chore: import upstream snapshot with attribution

.github/workflows/bleeding-edge.yml

@@ -0,0 +1,35 @@
+name: Bleeding Edge Dependencies Test
+on:
+  schedule:
+    - cron: '0 6 1 * *'   # First day of each month at 6:00 UTC
+  workflow_dispatch: # Allow manual triggering
+
+jobs:
+  test-bleeding-edge:
+    name: "Bleeding Edge Dependencies: Python Latest"
+    runs-on: ubuntu-latest
+    continue-on-error: true
+    steps:
+      - uses: actions/checkout@v6
+      - uses: snok/latest-python-versions@v1
+        id: python-versions
+        with:
+          min-version: '3.14'
+          include-prereleases: false
+      - uses: actions/setup-python@v6
+        with:
+          python-version: ${{ fromJson(steps.python-versions.outputs.latest-python-versions)[0] }}
+      - name: Download cleanvision test data
+        run: |
+          git clone https://github.com/cleanlab/assets.git
+          mv assets/cleanlab_test_data ./tests/datalab/data
+      - name: Install bleeding-edge dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install --pre --upgrade numpy scikit-learn pandas tqdm termcolor
+          pip install --pre torch torchvision || pip install torch torchvision
+          pip install -e ".[all]"
+      - name: Install test dependencies
+        run: pip install --no-cache-dir -r requirements-test-core.txt
+      - name: Test with bleeding-edge deps
+        run: pytest --verbose --maxfail=10 -m "not slow" --ignore=tests/test_frameworks.py --ignore=tests/test_model_pytorch_cnn.py

.github/workflows/ci.yml

@@ -0,0 +1,165 @@
+name: CI
+on:
+  push:
+  pull_request:
+  schedule:
+    - cron: '0 12 * * 1'  # Monday at 12:00 UTC
+jobs:
+  test-core:
+    name: "Core Tests: Python ${{ matrix.python }} on ${{ matrix.os }}"
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os:
+          - ubuntu-latest
+          - macos-latest
+          - windows-latest
+        python:
+          - "3.10"
+          - "3.11"
+          - "3.12"
+          - "3.13"
+          - "3.14"
+    steps:
+      - uses: actions/checkout@v6
+      - uses: actions/setup-python@v6
+        with:
+          python-version: ${{ matrix.python }}
+      - name: Download cleanvision test data
+        run: |
+          git clone https://github.com/cleanlab/assets.git
+          mv assets/cleanlab_test_data ./tests/datalab/data
+      - name: Install cleanlab
+        run: |
+          python -m pip install --upgrade pip
+          pip install -e ".[all]"
+      - name: Install core test dependencies
+        run: pip install --no-cache-dir -r requirements-test-core.txt
+      - name: Audit dependencies for vulnerabilities
+        uses: pypa/gh-action-pip-audit@v1.1.0
+        with:
+          inputs: requirements-test-core.txt
+      - name: Run core tests (excluding ML framework tests)
+        run: pytest --verbose --maxfail=30 --order-tests --cov=cleanlab/ --cov-config .coveragerc --cov-report=xml -m "not slow" --ignore=tests/test_frameworks.py --ignore=tests/test_model_pytorch_cnn.py
+      - uses: codecov/codecov-action@v5
+        with:
+          version: "v0.7.3"
+  
+  test-ml-frameworks:
+    name: "ML Framework Tests: Python ${{ matrix.python }} on ubuntu-latest"
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python:
+          - "3.10"
+          - "3.11"
+          - "3.12"
+          - "3.13"
+          - "3.14"
+    steps:
+      - uses: actions/checkout@v6
+      - uses: actions/setup-python@v6
+        with:
+          python-version: ${{ matrix.python }}
+      - name: Download cleanvision test data
+        run: |
+          git clone https://github.com/cleanlab/assets.git
+          mv assets/cleanlab_test_data ./tests/datalab/data
+      - name: Install cleanlab
+        run: |
+          python -m pip install --upgrade pip
+          pip install -e ".[all]"
+      - name: Install ML framework dependencies
+        run: pip install --no-cache-dir -r requirements-dev.txt
+      - name: Audit dependencies for vulnerabilities
+        uses: pypa/gh-action-pip-audit@v1.1.0
+        with:
+          inputs: requirements-dev.txt
+      - name: Run ML framework tests
+        run: pytest --verbose --maxfail=30 tests/test_frameworks.py tests/test_model_pytorch_cnn.py
+  test-without-extras-min-versions:
+    name: Test without optional dependencies and with minimum compatible versions of dependencies
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+      - uses: actions/setup-python@v6
+        with:
+          python-version: '3.10'
+      - name: Install cleanlab
+        run: |
+          python -m  pip install --upgrade pip
+          pip install .
+      - name: Install test dependencies
+        run: |
+          pip install pytest pipdeptree hypothesis
+          pipdeptree -j > deps.json
+      - name: Install minimum versions
+        run: |
+          python ./.github/get_min_dependencies.py
+          pip install -r requirements-min.txt
+      - name: Run tests
+        run: |
+          pytest tests/test_multilabel_classification.py tests/test_multiannotator.py tests/test_filter_count.py
+  test-max-versions:
+    name: "Maximum Dependency Versions: Python 3.14"
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: '3.14'
+      - name: Download cleanvision test data
+        run: |
+          git clone https://github.com/cleanlab/assets.git
+          mv assets/cleanlab_test_data ./tests/datalab/data
+      - name: Install latest stable versions
+        run: |
+          python -m pip install --upgrade pip
+          pip install --upgrade numpy scikit-learn pandas tqdm termcolor
+          pip install --upgrade torch torchvision
+          pip install -e ".[all]"
+      - name: Install test dependencies
+        run: pip install --no-cache-dir -r requirements-test-core.txt
+      - name: Test with latest stable versions
+        run: pytest --verbose --maxfail=10 -m "not slow" --ignore=tests/test_frameworks.py --ignore=tests/test_model_pytorch_cnn.py  
+  typecheck:
+    name: Type check
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+      - uses: actions/setup-python@v6
+        with:
+          python-version: '3.14'
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install ".[all]"  # install dependencies
+          pip install --no-cache-dir -r requirements-test-core.txt  # install core dependencies and type stubs
+      - name: Type check
+        run: mypy --install-types --non-interactive cleanlab
+  fmt:
+    name: Format
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+      - uses: psf/black@stable
+  flake8:
+    name: Check for unused/wildcard imports
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+      - uses: actions/setup-python@v6
+        with:
+          python-version: '3.14'
+      - name: Install flake8
+        run: pip install flake8
+      - name: Lint with flake8
+        run: flake8 cleanlab tests
+  nblint:
+    name: Lint Notebooks
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+      - uses: cleanlab/nblint-action@v1
+        with:
+          directory: 'docs'

.github/workflows/gh-pages.yaml

@@ -0,0 +1,210 @@
+name: GitHub Pages
+
+on:
+  push:
+    branches:
+      - "**"
+
+  release:
+    types: [published]
+
+  workflow_dispatch:
+
+jobs:
+  deploy:
+    runs-on: ubuntu-24.04
+    concurrency:
+      group: ${{ github.workflow }}-${{ github.ref }}
+    steps:
+      - uses: actions/checkout@v6
+        with:
+          fetch-depth: 0
+
+      - name: Install Pandoc
+        # pin Pandoc version
+        run: |
+          cd /tmp
+          wget https://github.com/jgm/pandoc/releases/download/2.19.2/pandoc-2.19.2-linux-amd64.tar.gz
+          sudo tar xzvf pandoc-2.19.2-linux-amd64.tar.gz --strip-components 1 -C /usr/local
+
+      - name: Setup Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: "3.11"
+
+      - name: Setup Node
+        uses: actions/setup-node@v6
+        with:
+          node-version: "22"
+  
+      - name: Install FFmpeg
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y ffmpeg
+
+      - name: Upgrade pip
+        run: python3 -m pip install --upgrade pip
+
+      # Caching removed to resolve disk space issues
+      # Heavy ML packages (TensorFlow, PyTorch, HuggingFace) will be downloaded fresh each run
+      # but this avoids the 3.7GB cache that causes "No space left on device" errors
+
+      - name: Install cleanlab
+        run: pip install -e ".[all]"
+
+      - name: Install dependencies
+        run: python3 -m pip install -r docs/requirements.txt
+
+      - name: Install katex
+        run: npm install -g katex
+
+      - name: Find and replace "%pip install cleanlab" with master branch in .ipynb files
+        if: ${{ github.ref_type == 'branch' }}
+        uses: jacobtomlinson/gha-find-replace@v2
+        with:
+          find: "%pip install cleanlab  ${{ '#' }} for colab"
+          replace: "%pip install git+https://github.com/${{ github.repository }}.git@${{ github.sha }}"
+          include: "docs/source**.ipynb"
+
+      - name: Find and replace "%pip install cleanlab" with release tag in .ipynb files
+        if: ${{ github.ref_type == 'tag' }}
+        uses: jacobtomlinson/gha-find-replace@v2
+        with:
+          find: "%pip install cleanlab  ${{ '#' }} for colab"
+          replace: "%pip install cleanlab==${{ github.ref_name }}"
+          include: "docs/source**.ipynb"
+
+      - name: Add and Commit changes to .ipynb files
+        run: |
+          git config --global user.email "actions@github.com"
+          git config --global user.name "GitHub Actions"
+          git add -A && git commit -m "Change cleanlab ver in .ipynb files"
+
+      - name: Change tag to the latest commit
+        if: ${{ github.ref_type == 'tag' }}
+        run: |
+          git tag -f $GITHUB_REF_NAME
+
+      - name: Run Sphinx with sphinx-multiversion wrapper - branch trigger
+        if: ${{ github.ref_type == 'branch' }}
+        env:
+          TF_CPP_MIN_LOG_LEVEL: 3
+          HF_HOME: /tmp/hf_cache
+          HF_HUB_DISABLE_SYMLINKS_WARNING: 1
+          TRANSFORMERS_CACHE: /tmp/transformers
+          TORCH_HOME: /tmp/torch
+        run: sphinx-multiversion docs/source cleanlab-docs -D smv_branch_whitelist=${{ github.ref_name }} -D smv_tag_whitelist=None
+
+      - name: Run Sphinx with sphinx-multiversion wrapper - tag trigger
+        if: ${{ github.ref_type == 'tag' }}
+        env:
+          TF_CPP_MIN_LOG_LEVEL: 3
+          HF_HOME: /tmp/hf_cache
+          HF_HUB_DISABLE_SYMLINKS_WARNING: 1
+          TRANSFORMERS_CACHE: /tmp/transformers
+          TORCH_HOME: /tmp/torch
+        run: sphinx-multiversion docs/source cleanlab-docs -D smv_branch_whitelist=None -D smv_tag_whitelist=${{ github.ref_name }}
+
+      - name: Debug temporary cache usage (not saved)
+        run: |
+          echo "=== Temporary Cache Usage ==="
+          echo "These caches are created in /tmp and cleaned up after build:"
+          echo "  /tmp/hf_cache: $(du -sh /tmp/hf_cache 2>/dev/null | cut -f1 || echo 'not found')"
+          echo "  /tmp/transformers: $(du -sh /tmp/transformers 2>/dev/null | cut -f1 || echo 'not found')"
+          echo "  /tmp/torch: $(du -sh /tmp/torch 2>/dev/null | cut -f1 || echo 'not found')"
+          echo "Pip cache directory: $(pip cache dir)"
+          echo "Pip cache size: $(du -sh $(pip cache dir) 2>/dev/null | cut -f1 || echo 'not found')"
+          echo "========================"
+
+      - name: Get latest stable release
+        id: stable_release
+        continue-on-error: true
+        uses: pozetroninc/github-action-get-latest-release@master
+        with:
+          owner: ${{ github.repository_owner }}
+          repo: cleanlab
+          excludes: prerelease, draft
+
+      - name: Create latest_release.txt file
+        if: ${{ steps.stable_release.outcome == 'success' }}
+        run: echo ${{ steps.stable_release.outputs.release }} > cleanlab-docs/latest_release.txt
+
+      - name: Find and replace "placeholder_version_number" in versioning.js
+        if: ${{ steps.stable_release.outcome == 'success' }}
+        uses: jacobtomlinson/gha-find-replace@v2
+        with:
+          find: "placeholder_version_number"
+          replace: ${{ steps.stable_release.outputs.release }}
+          include: "docs/source/_templates/versioning.js"
+
+      - name: Find and replace "placeholder_commit_hash" in versioning.js
+        uses: jacobtomlinson/gha-find-replace@v2
+        with:
+          find: "placeholder_commit_hash"
+          replace: ${{ github.sha }}
+          include: "docs/source/_templates/versioning.js"
+
+      - name: Copy versioning JS file to cleanlab-docs/
+        run: |
+          cp docs/source/_templates/versioning.js cleanlab-docs/versioning.js
+
+      - name: Find and replace "stable_url" in redirect-to-stable.html
+        if: ${{ github.ref_type == 'tag' && steps.stable_release.outcome == 'success' }}
+        uses: jacobtomlinson/gha-find-replace@v2
+        with:
+          find: "stable_url"
+          replace: ${{ env.DOCS_SITE_URL }}stable/index.html
+          include: "docs/source/_templates/redirect-to-stable.html"
+
+      - name: Copy redirecting HTML file to cleanlab-docs/
+        if: ${{ github.ref_type == 'tag' }}
+        run: cp docs/source/_templates/redirect-to-stable.html cleanlab-docs/index.html
+
+      - name: Update OpenGraph URLs in HTML files
+        # This code fixes the URLs in the HTML files with OpenGraph tags.
+        # It uses a regular expression to capture two parts of the URL:
+        #   $1: The first part of the URL, including the protocol and domain.
+        #   $2: The rest of the URL, including the path and query string.
+        # The replacement string combines the captured parts with the ref name.
+        # This ensures that the ref name is inserted into the URL correctly.
+        env:
+          REF_URL_SEGMENT: ${{ (github.ref_type == 'tag' && steps.stable_release.outcome == 'success') && 'stable' || github.ref_name }}
+        uses: jacobtomlinson/gha-find-replace@v2
+        with:
+          find: '(<meta property="og:url" content="https?://[^/]+)/([^"]+")'
+          replace: '$1/${{ env.REF_URL_SEGMENT }}/$2'
+          include: "cleanlab-docs/**/*.html"
+          regex: true
+      
+      - name: Deploy
+        if: ${{ (github.ref == 'refs/heads/master') || (github.ref_type == 'tag') }}
+        uses: peaceiris/actions-gh-pages@v4
+        with:
+          deploy_key: ${{ secrets.ACTIONS_DEPLOY_KEY }}
+          external_repository: ${{ github.repository_owner }}/cleanlab-docs
+          publish_branch: master
+          publish_dir: cleanlab-docs
+          keep_files: true
+          exclude_assets: ""
+
+      - uses: actions/upload-artifact@v5
+        with:
+          name: docs-html
+          path: cleanlab-docs
+          retention-days: 1 # don't need this except in link checking step below
+
+  links:
+    needs: deploy
+    runs-on: ubuntu-24.04
+    steps:
+      - uses: actions/download-artifact@v5
+        with:
+          name: docs-html
+          path: cleanlab-docs
+      - uses: anishathalye/proof-html@v1
+        with:
+          directory: ./cleanlab-docs
+          external_only: true
+          url_ignore_re: |
+            ^https://pradyunsg.me/
+            ^https://keras.io/

.github/workflows/links.yml

@@ -0,0 +1,35 @@
+name: GitHub Markdown Links
+on:
+  push:
+  pull_request:
+  schedule:
+    - cron: "0 8 * * 6"
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - run: sudo apt-get update -y
+      - run: >-
+          sudo apt-get install -y
+          pandoc
+      - uses: actions/checkout@v6
+      - run: |
+          find . -name '*.html' -delete
+      - run: |
+          find . -name '*.md' -exec pandoc -i {} -o {}.html \;
+      - uses: anishathalye/proof-html@v2
+        with:
+          directory: .
+          check_html: false
+          check_favicon: false
+          ignore_missing_alt: true
+          tokens: |
+            {"https://github.com": "${{ secrets.GITHUB_TOKEN }}"}
+          ignore_url: |
+            https://jair.org/index.php/jair/article/view/12125
+            https://stackoverflow.com/questions/41573587/what-is-the-difference-between-venv-pyvenv-pyenv-virtualenv-virtualenvwrappe
+          ignore_url_re: |
+            https://www.gnu.org/software/wget
+          swap_urls: |
+            {"^(\\..*)\\.md(#?.*)$": "\\1.md.html\\2",
+             "^(https://github\\.com/.*)#.*$": "\\1"}

.github/workflows/manual-testpypi.yml

@@ -0,0 +1,154 @@
+name: Manual TestPyPI Release
+
+on:
+  workflow_dispatch:  # Manual trigger only
+    inputs:
+      reason:
+        description: 'Reason for this test release'
+        required: false
+        default: 'Testing changes before official release'
+
+jobs:
+  extract_project_name:
+    runs-on: ubuntu-latest
+    outputs:
+      cleanlab_package_name: ${{ steps.extract_name.outputs.value }}
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v6
+
+      - name: TOML Reader
+        uses: SebRollen/toml-action@v1.2.0
+        id: extract_name
+        with:
+          file: "pyproject.toml"
+          field: "project.name"
+
+  build:
+    needs: extract_project_name
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v6
+
+      - name: Remove tests directory
+        run: |
+          TEST_DIR=$(find . -type d -name "tests")
+          if [ -n "$TEST_DIR" ]; then
+            rm -rf $TEST_DIR
+          else
+            echo "No tests directory found! This is unexpected."
+            exit 1
+          fi
+
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.11
+
+      - name: Modify version for TestPyPI
+        run: |
+          # Add .dev suffix with run number for TestPyPI
+          sed -i "s/__version__ = \"\(.*\)\"/__version__ = \"\1.dev${{ github.run_number }}\"/" cleanlab/version.py
+          echo "Modified version for TestPyPI:"
+          grep "__version__" cleanlab/version.py
+
+      - name: Install dependencies
+        run: pip install --upgrade "setuptools>=65.0,<70.0" wheel twine==5.0.0 pkginfo
+
+      - name: Clean build artifacts
+        run: |
+          rm -rf build/ dist/ *.egg-info cleanlab.egg-info/
+
+      - name: Perform some tests on the metadata of the package
+        run: |
+          python setup.py check --metadata --strict
+
+      - name: Build package
+        run: |
+          python setup.py sdist bdist_wheel
+
+      - name: Test package
+        run: |
+          twine check dist/*
+
+      - name: Display package info
+        run: |
+          echo "📦 Built packages:"
+          ls -lh dist/
+          echo ""
+          echo "📝 Reason: ${{ github.event.inputs.reason }}"
+
+      - name: Store build artifacts
+        uses: actions/upload-artifact@v5
+        with:
+          name: package
+          path: dist/*
+
+  publish-testpypi:
+    needs: build
+    runs-on: ubuntu-latest
+    permissions:
+      id-token: write  # IMPORTANT: this permission is mandatory for trusted publishing
+
+    steps:
+      - name: Fetch build artifacts
+        uses: actions/download-artifact@v5
+        with:
+          name: package
+          path: dist/
+
+      - name: Publish to TestPyPI
+        uses: pypa/gh-action-pypi-publish@e53eb8b103ffcb59469888563dc324e3c8ba6f06
+        with:
+          repository-url: https://test.pypi.org/legacy/
+          verify-metadata: true
+
+  test-basic-import:
+    needs: [publish-testpypi, extract_project_name]
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.11
+
+      - name: Install package from TestPyPI
+        env:
+          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
+        run: pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ $CLEANLAB_PACKAGE_NAME
+
+      - name: Test importing package
+        run: |
+          python -c "import cleanlab
+          from cleanlab.outlier import OutOfDistribution
+          ood = OutOfDistribution()
+          print(ood)
+          print('✅ Basic import test passed!')
+          print('Version:', cleanlab.__version__)"
+
+  test-extras-import:
+    needs: [publish-testpypi, extract_project_name]
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.11
+
+      - name: Install package with extras
+        env:
+          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
+        run: pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ $CLEANLAB_PACKAGE_NAME[all]
+
+      - name: Test importing package with extras
+        run: |
+          python -c "import cleanlab
+          from cleanlab import Datalab
+          lab = Datalab({'a': [1, 2, 3, 4, 5], 'b': ['a', 'b', 'c', 'b', 'a']})
+          print(lab)
+          print('✅ Extras import test passed!')
+          print('Version:', cleanlab.__version__)"

.github/workflows/release-build-publish.yml

@@ -0,0 +1,260 @@
+name: Release Build Publish
+
+on:
+    release:
+      types: [ published ]
+
+jobs:
+  extract_project_name:
+    runs-on: ubuntu-latest
+    outputs:
+      cleanlab_package_name: ${{ steps.extract_name.outputs.value }}
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v6
+
+      - name: TOML Reader
+        uses: SebRollen/toml-action@v1.2.0
+        id: extract_name
+        with:
+          file: "pyproject.toml"
+          field: "project.name"
+
+  store_published_version:
+    runs-on: ubuntu-latest
+    outputs:
+      version: ${{ steps.compare.outputs.version }}
+    steps:
+
+      - name: Checkout code
+        uses: actions/checkout@v6
+
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.11
+
+      - name: Install package
+        run: pip install .
+
+      - name: Compare the published version with the package version
+        id: compare
+        run: |
+          TAG_NAME="${{ github.ref_name }}"
+          # TAG_NAME must be in the form "v1.2.3[...]", error if it doesn't match
+            if [[ ! $TAG_NAME =~ ^v[0-9]+\.[0-9]+\.[0-9]+ ]]; then
+                echo "Tag name '$TAG_NAME' does not match the expected pattern 'vX.Y.Z'"
+                exit 1
+            fi
+          # Strip the 'v' prefix from the tag, and store the version in an output variable
+          VERSION=${TAG_NAME:1} # Strip the 'v' prefix from the tag
+          echo "version=$VERSION" >> $GITHUB_OUTPUT
+          # Check that the package version matches the tag version
+          PACKAGE_VERSION=$(python -c "import cleanlab; print(cleanlab.__version__)")
+          if [ "$PACKAGE_VERSION" != "$VERSION" ]; then
+            echo "Package version $PACKAGE_VERSION does not match tag $VERSION"
+            exit 1
+          fi
+
+  build:
+    needs: [store_published_version, extract_project_name]
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v6
+
+      - name: Remove tests directory
+        run: |
+          # Remove the tests directory from the package
+          TEST_DIR=$(find . -type d -name "tests")
+          if [ -n "$TEST_DIR" ]; then
+            rm -rf $TEST_DIR
+          else
+            echo "No tests directory found! This is unexpected."
+            exit 1
+          fi
+
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.11
+
+      - name: Install dependencies
+        run: pip install --upgrade "setuptools>=65.0,<70.0" wheel twine==5.0.0 pkginfo
+
+      - name: Clean build artifacts
+        run: |
+          rm -rf build/ dist/ *.egg-info cleanlab.egg-info/
+
+      - name: Perform some tests on the metadata of the package
+        run: |
+          python setup.py check --metadata --strict
+
+      - name: Build package
+        run: |
+          python setup.py sdist bdist_wheel
+
+      - name: Test package
+        run: |
+          twine check dist/*
+
+      - name: Verify contents of the tar.gz
+        env:
+          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
+        run: |
+          # Extract the tar.gz to a temporary directory
+          TMPDIR=$(mktemp -d)
+          tar -xzf dist/*.tar.gz -C $TMPDIR
+          PACKAGE_DIR=$TMPDIR/$(ls $TMPDIR) # Assuming there's only one directory extracted
+
+          # Define expected files and directories
+          EXPECTED_FILES="LICENSE PKG-INFO MANIFEST.in DEVELOPMENT.md CONTRIBUTING.md CODE_OF_CONDUCT.md README.md pyproject.toml setup.cfg setup.py"
+          EXPECTED_DIRS="cleanlab $CLEANLAB_PACKAGE_NAME.egg-info"
+
+          # Collect actual top-level files and directories
+          ACTUAL_CONTENTS=$(ls $PACKAGE_DIR)
+          ACTUAL_FILES=$(find $PACKAGE_DIR -maxdepth 1 -type f -exec basename {} \;)
+          ACTUAL_DIRS=$(find $PACKAGE_DIR -maxdepth 1 -type d -exec basename {} \; | sed "1d") # Remove the package directory itself from the list
+
+          # Function to check for unexpected or missing files and directories
+          check_contents() {
+            local missing=0
+            # Check for expected files
+            for expected_file in $EXPECTED_FILES; do
+              echo $ACTUAL_FILES | grep -qw $expected_file || { echo "Missing expected file: $expected_file"; missing=1; }
+            done
+            
+            # Check for expected directories
+            for expected_dir in $EXPECTED_DIRS; do
+              echo $ACTUAL_DIRS | grep -qw $expected_dir || { echo "Missing expected directory: $expected_dir"; missing=1; }
+            done
+
+            # Check for unexpected files and directories
+            for actual in $ACTUAL_CONTENTS; do
+              echo $EXPECTED_FILES $EXPECTED_DIRS | grep -qw $actual || { echo "Unexpected item in package: $actual"; missing=1; }
+            done
+
+            # Exit if anything unexpected or missing
+            if [ $missing -ne 0 ]; then
+              echo "Package content verification failed."
+              exit 1
+            fi
+          }
+
+          # Execute checks
+          check_contents
+
+      - name: Store build artifacts
+        uses: actions/upload-artifact@v5
+        with:
+          name: package
+          path: dist/*
+
+  publish-testpypi:
+    needs: build
+    runs-on: ubuntu-latest
+    permissions:
+        id-token: write  # IMPORTANT: this permission is mandatory for trusted publishing
+
+    steps:
+      - name: Fetch build artifacts
+        uses: actions/download-artifact@v5
+        with:
+          name: package
+          path: dist/
+
+      - name: Publish to TestPyPI
+        uses: pypa/gh-action-pypi-publish@e53eb8b103ffcb59469888563dc324e3c8ba6f06
+        with:
+          repository-url: https://test.pypi.org/legacy/
+          verify-metadata: true
+
+  verify-version:
+    needs: [publish-testpypi, store_published_version, extract_project_name]
+    runs-on: ubuntu-latest
+    environment: testpypi
+
+    steps:
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.11
+
+      - name: Install package from TestPyPI
+        env:
+          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
+        run: pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ $CLEANLAB_PACKAGE_NAME
+
+      - name: Verify the published version
+        env:
+          VERSION: ${{ needs.store_published_version.outputs.version }}
+        run: |
+          TEST_IMPORT_VERSION=$(python -c "import cleanlab; print(cleanlab.__version__)")
+          if [ "$TEST_IMPORT_VERSION" != "$VERSION" ]; then
+            echo "Imported version $TEST_IMPORT_VERSION does not match tag $VERSION"
+            exit 1
+          fi
+
+  test-basic-import:
+    needs: [verify-version, extract_project_name]
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.11
+
+      - name: Install package from TestPyPI
+        env:
+          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
+        run: pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ $CLEANLAB_PACKAGE_NAME
+
+      - name: Test importing package
+        run: |
+          python -c "import cleanlab
+          from cleanlab.outlier import OutOfDistribution
+          ood = OutOfDistribution()
+          print(ood)"
+
+  test-extras-import:
+    needs: [verify-version, extract_project_name]
+    runs-on: ubuntu-latest
+ 
+    steps:
+      - name: Set up Python
+        uses: actions/setup-python@v6
+        with:
+          python-version: 3.11
+
+      - name: Install package with extras
+        env:
+          CLEANLAB_PACKAGE_NAME: ${{ needs.extract_project_name.outputs.cleanlab_package_name }}
+        run: pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ $CLEANLAB_PACKAGE_NAME[all]
+
+      - name: Test importing package with extras
+        run: |
+          python -c "import cleanlab
+          from cleanlab import Datalab
+          lab = Datalab({'a': [1, 2, 3, 4, 5], 'b': ['a', 'b', 'c', 'b', 'a']})
+          print(lab)"
+
+  publish-pypi:
+    needs: [test-basic-import, test-extras-import]
+    runs-on: ubuntu-latest
+    environment: pypi
+    permissions:
+        id-token: write  # IMPORTANT: this permission is mandatory for trusted publishing
+
+    steps:
+      - name: Fetch build artifacts
+        uses: actions/download-artifact@v5
+        with:
+          name: package
+          path: dist/
+
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@e53eb8b103ffcb59469888563dc324e3c8ba6f06
+        with:
+          verify-metadata: true