Modernize Project

.github/workflows/ci.yml

@@ -9,27 +9,17 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.9", "3.10", "3.11"]
+        python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
 
     steps:
       - name: Checkout source
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
 
-      - name: Setup Conda Environment
-        uses: conda-incubator/setup-miniconda@v2.2.0
-        with:
-          miniforge-variant: Mambaforge
-          miniforge-version: latest
-          use-mamba: true
-          python-version: ${{ matrix.python-version }}
-          activate-environment: dask-glm
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
 
       - name: Install dask_glm
-        shell: bash -l {0}
-        run: pip install -e .
+        run: uv sync --python ${{ matrix.python-version }} --extra dev
 
       - name: Run pytest
-        shell: bash -l {0}
-        run: |
-          pip install pytest pytest-xdist
-          pytest dask_glm -n auto
+        run: uv run pytest dask_glm

.github/workflows/pre-commit.yml

@@ -7,6 +7,6 @@ jobs:
     name: "pre-commit hooks"
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v2
-      - uses: actions/setup-python@v2
-      - uses: pre-commit/action@v3.0.0
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+      - uses: pre-commit/action@v3.0.1

.github/workflows/wheels.yml

@@ -1,79 +1,46 @@
-name: Build and maybe upload to PyPI
+name: Build and publish to PyPI
 
 on:
   push:
+    tags:
+      - '*'
   pull_request:
-  release:
-    types:
-      - released
-      - prereleased
+  workflow_dispatch:
 
 jobs:
-  artifacts:
-    name: Build wheels on ${{ matrix.os }}
-    runs-on: ${{ matrix.os }}
-    strategy:
-      fail-fast: false
-      matrix:
-        os: [ubuntu-latest]
-
+  build:
+    name: Build distribution
+    runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
         with:
           fetch-depth: 0
-      - uses: actions/setup-python@v4
-        with:
-          python-version: '3.10' 
-      - name: Build wheels
-        run: pip wheel . -w dist
-      - name: Build Source Dist
-        run: python setup.py sdist
-      - uses: actions/upload-artifact@v3
-        with:
-          name: wheel
-          path: ./dist/dask_glm*
-      - uses: actions/upload-artifact@v3
-        with:
-          name: sdist
-          path: ./dist/dask-glm*
 
-  list_artifacts:
-    name: List build artifacts
-    needs: [artifacts]
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/download-artifact@v3
-        with:
-          name: sdist
-          path: dist
-      - uses: actions/download-artifact@v3
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
+
+      - name: Build wheel and sdist
+        run: uv build
+
+      - uses: actions/upload-artifact@v4
         with:
-          name: wheel
-          path: dist
-      - name: test
-        run: |
-          ls
-          ls dist
+          name: dist
+          path: dist/
 
-  upload_pypi:
-    needs: [artifacts]
-    if: "startsWith(github.ref, 'refs/tags/')"
+  publish:
+    name: Publish to PyPI
+    needs: [build]
+    if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags/')
     runs-on: ubuntu-latest
     environment:
       name: releases
       url: https://pypi.org/p/dask-glm
     permissions:
       id-token: write
     steps:
-      - uses: actions/download-artifact@v3
+      - uses: actions/download-artifact@v4
         with:
-          name: sdist
-          path: dist
-      - uses: actions/download-artifact@v3
-        with:
-          name: wheel
-          path: dist
+          name: dist
+          path: dist/
+
       - uses: pypa/gh-action-pypi-publish@release/v1
-        with:
-          packages-dir: dist
-          skip-existing: true

.gitignore

@@ -2,6 +2,7 @@
 *.swpi
 *.cache
 *.coverage
+uv.lock
 
 # Distribution / packaging
 .Python

.pre-commit-config.yaml

@@ -1,25 +1,25 @@
 repos:
 -   repo: https://github.com/psf/black
-    rev: 22.10.0
+    rev: 25.1.0
     hooks:
     - id: black
       language_version: python3
 
 -   repo: https://github.com/pycqa/flake8
-    rev: 5.0.4
+    rev: 7.2.0
     hooks:
     - id: flake8
       language_version: python3
 
 -   repo: https://github.com/pycqa/isort
-    rev: 5.12.0
+    rev: 6.0.1
     hooks:
     - id: isort
       language_version: python3
       args: ["--profile", "black"]
 
 -   repo: https://github.com/codespell-project/codespell
-    rev: v2.2.4
+    rev: v2.4.1
     hooks:
     - id: codespell
       additional_dependencies:

dask_glm/__init__.py

@@ -1,6 +1,6 @@
-from pkg_resources import DistributionNotFound, get_distribution
+from importlib.metadata import PackageNotFoundError, version
 
 try:
-    __version__ = get_distribution(__name__).version
-except DistributionNotFound:
+    __version__ = version(__name__)
+except PackageNotFoundError:
     pass

dask_glm/algorithms.py

@@ -1,7 +1,4 @@
-"""Optimization algorithms for solving minimizaiton problems.
-"""
-
-from __future__ import absolute_import, division, print_function
+"""Optimization algorithms for solving minimizaiton problems."""
 
 import functools
 
@@ -233,7 +230,7 @@ def admm(
     abstol=1e-4,
     reltol=1e-2,
     family=Logistic,
-    **kwargs
+    **kwargs,
 ):
     """
     Alternating Direction Method of Multipliers
@@ -358,7 +355,7 @@ def lbfgs(
     tol=1e-4,
     family=Logistic,
     verbose=False,
-    **kwargs
+    **kwargs,
 ):
     """L-BFGS solver using scipy.optimize implementation
 
@@ -428,7 +425,7 @@ def proximal_grad(
     family=Logistic,
     max_iter=100,
     tol=1e-8,
-    **kwargs
+    **kwargs,
 ):
     """
     Proximal Gradient Method

dask_glm/estimators.py

@@ -1,6 +1,7 @@
 """
 Models following scikit-learn's estimator API.
 """
+
 from sklearn.base import BaseEstimator
 
 from . import algorithms, families

dask_glm/tests/conftest.py

@@ -0,0 +1,47 @@
+import pytest
+
+
+def pytest_addoption(parser):
+    parser.addoption(
+        "--slow", action="store_true", default=False, help="run slow tests"
+    )
+
+
+def pytest_configure(config):
+    config.addinivalue_line("markers", "slow: mark test as slow to run")
+
+
+def pytest_collection_modifyitems(config, items):
+    if config.getoption("--slow"):
+        return
+    skip_slow = pytest.mark.skip(reason="need --slow option to run")
+    for item in items:
+        # Check for @pytest.mark.slow on the function
+        if item.get_closest_marker("slow"):
+            item.add_marker(skip_slow)
+            continue
+        # Check for pytest.param(..., marks=pytest.mark.slow) in parametrize
+        if not hasattr(item, "callspec"):
+            continue
+        for marker in item.iter_markers("parametrize"):
+            param_names = [n.strip() for n in marker.args[0].split(",")]
+            for param in marker.args[1]:
+                if not hasattr(param, "marks") or not param.marks:
+                    continue
+                has_slow = any(
+                    getattr(m, "name", None) == "slow"
+                    or (hasattr(m, "mark") and m.mark.name == "slow")
+                    for m in param.marks
+                )
+                if not has_slow:
+                    continue
+                # Check if all values from this param match the callspec
+                match = True
+                for i, name in enumerate(param_names):
+                    if i < len(param.values):
+                        if item.callspec.params.get(name) != param.values[i]:
+                            match = False
+                            break
+                if match:
+                    item.add_marker(skip_slow)
+                    break

dask_glm/tests/test_admm.py

@@ -9,7 +9,7 @@
 from dask_glm.utils import make_y, to_dask_cupy_array_xy
 
 
-@pytest.mark.parametrize("N", [1000, 10000])
+@pytest.mark.parametrize("N", [1000, pytest.param(10000, marks=pytest.mark.slow)])
 @pytest.mark.parametrize(
     "beta",
     [
@@ -47,6 +47,7 @@ def wrapped(beta, X, y, z, u, rho):
     assert np.allclose(result, z, atol=2e-3)
 
 
+@pytest.mark.slow
 @pytest.mark.parametrize("N", [1000, 10000])
 @pytest.mark.parametrize("nchunks", [5, 10])
 @pytest.mark.parametrize("p", [1, 5, 10])

dask_glm/tests/test_algos_families.py

@@ -105,12 +105,12 @@ def test_basic_unreg_descent(func, kwargs, N, nchunks, family, array_type):
 @pytest.mark.parametrize(
     "func,kwargs",
     [
-        (admm, {"abstol": 1e-4}),
+        pytest.param(admm, {"abstol": 1e-4}, marks=pytest.mark.slow),
         (proximal_grad, {"tol": 1e-7}),
     ],
 )
 @pytest.mark.parametrize("N", [1000])
-@pytest.mark.parametrize("nchunks", [1, 10])
+@pytest.mark.parametrize("nchunks", [1, pytest.param(10, marks=pytest.mark.slow)])
 @pytest.mark.parametrize("family", [Logistic, Normal, Poisson])
 @pytest.mark.parametrize("lam", [0.01, 1.2, 4.05])
 @pytest.mark.parametrize("reg", [r() for r in Regularizer.__subclasses__()])
@@ -138,6 +138,7 @@ def test_basic_reg_descent(func, kwargs, N, nchunks, family, lam, reg, array_typ
     assert opt < test_val
 
 
+@pytest.mark.slow
 @pytest.mark.parametrize(
     "func,kwargs",
     [
@@ -167,6 +168,7 @@ def test_determinism(func, kwargs, scheduler, array_type):
     pass
 else:
 
+    @pytest.mark.slow
     @pytest.mark.parametrize(
         "func,kwargs",
         [
@@ -186,6 +188,7 @@ def test_determinism_distributed(func, kwargs, loop_in_thread):  # noqa: F811
 
                 assert (a == b).all()
 
+    @pytest.mark.slow
     def test_broadcast_lbfgs_weight(loop_in_thread):  # noqa: F811
         with cluster() as (s, [a, b]):
             with Client(s["address"], loop=loop_in_thread) as c:

dask_glm/tests/test_estimators.py

@@ -54,9 +54,12 @@ def _maybe_skip_sparse_error(fit_intercept, is_sparse, is_cupy):
         pytest.xfail(f"TODO: {msg}")
 
 
-@pytest.mark.parametrize("fit_intercept", [True, False])
 @pytest.mark.parametrize(
-    "is_sparse,is_cupy", [(True, False), (False, False), (False, True)]
+    "fit_intercept", [pytest.param(True, marks=pytest.mark.slow), False]
+)
+@pytest.mark.parametrize(
+    "is_sparse,is_cupy",
+    [pytest.param(True, False, marks=pytest.mark.slow), (False, False), (False, True)],
 )
 def test_fit(fit_intercept, is_sparse, is_cupy):
     _maybe_skip_sparse_error(fit_intercept, is_sparse, is_cupy)
@@ -75,9 +78,12 @@ def test_fit(fit_intercept, is_sparse, is_cupy):
     lr.predict_proba(X)
 
 
-@pytest.mark.parametrize("fit_intercept", [True, False])
 @pytest.mark.parametrize(
-    "is_sparse,is_cupy", [(True, False), (False, False), (False, True)]
+    "fit_intercept", [pytest.param(True, marks=pytest.mark.slow), False]
+)
+@pytest.mark.parametrize(
+    "is_sparse,is_cupy",
+    [pytest.param(True, False, marks=pytest.mark.slow), (False, False), (False, True)],
 )
 def test_lm(fit_intercept, is_sparse, is_cupy):
     _maybe_skip_sparse_error(fit_intercept, is_sparse, is_cupy)
@@ -95,6 +101,7 @@ def test_lm(fit_intercept, is_sparse, is_cupy):
         assert lr.intercept_ is not None
 
 
+@pytest.mark.slow
 @pytest.mark.parametrize("fit_intercept", [True, False])
 @pytest.mark.parametrize(
     "is_sparse,is_cupy", [(True, False), (False, False), (False, True)]
@@ -115,6 +122,7 @@ def test_big(fit_intercept, is_sparse, is_cupy):
         assert lr.intercept_ is not None
 
 
+@pytest.mark.slow
 @pytest.mark.parametrize("fit_intercept", [True, False])
 @pytest.mark.parametrize(
     "is_sparse,is_cupy", [(True, False), (False, False), (False, True)]