tasks/audio.py

from fastapi import APIRouter
from datetime import datetime
from datasets import load_dataset
from sklearn.metrics import accuracy_score
import random
import os
import torch

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 = "Random Baseline"
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: Random Baseline
    - Makes random predictions from the label space (0-1)
    - 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.
    #--------------------------------------------------------------------------------------------   
    
    # Make random predictions (placeholder for actual model inference)
    def preprocess_audio(example):
    """Convert dataset into tensors."""
        waveform = torch.tensor(example["audio"]["array"], dtype=torch.float32).unsqueeze(0)  # Add batch dim
        label = torch.tensor(example["label"], dtype=torch.long)  # Ensure labels are `int64`
    return waveform, label

    model_path = "quantized_teacher_m5_static.pth"
    model, device = load_model(model_path)


    train_test = train_test.map(preprocess_audio)
    test_dataset = train_test.map(preprocess_audio)

    test_loader = DataLoader(test_dataset, batch_size=32, shuffle=True)

    
    true_labels = test_dataset["label"]
    predictions = []

    with torch.no_grad():
    for waveforms, labels in test_loader:
        waveforms, labels = waveforms.to(device), labels.to(device)
        
        # Run Model
        outputs = model(waveforms)
        predicted_label = torch.argmax(F.softmax(outputs, dim=1), dim=1)
        true_labels.extend(labels.cpu().numpy())
        predicted_labels.extend(predicted_label.cpu().numpy())
    
    #--------------------------------------------------------------------------------------------
    # YOUR MODEL INFERENCE STOPS HERE
    #--------------------------------------------------------------------------------------------   
    
    # Stop tracking emissions
    emissions_data = tracker.stop_task()
    
    # Calculate accuracy
    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