Update tasks/audio.py

tasks/audio.py

@@ -9,7 +9,13 @@ from .utils.evaluation import AudioEvaluationRequest
 from .utils.emissions import tracker, clean_emissions_data, get_space_info
 
 from dotenv import load_dotenv
-print("here I am")
+import logging
+
+# Configurer le logging
+logging.basicConfig(level=logging.INFO)
+
+# Utiliser le logging au lieu de print
+logging.info("Début du fichier python")
 load_dotenv()
 
 router = APIRouter()
@@ -40,9 +46,9 @@ async def evaluate_audio(request: AudioEvaluationRequest):
     }
     # Load and prepare the dataset
     # Because the dataset is gated, we need to use the HF_TOKEN environment variable to authenticate
-    print("here I am 2")
+    logging.info("Chargement des données")
     dataset = load_dataset(request.dataset_name,token=os.getenv("HF_TOKEN"))
-    print("here I am 3")
+    logging.info("Données chargées")
     # Split dataset
     train_test = dataset["train"]
     test_dataset = dataset["test"]
@@ -70,10 +76,10 @@ async def evaluate_audio(request: AudioEvaluationRequest):
     
     # Correctly access the audio data
     audio_data = [example["array"] for example in dataset["test"]["audio"]]
-    print("here I am prediction")
+    logging.info("Début des prédictions")
     # Prédiction sur tout le dataset
     results = classifier(audio_data, batch_size=8)
-    
+    logging.info("Fin des prédictions")
     predictions = []
     for result in results:
         # Check if result is a dictionary
@@ -118,5 +124,5 @@ async def evaluate_audio(request: AudioEvaluationRequest):
             "test_seed": request.test_seed
         }
     }
-    
+    logging.info("Returning results")
     return results