Backend/main.py

import os
import sys
import joblib 
import pandas as pd 
from fastapi import FastAPI, HTTPException

# Add the root directory to the system path so Python can find the 'backend' folder
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from backend.schema.user_input import WorkoutFeatures

MODEL_PATH = "backend/deploy_model.joblib"

try:
    # Look for the file in the same folder as main.py
    model = joblib.load(MODEL_PATH)
    print("✅ BACKEND: Model loaded successfully!")
except Exception as e:
    print(f"❌ BACKEND ERROR: Could not load model. Reason: {e}")
    model = None

app = FastAPI()

@app.post("/predict")
def predict_workout(data: WorkoutFeatures):
    if model is None:
        raise HTTPException(status_code=500, detail="Model not loaded on server.")

    # 1. Process Diet Columns (Same logic as before)
    diet_columns = ['diet_type_Keto', 'diet_type_Low-Carb', 'diet_type_Paleo', 'diet_type_Vegan', 'diet_type_Vegetarian']
    diet_values = [1 if col == f"diet_type_{data.diet_type}" else 0 for col in diet_columns]

    # 2. Build the exact feature list your model expects
    input_list = [
        data.Age, data.Session_Duration_hours, data.Calories_Burned,
        data.Fat_Percentage, data.Water_Intake_liters, data.workout_frequency,
        data.experience_level, data.bmi, data.daily_meals_frequency,
        data.carbs, data.proteins, data.fats, data.calories, data.Gender, 
        *diet_values
    ]

    try:
        # 3. Local Prediction (No more 'requests.post'!)
        # We wrap in [input_list] because sklearn models expect a 2D array
        prediction = model.predict([input_list])
        
        # Convert to float to make it JSON serializable
        return {"prediction": float(prediction[0])}

    except Exception as e:
        raise HTTPException(status_code=500, detail=f"Prediction Error: {str(e)}")