from fastapi import APIRouter
from datetime import datetime
from datasets import load_dataset
from sklearn.metrics import accuracy_score
import random
import os
import librosa
import joblib
import numpy as np

from .utils.evaluation import AudioEvaluationRequest
from .utils.emissions import tracker, clean_emissions_data, get_space_info

from dotenv import load_dotenv
load_dotenv()

router = APIRouter()

DESCRIPTION = "LGBM Classifier Baseline on Mel-frequency cepstral coefficients"
ROUTE = "/audio"



@router.post(ROUTE, tags=["Audio Task"],
             description=DESCRIPTION)
async def evaluate_audio(request: AudioEvaluationRequest):
    """
    Evaluate audio classification for rainforest sound detection.
    
    Current Model: LGBM
    - Used as a baseline for comparison
    """
    # Get space info
    username, space_url = get_space_info()

    # Define the label mapping
    LABEL_MAPPING = {
        "chainsaw": 0,
        "environment": 1
    }
    # Load and prepare the dataset
    # Because the dataset is gated, we need to use the HF_TOKEN environment variable to authenticate
    dataset = load_dataset(request.dataset_name,token=os.getenv("HF_TOKEN"))
    
    # Split dataset
    train_test = dataset["train"].train_test_split(test_size=request.test_size, seed=request.test_seed)
    test_dataset = train_test["test"]
    
    # Start tracking emissions
    tracker.start()
    tracker.start_task("inference")
    
    #--------------------------------------------------------------------------------------------
    # YOUR MODEL INFERENCE CODE HERE
    # Update the code below to replace the random baseline by your model inference within the inference pass where the energy consumption and emissions are tracked.
    #--------------------------------------------------------------------------------------------   
    
    def compute_mfcc(row, sr):
        audio_array = row['audio']['array']
        mfcc = librosa.feature.mfcc(y=audio_array, sr=sr, n_mfcc=5)
        return np.mean(mfcc, axis=1)

    def predict_new_audio(model, dataset, sr):
        list_mfcc = [extract_features(row, sr) for row in dataset]
        array_mfcc = np.vstack(list_mfcc)
        predictions = model.predict(array_mfcc)
        return predictions

    model_filename = "models/lightgbm_baseline_87_acc.pkl"
    clf = joblib.load(model_filename)

    predictions = predict_new_audio(clf, test_dataset, 12000)
    
    #--------------------------------------------------------------------------------------------
    # YOUR MODEL INFERENCE STOPS HERE
    #--------------------------------------------------------------------------------------------   
    
    # Stop tracking emissions
    emissions_data = tracker.stop_task()
    
    # Calculate accuracy
    true_labels = test_dataset["label"]
    accuracy = accuracy_score(true_labels, predictions)
    
    # Prepare results dictionary
    results = {
        "username": username,
        "space_url": space_url,
        "submission_timestamp": datetime.now().isoformat(),
        "model_description": DESCRIPTION,
        "accuracy": float(accuracy),
        "energy_consumed_wh": emissions_data.energy_consumed * 1000,
        "emissions_gco2eq": emissions_data.emissions * 1000,
        "emissions_data": clean_emissions_data(emissions_data),
        "api_route": ROUTE,
        "dataset_config": {
            "dataset_name": request.dataset_name,
            "test_size": request.test_size,
            "test_seed": request.test_seed
        }
    }
    
    return results