Update app.py

la version précédente provoque un "runtime error" pendant le build.
Remplacement complet de la fonction "process_images"

app.py

@@ -220,79 +220,32 @@ def select_preset(preset_id):
         return jsonify({'error': str(e)}), 500
 
 @app.route('/process', methods=['POST'])
-def process_image():
-    if 'image' not in request.files:
-        return jsonify({'error': 'No image provided'}), 400
+def process_images():
+    if 'image1' not in request.files or 'image2' not in request.files:
+        return jsonify({'error': 'Two images must be provided (image1 and image2)'}), 400
 
-    #Récupère l'image envoyée, la lit avec PIL (Image.open), la convertit en RGB
     try:
-        image_file = request.files['image']                                
-        image = Image.open(io.BytesIO(image_file.read())).convert('RGB')
-        
-        # Use default parameters if none provided
-        ### params = DEFAULT_PARAMS.copy()
-        ### if request.form:
-        ###    params.update({
-        ###        'subscription': float(request.form.get('subscription', params['subscription'])),
-        ###        'num_generations': int(request.form.get('num_generations', params['num_generations'])),
-        ###        'author_share': float(request.form.get('author_share', params['author_share'])),
-        ###        'ro_share': float(request.form.get('ro_share', params['ro_share'])),
-        ###        'num_users_k': int(request.form.get('num_users_k', params['num_users_k'])),
-        ###        'num_neighbors': int(request.form.get('num_neighbors', params['num_neighbors'])),
-        ###        'num_authors': int(request.form.get('num_authors', DEFAULT_PARAMS['num_authors'])),
-        ###    })
-        
-        # Process image
-        features = get_ft(model, image)
-        distances, indices = get_topk(index, features, topk=params['num_neighbors'])
-        
-        # Collect valid results first
-           valid_results = []                         #Pour chaque image trouvée : Vérifie que l'URL est valide (avec requests.head)
-        ### valid_similarities = []                    #Si l'URL est OK, ajoute l'image à valid_results et garde sa distance
-        ### for i in range(params['num_neighbors']):
-        ###     image_url = urls[indices[0][i]].strip()
-        ###     try:
-        ###         response = requests.head(image_url)
-        ###         if response.status_code == 200:
-        ###             valid_results.append({
-        ###                 'index': i,
-        ###                 'url': image_url
-        ###             })
-        ###             valid_similarities.append(distances[0][i])  
-        ###     except requests.RequestException:
-        ###         continue
-        
-        # Renormalize similarities for valid results
-        ### if valid_similarities:
-        ###     similarities = distance_to_similarity(np.array([valid_similarities]), temp=1e-5)
-            
-            # Calculate rewards with renormalized similarities
-            ### rewards = calculate_rewards(
-            ###     params['subscription'],
-            ###     params['num_generations'],
-            ###     params['author_share'],
-            ###     params['ro_share'],
-            ###     params['num_users_k'],
-            ###     similarities,
-            ###     params['num_authors']
-            ### )
-            
-            # Build final results
-            # Pour chaque image valide, associe l'URL, la récompense calculée, et des infos supplémentaires récupérées avec get_other_info()
-            results = []
-            for i, result in enumerate(valid_results):
-                other_info = get_other_info(result['url'])
-                results.append({
-                    'image_url': result['url'],
-                    'rewards': rewards[i],
-                    'other': other_info
-                })
-            
-            return jsonify({'results': results})
-        
+        # Charger les deux images
+        image_file1 = request.files['image1']
+        image1 = Image.open(io.BytesIO(image_file1.read())).convert('RGB')
+
+        image_file2 = request.files['image2']
+        image2 = Image.open(io.BytesIO(image_file2.read())).convert('RGB')
+
+        # Extraire les features des deux images
+        features1 = get_ft(model, image1)
+        features2 = get_ft(model, image2)
+
+        # Calculer la distance euclidienne entre les deux feature vectors
+        distance = float(np.linalg.norm(features1 - features2))  # Convertir en float Python natif pour JSON
+
+        # Retourner la distance
+        return jsonify({'distance': distance})
+
     except Exception as e:
         return jsonify({'error': str(e)}), 500
 
+
 if __name__ == '__main__':
     init_model()
     app.run(host='0.0.0.0', port=7860)