Update app.py

app.py

@@ -3,7 +3,9 @@ import streamlit as st
 import pandas as pd
 import numpy as np
 from tqdm.auto import tqdm
-from sentence_transformers import SentenceTransformer
+#from sentence_transformers import SentenceTransformer
+from transformers import AutoTokenizer, AutoModel
+import torch
 
 dataList = [
     {"Answer": "", "Distance": 0},
@@ -27,9 +29,12 @@ def load_documents_from_jsonl(embeddings_model, jsonl_path, createEmbeddings=Fal
 
     return df
        
-def generate_embeddings(model, text):
+def generate_embeddings(tokenizer, model, text):
+#    with torch.no_grad():
+#        embeddings = model.encode(text, convert_to_tensor=True)
+    encoded_input = tokenizer(text, padding=True, truncation=True, return_tensors='pt')
     with torch.no_grad():
-        embeddings = model.encode(text, convert_to_tensor=True)
+        embeddings = model(**encoded_input)
     return embeddings.cpu().numpy()
 
 def save_to_faiss(df):
@@ -56,7 +61,9 @@ def main():
     # Заголовок приложения
     st.title("Demo for LLAMA-2 RAG with CPU only")
 
-    model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')
+    #model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')
+    tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')
+    model = AutoModel.from_pretrained('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')
 
     df_qa = load_documents_from_jsonl('ExportForAI1.jsonl', model, False)     
     save_to_faiss(df_qa)