Upload 4 files

Dockerfile

@@ -0,0 +1,8 @@
+FROM python:3.11-slim
+WORKDIR /app
+ENV PYTHONDONTWRITEBYTECODE=1 PYTHONUNBUFFERED=1 MODEL_PATH=/app/diabetes_prediction_model_20251007.pkl
+COPY requirements.txt ./
+RUN pip install --no-cache-dir -r requirements.txt
+COPY . .
+EXPOSE 7860
+CMD uvicorn main:app --host 0.0.0.0 --port ${PORT:-7860}

diabetes_prediction_model_20251007.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:483b1c6e46afa3aedba6565eec3cc80e49fcaa04e4384e14bc34a18c4926f6dd
+size 7290

main.py

@@ -0,0 +1,58 @@
+# main.py  — FastAPI model service
+# Run: uvicorn main:app --reload
+# Requires: pip install fastapi uvicorn pydantic joblib pandas
+
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+from typing import List, Dict, Any
+import pandas as pd
+import joblib
+import os
+
+# Path to your saved pipeline from the notebook
+MODEL_PATH = os.getenv("MODEL_PATH", "diabetes_prediction_model_20251007.pkl")
+pipe = joblib.load(MODEL_PATH)
+
+# Feature schema expected by the pipeline (must match training)
+NUMERIC_FEATURES = ['age','alcohol_consumption_per_week', 'physical_activity_minutes_per_week',
+                     'diet_score', 'bmi', 'cholesterol_total', 'insulin_level', 'map', 'glucose_fasting']
+
+CATEGORICAL_FEATURES = ['gender', 'ethnicity', 'education_level', 'income_level', 'employment_status',
+                         'smoking_status', 'family_history_diabetes', 'hypertension_history', 'cardiovascular_history']
+
+ALL_FEATURES = NUMERIC_FEATURES + CATEGORICAL_FEATURES
+
+class Batch(BaseModel):
+    data: List[Dict[str, Any]]
+
+app = FastAPI()
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"], allow_credentials=True,
+    allow_methods=["*"], allow_headers=["*"],
+)
+
+@app.get("/health")
+def health():
+    return {"status": "ok"}
+
+@app.get("/features")
+def features():
+    return {"features": ALL_FEATURES}
+
+@app.post("/predict_batch")
+def predict_batch(batch: Batch):
+    X = pd.DataFrame(batch.data)
+    # Ensure all expected columns exist
+    for col in ALL_FEATURES:
+        if col not in X.columns:
+            X[col] = 0
+    # Reorder and type-hints (numerics safe-cast, categoricals to str)
+    X = X[ALL_FEATURES].copy()
+    for col in NUMERIC_FEATURES:
+        X[col] = pd.to_numeric(X[col], errors="coerce").fillna(0)
+    for col in CATEGORICAL_FEATURES:
+        X[col] = X[col].astype(str).fillna("Unknown")
+    probs = pipe.predict_proba(X)[:, 1].tolist()
+    return {"probabilities": probs}

requirements.txt

@@ -0,0 +1,8 @@
+fastapi
+uvicorn
+pydantic
+joblib
+pandas
+numpy
+scikit-learn
+xgboost