model/predict.py

"""
Apple Price Prediction Model - Inference Script
Indian Market | Prices in INR (₹) per kg
Hybrid Prophet + ARIMA Ensemble
"""

import os
import warnings
warnings.filterwarnings('ignore')

import numpy as np
import pandas as pd
import joblib
from datetime import datetime, timedelta

# Paths
_BASE_DIR     = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
_PROPHET_PATH = os.path.join(_BASE_DIR, 'models', 'prophet_model.pkl')
_ARIMA_PATH   = os.path.join(_BASE_DIR, 'models', 'arima_model.pkl')
_SCALER_PATH  = os.path.join(_BASE_DIR, 'models', 'scaler.pkl')

# Lazy-loaded singletons
_prophet_model = None
_arima_model   = None
_scaler        = None


def _load_models():
    global _prophet_model, _arima_model, _scaler
    if _prophet_model is None:
        _prophet_model = joblib.load(_PROPHET_PATH)
    if _arima_model is None:
        _arima_model = joblib.load(_ARIMA_PATH)
    if _scaler is None:
        _scaler = joblib.load(_SCALER_PATH)


def _storage_cost_inr(days: int) -> float:
    """
    Indian cold-storage rate: ₹0.75/kg/day.
    Returns total storage cost for given number of days.
    """
    return days * 0.75


def predict_price(input_features: dict) -> dict:
    """
    Predict apple wholesale price 7 days ahead and recommend
    whether to SELL now or STORE for higher future returns.

    Parameters
    ----------
    input_features : dict
        Required:
            current_price     (float)  Current wholesale price in ₹/kg
        Optional:
            date              (str)    'YYYY-MM-DD'  (default: today)
            storage_time_days (int)    Days already in cold storage (default: 0)
            apple_variety     (str)    One of: Shimla, Kinnauri,
                                       Royal Delicious, Golden Delicious, Maharaji
            region            (str)    Indian growing region

    Returns
    -------
    dict
        {
            "predicted_price_7d":  float    # ₹/kg forecast 7 days ahead
            "recommendation":      str      # "SELL" or "STORE"
            "current_price":       float    # ₹/kg (input)
            "storage_cost_7d":     float    # ₹/kg cost for 7 more days
            "breakeven_price":     float    # min price needed to justify storing
            "currency":            str      # "INR"
            "confidence":          str
        }
    """
    _load_models()

    date_str      = input_features.get('date', datetime.today().strftime('%Y-%m-%d'))
    current_price = float(input_features.get('current_price', 90.0))   # ₹/kg default
    storage_days  = int(input_features.get('storage_time_days', 0))

    target_date = pd.to_datetime(date_str) + timedelta(days=7)

    # ── Prophet forecast ──────────────────────────────────────
    future_df = pd.DataFrame({'ds': [target_date]})
    prophet_forecast = _prophet_model.predict(future_df)
    prophet_pred = float(prophet_forecast['yhat'].iloc[0])

    # ── ARIMA forecast (7 steps ahead) ───────────────────────
    arima_pred = float(_arima_model.predict(n_periods=7)[-1])

    # ── Hybrid blend ─────────────────────────────────────────
    predicted_price_7d = 0.6 * prophet_pred + 0.4 * arima_pred
    predicted_price_7d = max(30.0, round(predicted_price_7d, 2))

    # ── Sell / Store decision ─────────────────────────────────
    # Store only if the predicted price after 7 days exceeds
    # current price + cost of storing 7 more days
    storage_cost_7d  = _storage_cost_inr(7)
    breakeven_price  = round(current_price + storage_cost_7d, 2)
    recommendation   = "STORE" if predicted_price_7d > breakeven_price else "SELL"

    return {
        "predicted_price_7d": predicted_price_7d,
        "recommendation":     recommendation,
        "current_price":      round(current_price, 2),
        "storage_cost_7d":    round(storage_cost_7d, 2),
        "breakeven_price":    breakeven_price,
        "currency":           "INR",
        "confidence":         "hybrid Prophet+ARIMA (0.6/0.4)",
    }


# ─── CLI demo ────────────────────────────────────────────────
if __name__ == '__main__':
    # Example: Kinnauri apples at ₹120/kg, 15 days in cold storage
    sample = {
        'date':              '2026-03-07',
        'current_price':     120.0,       # ₹/kg
        'storage_time_days': 15,
        'apple_variety':     'Kinnauri',
        'region':            'Himachal Pradesh',
    }
    result = predict_price(sample)

    print("\n=== Apple Price Prediction (Indian Market) ===")
    print(f"  Variety           : {sample['apple_variety']} ({sample['region']})")
    print(f"  Current Price     : ₹{result['current_price']:.2f}/kg")
    print(f"  Storage Cost (7d) : ₹{result['storage_cost_7d']:.2f}/kg")
    print(f"  Break-even Price  : ₹{result['breakeven_price']:.2f}/kg")
    print(f"  Predicted (7d)    : ₹{result['predicted_price_7d']:.2f}/kg")
    print(f"  Recommendation    : *** {result['recommendation']} ***")
    print(f"  Confidence        : {result['confidence']}")