Upload 4 files

Dockerfile

@@ -0,0 +1,12 @@
+FROM python:3.11-slim
+
+WORKDIR /app
+
+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", "7860"]

main.py

@@ -0,0 +1,136 @@
+from fastapi import FastAPI, HTTPException, Query
+from fastapi.middleware.cors import CORSMiddleware
+import pandas as pd
+import numpy as np
+import os
+
+app = FastAPI(
+    title="Districtmaps.ai API",
+    description="District-level health risk intelligence across 708 Indian districts. Powered by NFHS-5 data and validated ML models.",
+    version="1.0.0"
+)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# Load data at startup
+DATA_PATH = os.getenv("DATA_PATH", "india_all_districts_risk.csv")
+df = None
+
+@app.on_event("startup")
+def load_data():
+    global df
+    df = pd.read_csv(DATA_PATH)
+    df.columns = [c.strip().lower().replace(" ", "_") for c in df.columns]
+    df["district_lower"] = df["district"].str.lower().str.strip()
+    df["state_lower"]    = df["state"].str.lower().str.strip()
+    print(f"Loaded {len(df)} districts.")
+
+def format_district(row):
+    return {
+        "district":               row.get("district", ""),
+        "state":                  row.get("state", ""),
+        "risk_scores": {
+            "diabetes":           safe_float(row.get("diabetes_risk")),
+            "blood_pressure":     safe_float(row.get("blood_pressure_risk")),
+            "obesity":            safe_float(row.get("obesity_risk")),
+            "anaemia":            safe_float(row.get("anaemia_risk")),
+        },
+        "composite_risk":         safe_float(row.get("composite_risk")),
+        "risk_percentile":        safe_float(row.get("diabetes_risk_norm")),
+    }
+
+def safe_float(val):
+    try:
+        f = float(val)
+        return round(f, 4) if not np.isnan(f) else None
+    except:
+        return None
+
+@app.get("/", tags=["Info"])
+def root():
+    return {
+        "product":     "Districtmaps.ai",
+        "description": "District-level NCD risk intelligence for India",
+        "districts":   len(df) if df is not None else 0,
+        "conditions":  ["diabetes", "blood_pressure", "obesity", "anaemia"],
+        "validation": {
+            "cross_sectional_r2": 0.7477,
+            "temporal_r2":        0.6279,
+            "temporal_gap":       "4 years (NFHS-4 2015-16 → NFHS-5 2019-21)",
+            "districts_covered":  708
+        },
+        "endpoints": {
+            "GET /risk":       "Risk scores for a specific district",
+            "GET /districts":  "Full ranked list of all districts",
+            "GET /top":        "Top N highest risk districts",
+            "GET /state/{state}": "All districts within a state",
+        }
+    }
+
+@app.get("/risk", tags=["Risk Scores"])
+def get_district_risk(
+    district: str = Query(..., description="District name e.g. Mumbai"),
+    state:    str = Query(None, description="Optional state filter to disambiguate")
+):
+    mask = df["district_lower"] == district.lower().strip()
+    if state:
+        mask &= df["state_lower"] == state.lower().strip()
+    results = df[mask]
+    if results.empty:
+        # Fuzzy fallback — partial match
+        mask2 = df["district_lower"].str.contains(district.lower().strip(), na=False)
+        if state:
+            mask2 &= df["state_lower"].str.contains(state.lower().strip(), na=False)
+        results = df[mask2]
+    if results.empty:
+        raise HTTPException(status_code=404, detail=f"District '{district}' not found.")
+    return {
+        "query":   district,
+        "matches": [format_district(row) for _, row in results.iterrows()]
+    }
+
+@app.get("/districts", tags=["Rankings"])
+def get_all_districts(
+    sort_by: str = Query("composite_risk", description="Field to sort by"),
+    order:   str = Query("desc", description="asc or desc"),
+    limit:   int = Query(708, ge=1, le=708)
+):
+    ascending = order == "asc"
+    col = sort_by if sort_by in df.columns else "composite_risk"
+    sorted_df = df.sort_values(col, ascending=ascending).head(limit)
+    return {
+        "total":     len(sorted_df),
+        "sorted_by": col,
+        "order":     order,
+        "districts": [format_district(row) for _, row in sorted_df.iterrows()]
+    }
+
+@app.get("/top", tags=["Rankings"])
+def get_top_districts(
+    n:         int = Query(10, ge=1, le=100, description="Number of districts"),
+    condition: str = Query("composite_risk", description="diabetes_risk | blood_pressure_risk | obesity_risk | anaemia_risk | composite_risk")
+):
+    col = condition if condition in df.columns else "composite_risk"
+    top = df.nlargest(n, col)
+    return {
+        "condition": col,
+        "top_n":     n,
+        "districts": [format_district(row) for _, row in top.iterrows()]
+    }
+
+@app.get("/state/{state}", tags=["State"])
+def get_state_districts(state: str):
+    mask    = df["state_lower"].str.contains(state.lower().strip(), na=False)
+    results = df[mask].sort_values("composite_risk", ascending=False)
+    if results.empty:
+        raise HTTPException(status_code=404, detail=f"State '{state}' not found.")
+    return {
+        "state":     state,
+        "districts": len(results),
+        "ranked":    [format_district(row) for _, row in results.iterrows()]
+    }

requirements.txt

@@ -0,0 +1,5 @@
+fastapi==0.111.0
+uvicorn==0.29.0
+pandas==2.2.2
+numpy==1.26.4
+python-multipart==0.0.9