v 0.1

- добавлен llama-index

app.py

@@ -18,14 +18,18 @@ logger = logging.getLogger(__name__)
 logger.info("Запускаемся... 🥳🥳🥳")
 
 app = FastAPI()
+
 model_url = "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct-GGUF/resolve/main/qwen2.5-7b-instruct-q3_k_m.gguf"
+
 SYSTEM_PROMPT = ''
+
 tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-7B-Instruct")
+
 set_global_tokenizer(
     AutoTokenizer.from_pretrained("Qwen/Qwen2.5-7B-Instruct").encode
 )
 
-embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")
+embed_model = HuggingFaceEmbedding(model_name="sentence-transformers/all-mpnet-base-v2")
 
 documents = SimpleDirectoryReader("./data/").load_data()
 
@@ -94,8 +98,12 @@ async def set_system_prompt(text: str):
 async def predict(text: str):
     # Генерация ответа с помощью модели
     logger.info('post/predict')
+    logger.info('ЗАПРОС:')
+    logger.info(text)
     # response = llm.complete(text)
     response = query_engine.query(text)
+    logger.info('ОТВЕТ:')
+    logger.info(response)
     # text_response = generate_response(response)
     # return {"response": text_response}
     return {"response": response}