rebase on master

app.py

@@ -1,59 +1,13 @@
-<<<<<<< HEAD
-import logging
-import sys
-
-import streamlit as st
-from dotenv import load_dotenv
-
-from xpipe_wiki.manager_factory import XPipeRobotManagerFactory, XPipeRobotRevision
-
-logging.basicConfig(
-    stream=sys.stdout, level=logging.INFO
-)  # logging.DEBUG for more verbose output
-logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))
-
-# Sidebar contents
-with st.sidebar:
-    st.title("🤗💬 LLM Chat App")
-    st.markdown(
-        """
-    ## About
-    This app is an LLM-powered chatbot built using:
-    - [Streamlit](https://streamlit.io/)
-    - [LangChain](https://python.langchain.com/)
-    - [X-Pipe](https://github.com/ctripcorp/x-pipe)
-    """
-    )
-    # add_vertical_space(5)
-    st.write("Made by Nick")
-
-
-def main() -> None:
-    st.header("X-Pipe Wiki 机器人 💬")
-    robot_manager = XPipeRobotManagerFactory.get_or_create(
-        XPipeRobotRevision.HUGGINGFACE_VERSION_0
-    )
-    robot = robot_manager.get_robot()
-    query = st.text_input("X-Pipe Wiki 问题:")
-    if query:
-        response = robot.ask(question=query)
-        st.write(response)
-
-
-if __name__ == "__main__":
-    load_dotenv()
-    main()
-=======
 import logging
 import sys
 
 import streamlit as st
+from dotenv import load_dotenv
 
 from xpipe_wiki.manager_factory import XPipeRobotManagerFactory, XPipeRobotRevision
-from xpipe_wiki.robot_manager import XPipeWikiRobot, AzureOpenAIXPipeWikiRobot
 
 logging.basicConfig(
-    stream=sys.stdout, level=logging.DEBUG
+    stream=sys.stdout, level=logging.INFO
 )  # logging.DEBUG for more verbose output
 logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))
 
@@ -76,7 +30,7 @@ with st.sidebar:
 def main() -> None:
     st.header("X-Pipe Wiki 机器人 💬")
     robot_manager = XPipeRobotManagerFactory.get_or_create(
-        XPipeRobotRevision.SIMPLE_OPENAI_VERSION_0
+        XPipeRobotRevision.HUGGINGFACE_VERSION_0
     )
     robot = robot_manager.get_robot()
     query = st.text_input("X-Pipe Wiki 问题:")
@@ -86,5 +40,5 @@ def main() -> None:
 
 
 if __name__ == "__main__":
+    load_dotenv()
     main()
->>>>>>> 2bf20e0 (Update app.py)

llama/context.py

@@ -1,4 +1,5 @@
 from abc import abstractmethod, ABC
+from typing import List
 
 from llama_index import ServiceContext, LLMPredictor, LangchainEmbedding, Document
 from llama_index import StorageContext
@@ -8,7 +9,6 @@ from langchain_manager.manager import BaseLangChainManager
 
 
 class ServiceContextManager(Lifecycle, ABC):
-
     @abstractmethod
     def get_service_context(self) -> ServiceContext:
         pass
@@ -40,23 +40,33 @@ class AzureServiceContextManager(ServiceContextManager):
         )
 
     def do_start(self) -> None:
-        self.logger.info("[do_start][embedding] last used usage: %d",
-                         self.service_context.embed_model.total_tokens_used)
-        self.logger.info("[do_start][predict] last used usage: %d",
-                         self.service_context.llm_predictor.total_tokens_used)
+        self.logger.info(
+            "[do_start][embedding] last used usage: %d",
+            self.service_context.embed_model.total_tokens_used,
+        )
+        self.logger.info(
+            "[do_start][predict] last used usage: %d",
+            self.service_context.llm_predictor.total_tokens_used,
+        )
 
     def do_stop(self) -> None:
-        self.logger.info("[do_stop][embedding] last used usage: %d",
-                         self.service_context.embed_model.total_tokens_used)
-        self.logger.info("[do_stop][predict] last used usage: %d",
-                         self.service_context.llm_predictor.total_tokens_used)
+        self.logger.info(
+            "[do_stop][embedding] last used usage: %d",
+            self.service_context.embed_model.total_tokens_used,
+        )
+        self.logger.info(
+            "[do_stop][predict] last used usage: %d",
+            self.service_context.llm_predictor.total_tokens_used,
+        )
 
     def do_dispose(self) -> None:
-        self.logger.info("[do_dispose] total used token: %d", self.service_context.llm_predictor.total_tokens_used)
+        self.logger.info(
+            "[do_dispose] total used token: %d",
+            self.service_context.llm_predictor.total_tokens_used,
+        )
 
 
 class StorageContextManager(Lifecycle, ABC):
-
     @abstractmethod
     def get_storage_context(self) -> StorageContext:
         pass
@@ -65,9 +75,11 @@ class StorageContextManager(Lifecycle, ABC):
 class LocalStorageContextManager(StorageContextManager):
     storage_context: StorageContext
 
-    def __init__(self,
-                 dataset_path: str = "./dataset",
-                 service_context_manager: ServiceContextManager = None) -> None:
+    def __init__(
+        self,
+        service_context_manager: ServiceContextManager,
+        dataset_path: str = "./dataset",
+    ) -> None:
         super().__init__()
         self.dataset_path = dataset_path
         self.service_context_manager = service_context_manager
@@ -77,8 +89,11 @@ class LocalStorageContextManager(StorageContextManager):
 
     def do_init(self) -> None:
         from llama.utils import is_local_storage_files_ready
+
         if is_local_storage_files_ready(self.dataset_path):
-            self.storage_context = StorageContext.from_defaults(persist_dir=self.dataset_path)
+            self.storage_context = StorageContext.from_defaults(
+                persist_dir=self.dataset_path
+            )
         else:
             docs = self._download()
             self._indexing(docs)
@@ -94,14 +109,17 @@ class LocalStorageContextManager(StorageContextManager):
     def do_dispose(self) -> None:
         self.storage_context.persist(self.dataset_path)
 
-    def _download(self) -> [Document]:
+    def _download(self) -> List[Document]:
         from llama.data_loader import GithubLoader
+
         loader = GithubLoader()
         return loader.load()
 
-    def _indexing(self, docs: [Document]):
+    def _indexing(self, docs: List[Document]) -> None:
         from llama_index import GPTVectorStoreIndex
-        index = GPTVectorStoreIndex.from_documents(docs,
-                                                   service_context=self.service_context_manager.get_service_context())
+
+        index = GPTVectorStoreIndex.from_documents(
+            docs, service_context=self.service_context_manager.get_service_context()
+        )
         index.storage_context.persist(persist_dir=self.dataset_path)
         self.storage_context = index.storage_context

llama/data_loader.py

@@ -16,10 +16,10 @@ class WikiLoader(ABC):
 
 class GithubLoader(WikiLoader):
     def __init__(
-            self,
-            github_owner: Optional[str] = None,
-            repo: Optional[str] = None,
-            dirs: Optional[Sequence[str]] = None,
+        self,
+        github_owner: Optional[str] = None,
+        repo: Optional[str] = None,
+        dirs: Optional[Sequence[str]] = None,
     ):
         super().__init__()
         self.owner = (

requirements.txt

@@ -4,4 +4,5 @@ streamlit
 ruff
 black
 mypy
-accelerate
+accelerate
+python-dotenv

xpipe_wiki/robot_manager.py

@@ -59,10 +59,6 @@ class AzureXPipeWikiRobotManager(XPipeWikiRobotManager):
         index = load_index_from_storage(
             storage_context=self.storage_context_manager.get_storage_context(),
             service_context=self.service_context_manager.get_service_context(),
-<<<<<<< HEAD
-=======
-
->>>>>>> 13f45af (Update xpipe_wiki/robot_manager.py)
         )
         self.query_engine = index.as_query_engine(
             service_context=self.service_context_manager.get_service_context()