app.py

from fastapi import FastAPI, HTTPException
import joblib
import pandas as pd
from typing import Dict

app = FastAPI(
    title="Spending Risk ML Backend",
    description="Predicts future spend, spike risk, and spending acceleration",
    version="1.0.0"
)

# -----------------------------
# Load models ONCE at startup
# -----------------------------
try:
    future_spend_model = joblib.load("future_spend_7d.pkl")
    spike_model = joblib.load("spike_probability.pkl")
    acceleration_model = joblib.load("acceleration.pkl")
    FEATURES = joblib.load("model_features.pkl")
except Exception as e:
    raise RuntimeError(f"❌ Model loading failed: {e}")

# -----------------------------
# Health check (HF requirement)
# -----------------------------
@app.get("/")
def health_check():
    return {
        "status": "running",
        "service": "spending-risk-backend"
    }

# -----------------------------
# Prediction endpoint
# -----------------------------
@app.post("/predict")
def predict(payload: Dict):
    try:
        # 1. Build feature vector safely
        #    Missing features -> default 0
        input_row = {feat: payload.get(feat, 0) for feat in FEATURES}
        X = pd.DataFrame([input_row])

        # 2. Predictions
        future_spend = float(future_spend_model.predict(X)[0])
        spike_prob = float(spike_model.predict_proba(X)[0][1])
        acceleration = float(acceleration_model.predict(X)[0])

        # 3. Response (frontend-friendly)
        return {
            "future_7d_spend": round(future_spend, 2),
            "spike_probability": round(spike_prob, 3),
            "acceleration": round(acceleration, 2)
        }

    except Exception as e:
        raise HTTPException(
            status_code=400,
            detail=f"Prediction failed: {str(e)}"
        )