ci(datasets): Migrate Flower Datasets to uv

.github/workflows/build-deploy-non-framework-docs.yml

@@ -32,8 +32,8 @@ jobs:
       - name: Install docs dependencies
         run: |
           cd framework
-          python -m poetry add ../datasets
           python -m poetry install
+          python -m pip install -e ../datasets
       - name: Update HTML theme options
         run: python dev/update-html-themes.py
       - name: Build baselines docs

.github/workflows/datasets-test.yml

@@ -20,6 +20,8 @@ concurrency:
 
 env:
   FLWR_TELEMETRY_ENABLED: 0
+  UV_NO_MANAGED_PYTHON: 1
+  UV_PYTHON_DOWNLOADS: never
 
 jobs:
   test_core:
@@ -42,6 +44,11 @@ jobs:
         uses: ./.github/actions/bootstrap
         with:
           python-version: ${{ matrix.python }}
+          poetry-skip: "true"
+      - name: Set up uv
+        uses: astral-sh/setup-uv@v7
+        with:
+          version: "0.10.2"
       - name: Free Disk Space (Ubuntu)
         uses: jlumbroso/free-disk-space@v1.3.1
         with:
@@ -56,10 +63,10 @@ jobs:
         run: |
           sudo apt-get update
           sudo apt-get install -y ffmpeg
-      - name: Install dependencies (mandatory only)
+      - name: Install dependencies
         run: |
           cd datasets
-          python -m poetry install --all-extras
+          uv sync --frozen --all-extras
       - name: Cache Hugging Face datasets
         uses: actions/cache@v3
         with:
@@ -68,11 +75,14 @@ jobs:
           restore-keys: hf-datasets-
       - name: Set Hugging Face token
         run: |
+          cd datasets
           if [ -n "${{ secrets.HF_TOKEN }}" ]; then
             echo "Logging into Hugging Face..."
-            hf auth login --token ${{ secrets.HF_TOKEN }}
+            uv run hf auth login --token ${{ secrets.HF_TOKEN }}
           else
             echo "Skipping Hugging Face login stage (HF_TOKEN not set)"
           fi
       - name: Test (formatting + unit tests)
-        run: ./datasets/dev/test.sh
+        run: |
+          cd datasets
+          uv run ./dev/test.sh

datasets/dev/build.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+
+# Copyright 2026 Flower Labs GmbH. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+
+set -e
+cd "$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"/../
+
+uv build --clear

datasets/dev/publish.sh

@@ -18,4 +18,4 @@
 set -e
 cd "$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"/../
 
-python -m poetry publish -u __token__ -p ${PYPI_TOKEN}
+uv publish --token "${PYPI_TOKEN}"

datasets/docs/source/conf.py

@@ -46,7 +46,7 @@
 _pyproject_path = Path(__file__).parent.parent.parent / "pyproject.toml"
 with open(_pyproject_path, "rb") as f:
     _pyproject = tomllib.load(f)
-release = _pyproject["tool"]["poetry"]["version"]
+release = _pyproject["project"]["version"]
 
 # Make version available as a substitution in rst files (e.g., |release|)
 rst_prolog = f"""

datasets/docs/source/contributor-how-to-develop-flwr-datasets.rst

@@ -0,0 +1,71 @@
+How to develop flwr-datasets
+============================
+
+Flower Datasets uses ``uv`` for development and CI. The commands presented in this guide assume you have cloned the `Flower Github repository <https://github.com/adap/flower>`_.
+
+
+Setup
+-----
+
+.. code-block:: bash
+
+   cd datasets
+   uv sync --all-extras
+
+.. tip::
+
+   Use ``uv sync --frozen --all-extras`` to ensure ``uv.lock`` is not modified.
+
+Run checks (formatting + unit tests)
+------------------------------------
+
+.. code-block:: bash
+
+   cd datasets
+   uv run ./dev/test.sh
+
+Format
+------
+
+.. code-block:: bash
+
+   cd datasets
+   uv run ./dev/format.sh
+
+Build docs
+----------
+
+.. code-block:: bash
+
+   cd datasets
+   uv run ./dev/build-flwr-datasets-docs.sh
+
+Run E2E tests
+-------------
+
+.. code-block:: bash
+
+   cd datasets/e2e/pytorch
+   uv sync --frozen
+   uv run python -m unittest discover -p "*_test.py"
+
+Repeat for ``datasets/e2e/scikit-learn`` and ``datasets/e2e/tensorflow``.
+
+Dependency management (no ``uv pip``)
+-------------------------------------
+
+.. code-block:: bash
+
+   cd datasets
+
+   # Add a runtime dependency
+   uv add <package>
+
+   # Add a dev dependency
+   uv add --dev <package>
+
+   # Add a dependency to an extra (e.g. "vision")
+   uv add --optional vision <package>
+
+   # Update lockfile (commit the result)
+   uv lock

datasets/docs/source/index.rst

@@ -70,6 +70,7 @@ Information-oriented API reference and other reference material.
    :maxdepth: 1
    :caption: Contributor tutorials
 
+   contributor-how-to-develop-flwr-datasets
    contributor-how-to-contribute-dataset
 
 

datasets/pyproject.toml

@@ -1,17 +1,15 @@
 [build-system]
-requires = ["poetry-core>=2.1.3"]
-build-backend = "poetry.core.masonry.api"
+requires = ["hatchling"]
+build-backend = "hatchling.build"
 
-[tool.poetry]
+[project]
 name = "flwr-datasets"
 version = "0.6.0"
 description = "Flower Datasets"
 license = "Apache-2.0"
-authors = ["The Flower Authors <hello@flower.ai>"]
+authors = [{ name = "The Flower Authors", email = "hello@flower.ai" }]
 readme = "README.md"
-homepage = "https://flower.ai"
-repository = "https://github.com/adap/flower"
-documentation = "https://flower.ai/docs/datasets"
+requires-python = ">=3.10"
 keywords = [
     "Artificial Intelligence",
     "Dataset",
@@ -46,48 +44,59 @@ classifiers = [
     "Topic :: Software Development :: Libraries :: Python Modules",
     "Typing :: Typed",
 ]
-packages = [{ include = "flwr_datasets", from = "./" }]
-exclude = ["./**/*_test.py"]
+dependencies = [
+    "numpy>=1.26.0,<3.0.0",
+    "datasets>=4.0.0,<5.0.0",
+    "tqdm>=4.66.1,<5.0.0",
+    "rich>=13.5.0",
+    "matplotlib>=3.7.5,<4.0.0",
+    "seaborn>=0.13.0,<0.14.0",
+]
 
-[tool.poetry.scripts]
+[project.optional-dependencies]
+vision = ["pillow>=12.1.1"]
+audio = ["torch>=2.8.0", "torchcodec>=0.7.0"]
+
+[project.scripts]
 # `flwr-datasets` CLI
 flwr-datasets = "flwr_datasets.cli.app:app"
 
-[tool.poetry.dependencies]
-python = "^3.10"
-# Mandatory dependencies
-numpy = ">=1.26.0,<3.0.0"
-datasets = ">=4.0.0,<5.0.0"
-tqdm = "^4.66.1"
-rich = "^13.5.0"
-matplotlib = "^3.7.5"
-seaborn = "^0.13.0"
-# Optional dependencies (vision)
-pillow = { version = ">=12.1.1", optional = true }
-# Optional dependencies (audio)
-torch = { version = ">=2.8.0", python = ">=3.10,<3.14", optional = true }
-torchcodec = { version = ">=0.7.0", python = ">=3.10,<3.14", optional = true }
-
-[tool.poetry.extras]
-vision = ["pillow"]
-audio = ["torch", "torchcodec"]
-
-[tool.poetry.group.dev.dependencies]
-tomli = { version = "^2.0.1", python = "<3.11" }
-types-requests = "==2.31.0.20240125"
-types-setuptools = "==80.9.0.20250822"
-setuptools = "==80.9.0"
-isort = "==5.13.2"
-black = { version = "==25.11.0", extras = ["jupyter"] }
-taplo = "==0.9.3"
-docformatter = "==1.7.5"
-mypy = "==1.8.0"
-pylint = "==3.3.1"
-parameterized = "==0.9.0"
-pytest = "==7.4.4"
-pytest-watcher = "==0.4.3"
-ruff = "==0.14.5"
-devtool = [{ path = "./devtool", develop = true }]
+[project.urls]
+homepage = "https://flower.ai"
+repository = "https://github.com/adap/flower"
+documentation = "https://flower.ai/docs/datasets"
+
+[dependency-groups]
+dev = [
+    "types-requests==2.31.0.20240125",
+    "types-setuptools==80.9.0.20250822",
+    "isort==5.13.2",
+    "black[jupyter]==25.11.0",
+    "taplo==0.9.3",
+    "docformatter==1.7.5",
+    "mypy==1.8.0",
+    "pylint==3.3.1",
+    "parameterized==0.9.0",
+    "pytest==7.4.4",
+    "pytest-watcher==0.4.3",
+    "ruff==0.14.5",
+    "devtool",
+]
+
+[tool.uv]
+default-groups = ["dev"]
+
+[tool.uv.sources]
+devtool = { path = "devtool", editable = true }
+
+[tool.hatch.build.targets.wheel]
+packages = ["flwr_datasets"]
+
+[tool.hatch.build.targets.sdist]
+include = ["README.md", "pyproject.toml", "flwr_datasets/**"]
+
+[tool.hatch.build]
+exclude = ["/flwr_datasets/**/*_test.py"]
 
 [tool.isort]
 line_length = 88