Update util.py

util.py

@@ -4,11 +4,11 @@ from langchain_community.embeddings import HuggingFaceHubEmbeddings
 from langchain_community.vectorstores import Chroma
 from langchain.chains import RetrievalQA
 from langchain_google_genai import ChatGoogleGenerativeAI, GoogleGenerativeAIEmbeddings
-import shutil
+
 import git 
 
+# embeddings = HuggingFaceHubEmbeddings(model="thuan9889/llama_embedding_model_v1")
 from chromadb.utils import embedding_functions
-
 embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=os.environ['GOOGLE_API_KEY'], task_type="retrieval_query")
 
 model = ChatGoogleGenerativeAI(model="gemini-pro",google_api_key=os.environ['GOOGLE_API_KEY'],temperature=0.2,convert_system_message_to_human=True)
@@ -18,7 +18,7 @@ def get_folder_paths(directory = "githubCode"):
     for root, dirs, files in os.walk(directory):
         if '.git' in dirs:
             # Skip the directory if a .git folder is found
-            dirs.remove('.git')
+            dirs.remove('.git') 
         for dir_name in dirs:
             folder_paths.append(os.path.join(root, dir_name))
     return folder_paths
@@ -50,17 +50,19 @@ texts = text_splitter.split_text(context)
 
 vector_index = Chroma.from_texts(texts, embeddings).as_retriever(search_kwargs={"k":5})
 
-shutil.rmtree("githubCode")
+# import shutil
+# shutil.rmtree('githubCode')
+# print("Directory removed!!")
 qa_chain = RetrievalQA.from_chain_type(
     model,
     retriever=vector_index,
     return_source_documents=True
-
 )
 
 # Function to generate assistant's response using ask function
 def generate_assistant_response(question):
     answer = qa_chain({"query": question})
+    print(answer)
     return answer['result']
 
 # print(generate_assistant_response("Tell me about the instructor_embeddings function."))