 0.66.0); lift cap when numba supports 2.5
+    "pandas>=3.0.5",
+    "preliz>=0.27.1",
     "psutil>=7.2.2",
     "pyarrow>=24.0.0",  # pandas 3.x's default `str` dtype is Arrow-backed; matches the conda core floor
-    "pyyaml>=6.0",  # dse_research_utils.environment.check (dse-check-env console script)
+    "pyyaml>=6.0.3",  # dse_research_utils.environment.check (dse-check-env console script)
     "rich>=15.0.0",
     "scikit-learn>=1.9.0",
     "scipy>=1.18.0",
@@ -33,7 +33,7 @@ dependencies = [
 # and raise a clear error when the extra is absent.
 viz = ["seaborn>=0.13"]            # dse_research_utils.plot.grids (histogram grids)
 graphs = ["graphviz", "networkx"]  # dse_research_utils.plot.graphs (needs system Graphviz)
-notebook = ["jupyter>=1.1.0", "jupytext>=1.19.4"]  # notebook/Jupyter stack (ipython drives plot.io.display_image)
+notebook = ["jupyter>=1.1.1", "jupytext>=1.19.5"]  # notebook/Jupyter stack (ipython drives plot.io.display_image)
 dependence = ["dcor"]              # dse_research_utils.ml.feature_dependence.distance_corr_matrix
 # Shared pure-Python tail for consuming projects (the pip layer of the hybrid env
 # model); the compiled scientific core lives in conda-forge (data/environment-core.yml).

src/python/src/dse_research_utils/__init__.py

@@ -1,4 +1,4 @@
 # Copyright (c) 2026 Down Syndrome Education International and contributors
 # SPDX-License-Identifier: AGPL-3.0-or-later
 
-__version__ = "0.8.0"
+__version__ = "0.9.0"

src/python/src/dse_research_utils/data/environment-core.yml

@@ -20,9 +20,14 @@ channels:
 dependencies:
   - python=3.14.6
   # --- core numerics (compiled) ---
-  - numpy>=2.4.6,<2.5 # pymc -> pytensor 3.x -> numba<=0.65.1 -> numpy<2.5
+  # numpy 2.5 is released but unreachable: pymc -> pytensor 3.x -> numba, and
+  # every numba that pytensor accepts pins numpy <2.5. Note the cap does not
+  # move when the numba cap below does — numba 0.66.0 still pins numpy <2.5 —
+  # so lifting this needs a numba release that supports 2.5, not merely a
+  # pytensor release that accepts a newer numba.
+  - numpy>=2.4.6,<2.5
   - scipy>=1.18.0
-  - pandas>=3.0.3
+  - pandas>=3.0.5
   # pandas 3.x's default `str` dtype is Arrow-backed, so PyArrow underpins every
   # string column. Declared explicitly because pandas only *constrains* pyarrow
   # (>=13.0.0) rather than depending on it — without this line its presence rests
@@ -36,23 +41,30 @@ dependencies:
   # 0.10.2) still pins libabseil <20260108 — the two do not co-solve. Raise this
   # once jaxlib is rebuilt against the newer abseil.
   - pyarrow>=24.0.0
-  - numba>=0.58,<=0.65.1 # every pytensor 3.x caps numba here (verified 3.1.3, 3.2.1, 3.2.2); this is what pymc commits us to
-  - matplotlib>=3.11.0
+  # Cap holds even though numba 0.66.0 is released: every pytensor 3.x caps
+  # numba at 0.65.1 (verified 3.1.3, 3.2.1, 3.2.2, 3.2.3), and that is what pymc
+  # commits us to.
+  - numba>=0.58,<=0.65.1
+  - matplotlib>=3.11.1
   - scikit-learn>=1.9.0
   - statsmodels>=0.14.6
   # --- Bayesian modelling stack ---
-  - pymc>=6.1.0 # pulls pytensor + a C toolchain from conda-forge; each pymc 6.x pins one pytensor 3.x minor (6.1 -> 3.1, 6.2 -> 3.2), which caps numba <=0.65.1 -> numpy <2.5. Floor is the minimum supported, not what solves: 6.2.0/pytensor 3.2.2 is current
+  # pymc pulls pytensor + a C toolchain from conda-forge. Each pymc 6.x admits
+  # one pytensor 3.x minor (6.1 -> 3.1, 6.2 -> >=3.2.2,<3.3), which caps numba
+  # <=0.65.1 -> numpy <2.5. Floor is the minimum supported, not what solves:
+  # 6.2.0/pytensor 3.2.3 is current.
+  - pymc>=6.2.0
   - nutpie>=0.16.11
   - numpyro>=0.21.0
   - jax>=0.10.2 # CPU build; GPU acceleration via environment-gpu.yml
   - arviz>=1.2.0
   - arviz-base>=1.2.0
   - arviz-stats>=1.2.0
   - arviz-plots>=1.2.0
-  - preliz>=0.27.0
+  - preliz>=0.27.1
   # --- serialization / storage (system libraries) ---
-  - h5py>=3.12.0
-  - h5netcdf>=1.3.0
+  - h5py>=3.16.0
+  - h5netcdf>=1.8.1
   # --- graph rendering (system Graphviz) ---
   - graphviz>=14.1.2
-  - python-graphviz>=0.20
+  - python-graphviz>=0.21