Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -6,10 +6,7 @@ from langchain_core.messages import AIMessage, HumanMessage
|
|
6 |
from langchain_community.document_loaders import WebBaseLoader
|
7 |
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
8 |
from langchain_community.vectorstores import Chroma
|
9 |
-
# from langchain_openai import OpenAIEmbeddings, ChatOpenAI
|
10 |
-
from langchain.llms import HuggingFaceHub
|
11 |
from langchain.embeddings import HuggingFaceEmbeddings
|
12 |
-
from dotenv import load_dotenv
|
13 |
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
|
14 |
from langchain.chains import create_history_aware_retriever, create_retrieval_chain
|
15 |
from langchain.chains.combine_documents import create_stuff_documents_chain
|
@@ -32,7 +29,11 @@ def get_vectorstore_from_url(url):
|
|
32 |
return vector_store
|
33 |
|
34 |
def get_context_retriever_chain(vector_store):
|
35 |
-
llm = HuggingFaceHub(repo_id = "HuggingFaceH4/zephyr-7b-beta", model_kwargs = {"temperature":0.5, "max_length":512})
|
|
|
|
|
|
|
|
|
36 |
retriever = vector_store.as_retriever()
|
37 |
|
38 |
prompt = ChatPromptTemplate.from_messages([
|
@@ -46,8 +47,11 @@ def get_context_retriever_chain(vector_store):
|
|
46 |
return retriever_chain
|
47 |
|
48 |
def get_conversational_rag_chain(retriever_chain):
|
49 |
-
llm = HuggingFaceHub(repo_id = "HuggingFaceH4/zephyr-7b-beta", model_kwargs = {"temperature":0.5, "max_length":512})
|
50 |
-
|
|
|
|
|
|
|
51 |
prompt = ChatPromptTemplate.from_messages([
|
52 |
("system", "Answer the user's questions based on the below context:\n\n{context}"),
|
53 |
MessagesPlaceholder(variable_name="chat_history"),
|
@@ -100,14 +104,10 @@ else:
|
|
100 |
|
101 |
|
102 |
# conversation
|
103 |
-
|
104 |
-
|
105 |
-
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
|
110 |
-
st.write(message.content)
|
111 |
-
elif isinstance(message, HumanMessage):
|
112 |
-
with st.chat_message("Human"):
|
113 |
-
st.write(message.content)
|
|
|
6 |
from langchain_community.document_loaders import WebBaseLoader
|
7 |
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
8 |
from langchain_community.vectorstores import Chroma
|
|
|
|
|
9 |
from langchain.embeddings import HuggingFaceEmbeddings
|
|
|
10 |
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
|
11 |
from langchain.chains import create_history_aware_retriever, create_retrieval_chain
|
12 |
from langchain.chains.combine_documents import create_stuff_documents_chain
|
|
|
29 |
return vector_store
|
30 |
|
31 |
def get_context_retriever_chain(vector_store):
|
32 |
+
# llm = HuggingFaceHub(repo_id = "HuggingFaceH4/zephyr-7b-beta", model_kwargs = {"temperature":0.5, "max_length":512})
|
33 |
+
llm = HuggingFaceEndpoint(
|
34 |
+
repo_id="HuggingFaceH4/zephyr-7b-beta",
|
35 |
+
task="text-generation",
|
36 |
+
max_new_tokens=512)
|
37 |
retriever = vector_store.as_retriever()
|
38 |
|
39 |
prompt = ChatPromptTemplate.from_messages([
|
|
|
47 |
return retriever_chain
|
48 |
|
49 |
def get_conversational_rag_chain(retriever_chain):
|
50 |
+
# llm = HuggingFaceHub(repo_id = "HuggingFaceH4/zephyr-7b-beta", model_kwargs = {"temperature":0.5, "max_length":512})
|
51 |
+
llm = HuggingFaceEndpoint(
|
52 |
+
repo_id="HuggingFaceH4/zephyr-7b-beta",
|
53 |
+
task="text-generation",
|
54 |
+
max_new_tokens=512)
|
55 |
prompt = ChatPromptTemplate.from_messages([
|
56 |
("system", "Answer the user's questions based on the below context:\n\n{context}"),
|
57 |
MessagesPlaceholder(variable_name="chat_history"),
|
|
|
104 |
|
105 |
|
106 |
# conversation
|
107 |
+
for message in st.session_state.chat_history:
|
108 |
+
if isinstance(message, AIMessage):
|
109 |
+
with st.chat_message("AI"):
|
110 |
+
st.write(message.content)
|
111 |
+
elif isinstance(message, HumanMessage):
|
112 |
+
with st.chat_message("Human"):
|
113 |
+
st.write(message.content)
|
|
|
|
|
|
|
|