Merge branch 'master' into feature/mypy-types-cleanlab-internal-util

.github/get_min_dependencies.py

@@ -0,0 +1,15 @@
+"""This script fetches minimum dependencies of cleanlab package and writes them to the file requirements-min.txt"""
+import json
+
+
+if __name__ == "__main__":
+    with open("./deps.json", "r") as f:
+        deps = json.load(f)
+
+    for package in deps:
+        if package["package"]["package_name"] == "cleanlab":
+            for dep in package["dependencies"]:
+                req_version = dep["required_version"]
+                with open("requirements-min.txt", "a") as f:
+                    if req_version.startswith(">="):
+                        f.write(f"{dep['package_name']}=={req_version[2:]}\n")

.github/workflows/ci.yml

@@ -5,6 +5,53 @@ on:
   schedule:
     - cron: '0 12 * * 1'
 jobs:
+  test37:
+    name: "Test: Python 3.7 on ${{ matrix.os }}"
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os:
+          - ubuntu-latest
+          - macos-latest
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4
+        with:
+          python-version: 3.7.16
+      - name: Install cleanlab
+        run: pip install -e .
+      - name: Install development dependencies
+        run: pip install -r requirements-dev.txt
+      - name: Install fasttext for non-Windows machines
+        if: matrix.os != 'windows-latest'
+        run: |
+          pip install fasttext
+      - name: Test with coverage
+        run: pytest --verbose --cov=cleanlab/ --cov-config .coveragerc --cov-report=xml
+        env:
+          TEST_FASTTEXT: true
+      - uses: codecov/codecov-action@v3
+  test37-windows:
+    name: "Test: Python 3.7 on windows"
+    runs-on: windows-latest
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4
+        with:
+          python-version: 3.7.9
+      - name: Install cleanlab
+        run: pip install -e .
+      - name: Install development dependencies
+        run: pip install -r requirements-dev.txt
+      - name: Overwrite tensorflow version on Windows
+        run: |
+          pip uninstall -y tensorflow
+          pip install tensorflow-cpu
+      - name: Test with coverage
+        run: pytest --verbose --cov=cleanlab/ --cov-config .coveragerc --cov-report=xml
+        env:
+          TEST_FASTTEXT: true
+      - uses: codecov/codecov-action@v3
   test:
     name: "Test: Python ${{ matrix.python }} on ${{ matrix.os }}"
     runs-on: ${{ matrix.os }}
@@ -15,7 +62,6 @@ jobs:
           - macos-latest
           - windows-latest
         python:
-          - "3.7"
           - "3.8"
           - "3.9"
           - "3.10"
@@ -26,18 +72,45 @@ jobs:
           python-version: ${{ matrix.python }}
       - name: Install cleanlab
         run: pip install -e .
-      - name: Check cleanlab runs without optional dependencies
-        run: python3 -c "import cleanlab"
       - name: Install development dependencies
         run: pip install -r requirements-dev.txt
+      - name: Install fasttext for non-Windows machines
+        if: matrix.os != 'windows-latest'
+        run: |
+          pip install fasttext
       - name: Overwrite tensorflow version on Windows
         if: matrix.os == 'windows-latest'
         run: |
           pip uninstall -y tensorflow
           pip install tensorflow-cpu
       - name: Test with coverage
         run: pytest --verbose --cov=cleanlab/ --cov-config .coveragerc --cov-report=xml
+        env:
+          TEST_FASTTEXT: true
       - uses: codecov/codecov-action@v3
+  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@v3
+      - uses: actions/setup-python@v4
+        with:
+          python-version: '3.7'
+      - name: Install cleanlab
+        run: |
+          python -m  pip install --upgrade pip
+          pip install .
+      - name: Install test dependencies
+        run: |
+          pip install pytest pytest-lazy-fixture pipdeptree
+          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
   typecheck:
     name: Type check
     runs-on: ubuntu-latest

.github/workflows/links.yml

@@ -17,11 +17,13 @@ jobs:
           find . -name '*.html' -delete
       - run: |
           find . -name '*.md' -exec pandoc -i {} -o {}.html \;
-      - uses: anishathalye/proof-html@v1
+      - uses: anishathalye/proof-html@v2
         with:
           directory: .
+          check_html: false
           check_favicon: false
-          empty_alt_ignore: true
-          url_ignore_re: |
-            ^https:\/\/docs\.github\.com\/
-            ^https:\/\/github\.com\/cleanlab\/cleanlab\/projects
+          ignore_missing_alt: true
+          tokens: |
+            {"https://github.com": "${{ secrets.GITHUB_TOKEN }}"}
+          swap_urls: |
+            {"^\.\/\(.*\).md": "\\1.md.html"}

.gitignore

@@ -118,6 +118,7 @@ venv.bak/
 /docs/source/notebooks/*.gz
 /docs/source/notebooks/spoken_digits
 /docs/source/notebooks/pretrained_models
+/docs/source/tutorials/datalab/datalab-files/
 
 # Editor files
 .vscode/

.mypy.ini

@@ -21,3 +21,12 @@ ignore_missing_imports = True
 
 [mypy-tqdm.*]
 ignore_missing_imports = True
+
+[mypy-matplotlib.*]
+ignore_missing_imports = True
+
+[mypy-datasets.*]
+ignore_missing_imports = True
+
+[mypy-scipy.*]
+ignore_missing_imports = True