Update doc_preprocessing.py

doc_preprocessing.py

@@ -6,6 +6,8 @@ import streamlit as st
 import numpy as np
 import os
 
+emb_model = "google/Gemma-Embeddings-v1.0"
+
 def extract_text(file):
     text = ""
     # Check if the input is a file path (string) or a file-like object
@@ -64,7 +66,7 @@ def get_embeddings(texts)-> np.ndarray:
         # embeddings = embedding_model(texts)
 
 
-        model = SentenceTransformer("sujet-ai/Marsilia-Embeddings-FR-Base", trust_remote_code=True)
+        model = SentenceTransformer(emb_model, trust_remote_code=True)
         embeddings = model.encode(texts)
 
         print(f"Generated {len(embeddings)} embeddings.")