Reduce model size

app.py

@@ -38,19 +38,16 @@ def getVectorIndex(docs):
     Settings.chunk_size = 512
     index_set = {}
 
-    storage_context = StorageContext.from_defaults(persist_dir=f"./storage/book_data")
-    cur_index = load_index_from_storage(
-        storage_context,embed_model = getEmbedModel()
-    )
-    #cur_index = VectorStoreIndex.from_documents(docs, embed_model = getEmbedModel(), storage_context=storage_context)
-    #storage_context.persist(persist_dir=f"./storage/book_data")
+    storage_context = StorageContext.from_defaults()
+    cur_index = VectorStoreIndex.from_documents(docs, embed_model = getEmbedModel(), storage_context=storage_context)
+    storage_context.persist(persist_dir=f"./storage/book_data")
     return cur_index
 
 
 def getLLM():
 
-    model_path = "NousResearch/Llama-2-13b-chat-hf"
-    # model_path = "meta-llama/Llama-2-13b-chat-hf"
+    # model_path = "NousResearch/Llama-2-13b-chat-hf"
+    model_path = "NousResearch/Llama-2-7b-chat-hf"
 
     llm = HuggingFaceLLM(
     context_window=3900,