import mlflow 
import uvicorn
import pandas as pd 
from pydantic import BaseModel
from typing import Literal, List, Union
from fastapi import FastAPI, File, UploadFile, HTTPException
import joblib
import traceback 
import logging
import os

# Création de l'application FastAPI

app = FastAPI()

# MLFLOW URI

# Set your variables for your environment
#EXPERIMENT_NAME="hyperparameter_tuning"

# Set tracking URI to your Hugging Face application
#mlflow.set_tracking_uri(os.environ["APP_URI"])

# Set experiment's info 
#mlflow.set_experiment(EXPERIMENT_NAME)

# Get our experiment info
#experiment = mlflow.get_experiment_by_name(EXPERIMENT_NAME)

# Exemple de données 
df = pd.DataFrame({
                "MedInc": [8.3252],
                "HouseAge": [41.0],                         
                "AveRooms": [6.984126984126984],
                "AveBedrms": [1.0238095238095237],
                "Population": [322.0],
                "AveOccup": [2.5555555555555554],
                "Latitude": [37.88],
                "Longitude": [-122.23],
                            })

class PredictionFeatures(BaseModel):
    MedInc: float
    HouseAge: float
    AveRooms: float
    AveBedrms: float
    Population: float
    AveOccup: float
    Latitude: float
    Longitude: float

@app.get("/")
async def root():
    logging.debug("Request")
    return {"message": "L'application FastAPI est déployée avec succès !"}

@app.post("/predict", tags=["Machine Learning"])
async def predict(predictionFeatures: PredictionFeatures):
    """
    Prédiction du prix d'une maison.
    """
    try:
    # Préparer les caractéristiques d'entrée pour la prédiction
        caracts = pd.DataFrame({"MedInc": [predictionFeatures.MedInc],
                            "HouseAge": [predictionFeatures.HouseAge],
                            "AveRooms": [predictionFeatures.AveRooms],
                            "AveBedrms": [predictionFeatures.AveBedrms],
                            "Population": [predictionFeatures.Population],
                            "AveOccup": [predictionFeatures.AveOccup],
                            "Latitude": [predictionFeatures.Latitude],
                            "Longitude": [predictionFeatures.Longitude],
                            })

    # Charger le modèle depuis mlflow
        logged_model = 'runs:/0abb54594cf24179b220ff2e5acff9d7/best_estimator'  # Remplace par ton propre run ID
        loaded_model = mlflow.pyfunc.load_model(logged_model)

    # Faire la prédiction
        prediction_price = loaded_model.predict(caracts)

    # Formater la réponse
        response = {"prediction": prediction_price.tolist()[0]}
        return response

    except Exception as e:
    # Afficher l'erreur complète dans les logs
        error_message = str(e) + "\n" + traceback.format_exc()
        raise HTTPException(status_code=500, detail=error_message)
    
print("Le model est là !")