ajout nouveau type cache

.gitignore

@@ -19,4 +19,4 @@ app/__pycache__/
 *.log
 
 #cache des heatmaps
- cache_heatmaps/
+cache/

Dockerfile

@@ -13,7 +13,7 @@ ENV HOME=/home/user
 ENV PATH=/home/user/.local/bin:$PATH
 
 # Créer la structure de dossiers complète avec les bonnes permissions (MODIFIÉ)
-RUN mkdir -p /home/user/app/cache_heatmaps && \
+RUN mkdir -p /home/user/app/cache && \
     chown -R user:user /home/user/app && \
     chmod -R 755 /home/user/app
 

app/model.py

@@ -21,13 +21,12 @@ from tf_keras_vis.utils import normalize
 import logging
 import time
 import os
-
+import diskcache as dc
 from typing import TypedDict, Callable, Any
 
-HEATMAP_CACHE = {}
-CACHE_DIR = "./cache_heatmaps"
+CACHE_DIR = './cache'
 os.makedirs(CACHE_DIR, exist_ok=True)
-
+cache = dc.Cache(CACHE_DIR)
 
 logging.basicConfig(
     level=logging.INFO,  # ou logging.DEBUG
@@ -336,26 +335,17 @@ def hash_image_bytes(image_bytes):
 def get_heatmap(config, image_bytes: bytes, predicted_class_index):
     result = {}
     try:
-        image_hash = hash_image_bytes(image_bytes)
-        cache_key = f"{image_hash}_{predicted_class_index}"
+        hash_key = hash_image_bytes(image_bytes)
+        heatmap_key = f"{hash_key}_heatmap"
 
-        # Vérification cache mémoire d'abord
-        if cache_key in HEATMAP_CACHE:
-            logger.info(f"✅ Heatmap trouvée dans cache mémoire pour {cache_key}")
-            result["heatmap"] = HEATMAP_CACHE[cache_key]
+       # Vérification cache mémoire d'abord
+        if heatmap_key in cache:
+            logger.info(f"✅ Heatmap trouvée dans le cache {heatmap_key}")
+            result["heatmap"] = cache[heatmap_key]
             return result
 
-        # Vérification cache disque ensuite
-        cache_file_path = os.path.join(CACHE_DIR, f"{cache_key}.pkl")
-        if os.path.exists(cache_file_path):
-            logger.info(f"✅ Heatmap trouvée sur disque pour {cache_key}")
-            with open(cache_file_path, "rb") as f:
-                cached_heatmap = pickle.load(f)
-            result["heatmap"] = cached_heatmap
-            # On remet aussi en mémoire pour accélérer prochaines requêtes
-            HEATMAP_CACHE[cache_key] = cached_heatmap
-            return result
 
+        #
         # Calcul si non trouvé dans le cache
         _, raw_input = preprocess_image(
             image_bytes, config["target_size"], config["preprocess_input"]
@@ -377,14 +367,7 @@ def get_heatmap(config, image_bytes: bytes, predicted_class_index):
         # Conversion en liste pour le JSON
         heatmap_list = heatmap.tolist()
         result["heatmap"] = heatmap_list
-
-        # Sauvegarde dans cache mémoire
-        HEATMAP_CACHE[cache_key] = heatmap_list
-
-        # Sauvegarde sur disque
-        with open(cache_file_path, "wb") as f:
-            pickle.dump(heatmap_list, f)
-
+        cache[heatmap_key] = heatmap_list
     except Exception as e:
         logger.error(f"❌ Erreur lors de la génération de la heatmap: {e}")
         result["heatmap"] = []
@@ -418,10 +401,17 @@ def get_heatmap_old(config, image_bytes: bytes,predicted_class_index):
         result["heatmap"] = []
     return result
 
-def predict_with_model(config, image_bytes: bytes):
-   
-    #input_array,raw_input = preprocess_image(image_bytes,config["target_size"],config["preprocess_input"])
 
+
+
+def predict_with_cache(config, image_bytes: bytes):
+    hash_key = hash_image_bytes(image_bytes)
+    pred_key = f"{hash_key}_pred"
+
+    if pred_key in cache:
+        logger.info(f"✅ prédiction trouvée dans le cache {hash_key}")
+        return cache[pred_key]
+    
     try:
         logger.info("📤 Lecture des bytes et conversion en image PIL")
         image = Image.open(io.BytesIO(image_bytes)).convert("RGB")
@@ -436,13 +426,18 @@ def predict_with_model(config, image_bytes: bytes):
     confidence = float(np.max(preds))
     entropy=float(compute_entropy_safe(preds))
     logger.info(f"✅ Prédiction : classe={predicted_class_index}, confiance={confidence:.4f},entropy={entropy:.4f}")
-
-    return  {
+    result={
         "preds": preds.tolist(),
         "predicted_class": predicted_class_index,
         "confidence": confidence,
         "entropy":entropy
     }
+    cache[pred_key] = result
+    return  
+
+def predict_with_model(config, image_bytes: bytes):
+   return predict_with_cache(config, image_bytes)
+   
 
 
 def predict_with_model_old(config, image_bytes: bytes):

requirements.txt

@@ -95,3 +95,4 @@ Werkzeug==3.1.3
 wheel==0.45.1
 wrapt==1.17.2
 yarl==1.20.1
+diskcache