Update app.py

app.py

@@ -72,7 +72,8 @@ def load_data():
 @st.cache_resource
 def load_model(_docs):
     #embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-large",model_kwargs={"device":DEVICE})
-    embeddings = HuggingFaceInstructEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2",model_kwargs={"device":DEVICE})
+    #embeddings = HuggingFaceInstructEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2",model_kwargs={"device":DEVICE})
+    embeddings = HuggingFaceInstructEmbeddings(model_name="/home/user/app/all-MiniLM-L6-v2/",model_kwargs={"device":DEVICE})
     print(DEVICE)
 
     text_splitter = RecursiveCharacterTextSplitter(chunk_size=1024, chunk_overlap=256)
@@ -86,7 +87,8 @@ def load_model(_docs):
     print('db done')
     
     
-    model_name_or_path = "TheBloke/Llama-2-13B-chat-GPTQ"
+    #model_name_or_path = "TheBloke/Llama-2-13B-chat-GPTQ"
+    model_name_or_path = "/home/user/app/Llama-2-13B-chat-GPTQ/"
     model_basename = "model"
 
     tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)