Update app.py

app.py

@@ -10,7 +10,6 @@ from functools import lru_cache
 from dotenv import load_dotenv
 from fastapi import FastAPI, Request, HTTPException
 from fastapi.responses import JSONResponse
-import time
 from tqdm import tqdm
 
 urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
@@ -49,10 +48,11 @@ class ModelManager:
 
     def load_all_models(self):
         with ThreadPoolExecutor(max_workers=len(global_data['model_configs'])) as executor:
-            futures = [executor.submit(self._load_model, config) for config in tqdm(global_data['model_configs'], desc="Cargando modelos")]
-            for future in as_completed(futures):
+            futures = [executor.submit(self._load_model, config) for config in tqdm(global_data['model_configs'], desc="Loading models")]
+            for future in tqdm(as_completed(futures), total=len(global_data['model_configs']), desc="Loading models complete"):
                 future.result()
 
+
     def _load_model(self, model_config):
         model_name = model_config['name']
         cache_file = os.path.join(self.model_cache_dir, f"{model_name}.pkl")
@@ -66,7 +66,7 @@ class ModelManager:
                     with open(cache_file, "wb") as f:
                         pickle.dump(self.models[model_name], f)
             except Exception as e:
-                print(f"Error al cargar el modelo {model_name}: {e}")
+                print(f"Error loading model {model_name}: {e}")
                 self.models[model_name] = None
 
     def get_model(self, model_name):
@@ -81,8 +81,8 @@ def normalize_input(input_text):
     return input_text.strip()
 
 def remove_duplicates(text):
-    text = re.sub(r'(Hello there, how are you\? \[/INST\]){2,}', 'Hello there, how are you? [/INST]', text)
-    text = re.sub(r'(How are you\? \[/INST\]){2,}', 'How are you? [/INST]', text)
+    text = re.sub(r'(Hello there, how are you\? \[/INST\]){2,}', 'Hello there, how are you?', text)
+    text = re.sub(r'(How are you\? \[/INST\]){2,}', 'How are you?', text)
     text = text.replace('[/INST]', '')
     lines = text.split('\n')
     unique_lines = []
@@ -107,7 +107,7 @@ async def process_message(message):
 
     with ThreadPoolExecutor(max_workers=len(global_data['model_configs'])) as executor:
         futures = [executor.submit(generate_model_response, model_manager.get_model(config['name']), inputs) for config in global_data['model_configs'] if model_manager.get_model(config['name'])]
-        for i, future in enumerate(as_completed(futures)):
+        for i, future in enumerate(tqdm(as_completed(futures), total=len([f for f in futures]), desc="Generating responses")):
              model_name = global_data['model_configs'][i]['name']
              responses[model_name] = future.result()
 
@@ -122,7 +122,7 @@ async def api_generate_multimodel(request: Request):
         data = await request.json()
         message = data.get("message")
         if not message:
-            raise HTTPException(status_code=400, detail="Mensaje faltante")
+            raise HTTPException(status_code=400, detail="Missing message")
         response = await process_message(message)
         return JSONResponse({"response": response})
     except HTTPException as e: