Update app.py

app.py

@@ -35,24 +35,7 @@ if 'history' not in st.session_state:
 if 'feedback' not in st.session_state:
     st.session_state.feedback = {}
 
-@st.cache_data
-def generate_embedding(_model, _tokenizer, text):
-    inputs = _tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=512)
-    if torch.cuda.is_available():
-        inputs = {k: v.to('cuda') for k, v in inputs.items()}
-    with torch.no_grad():
-        outputs = _model.encoder(**inputs)
-    embedding = outputs.last_hidden_state.mean(dim=1).squeeze()
-    if torch.cuda.is_available():
-        embedding = embedding.cpu()
-    return embedding.numpy()
 
-#error handling
-try:
-    query_embedding = generate_embedding(model, tokenizer, user_query)
-except Exception as e:
-    st.error(f"Error generating embedding: {str(e)}")
-    st.stop()
     
    
 
@@ -95,6 +78,25 @@ def load_data_and_model():
     
     tokenizer, model = load_model_and_tokenizer()
 
+@st.cache_data
+def generate_embedding(_model, _tokenizer, text):
+    inputs = _tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=512)
+    if torch.cuda.is_available():
+        inputs = {k: v.to('cuda') for k, v in inputs.items()}
+    with torch.no_grad():
+        outputs = _model.encoder(**inputs)
+    embedding = outputs.last_hidden_state.mean(dim=1).squeeze()
+    if torch.cuda.is_available():
+        embedding = embedding.cpu()
+    return embedding.numpy()
+
+#error handling
+try:
+    query_embedding = generate_embedding(model, tokenizer, user_query)
+except Exception as e:
+    st.error(f"Error generating embedding: {str(e)}")
+    st.stop()
+
     # Precompute embeddings with GPU support
     @st.cache_data
     def generate_embedding(text):