feat: upgrade to Python 3.13

- Update Python version from 3.8 to 3.13.0
- Update dependencies to latest compatible versions
- Update scikit-learn OneHotEncoder parameter for compatibility
- Add version upgrade documentation

Dependencies updated:
- numpy >= 1.26.0
- pandas >= 2.1.0
- scikit-learn >= 1.3.0
- fastapi >= 0.103.0
- uvicorn >= 0.23.0
- gunicorn >= 21.2.0

Author: SudKul

PYTHON_3_13_UPDATES.md

@@ -0,0 +1,20 @@
+# Python 3.13 Compatibility Updates
+
+This document summarizes the changes made to make the codebase compatible with Python 3.13.
+
+## Dependencies Updated
+- Python: 3.8 → 3.13.0
+- NumPy: Latest compatible (1.26.0+)
+- Pandas: Latest compatible (2.1.0+)
+- scikit-learn: Latest compatible (1.3.0+)
+- FastAPI: 0.63.0 → 0.103.0+
+- Other dependencies updated to latest stable versions
+
+## Code Changes
+1. **ML Data Processing (`ml/data.py`)**
+   - Updated OneHotEncoder parameters from `sparse=False` to `sparse_output=False` to match newer scikit-learn API
+
+## Notes
+- The core ML functionality and boilerplate structure remains unchanged
+- Dependencies updated to ensure compatibility with Python 3.13
+- Starter code structure maintained for student implementation

starter/requirements.txt

@@ -1,9 +1,9 @@
-python==3.8
-numpy
-pandas
-scikit-learn
-pytest
-requests
-fastapi==0.63.0
-uvicorn
-gunicorn
+python>=3.13.0
+numpy>=1.26.0
+pandas>=2.1.0
+scikit-learn>=1.3.0
+pytest>=7.4.0
+requests>=2.31.0
+fastapi>=0.103.0
+uvicorn>=0.23.0
+gunicorn>=21.2.0

starter/starter/ml/data.py

@@ -54,7 +54,7 @@ def process_data(
     X_continuous = X.drop(*[categorical_features], axis=1)
 
     if training is True:
-        encoder = OneHotEncoder(sparse=False, handle_unknown="ignore")
+        encoder = OneHotEncoder(sparse_output=False, handle_unknown="ignore")
         lb = LabelBinarizer()
         X_categorical = encoder.fit_transform(X_categorical)
         y = lb.fit_transform(y.values).ravel()