Merge pull request #394 from MaxGhenis/fix-readme-typo

Fix README typo and add PyPI auto-publish workflow

Author: MaxGhenis

.github/workflows/code_changes.yaml

@@ -67,3 +67,26 @@ jobs:
               branch: gh-pages
               folder: docs/_build/html
               clean: true
+  Publish:
+      runs-on: ubuntu-latest
+      needs: [Lint, Test]
+      if: github.event.head_commit.message == 'Update package version'
+      steps:
+          - name: Checkout repo
+            uses: actions/checkout@v4
+          - name: Set up Python
+            uses: actions/setup-python@v5
+            with:
+                python-version: 3.13
+          - name: Install uv
+            uses: astral-sh/setup-uv@v5
+          - name: Install package
+            run: uv pip install -e .[dev] --system
+          - name: Build package
+            run: python -m build
+          - name: Publish a Python distribution to PyPI
+            uses: pypa/gh-action-pypi-publish@release/v1
+            with:
+                user: __token__
+                password: ${{ secrets.PYPI }}
+                skip-existing: true

README.md

@@ -6,7 +6,7 @@ While it is possible to install via PyPi:
 ```bash
 pip install policyengine-us-data
 ```
-the recommended installion is 
+the recommended installation is 
 ```
 pip install -e .[dev]
 ```

changelog_entry.yaml

@@ -0,0 +1,6 @@
+- bump: patch
+  changes:
+    added:
+    - PyPI auto-publish workflow in GitHub Actions
+    fixed:
+    - README typo (installion -> installation)

policyengine_us_data/datasets/cps/cps.py

@@ -122,8 +122,8 @@ def downsample(self, frac: float):
                     original_data[key] = values.astype(original_dtypes[key])
                 except:
                     # If conversion fails, log it but continue
-                    print(
-                        f"Warning: Could not convert {key} back to {original_dtypes[key]}"
+                    logging.warning(
+                        f"Could not convert {key} back to {original_dtypes[key]}"
                     )
                     original_data[key] = values
             else:
@@ -175,11 +175,11 @@ def add_rent(self, cps: h5py.File, person: DataFrame, household: DataFrame):
     inference_df = inference_df[mask]
 
     qrf = QRF()
-    print("Training imputation model for rent and real estate taxes.")
+    logging.info("Training imputation model for rent and real estate taxes.")
     qrf.fit(train_df[PREDICTORS], train_df[IMPUTATIONS])
-    print("Imputing rent and real estate taxes.")
+    logging.info("Imputing rent and real estate taxes.")
     imputed_values = qrf.predict(inference_df[PREDICTORS])
-    print("Imputation complete.")
+    logging.info("Imputation complete.")
     cps["rent"] = np.zeros_like(cps["age"])
     cps["rent"][mask] = imputed_values["rent"]
     # Assume zero housing assistance since
@@ -645,7 +645,7 @@ def add_household_variables(cps: h5py.File, household: DataFrame) -> None:
 
 def add_previous_year_income(self, cps: h5py.File) -> None:
     if self.previous_year_raw_cps is None:
-        print(
+        logging.info(
             "No previous year data available for this dataset, skipping previous year income imputation."
         )
         return