Spaces:
Paused
Paused
code update
Browse files
app.py
CHANGED
@@ -2,12 +2,13 @@ import io
|
|
2 |
import streamlit as st
|
3 |
from PyPDF2 import PdfReader
|
4 |
from langchain.text_splitter import CharacterTextSplitter
|
5 |
-
from langchain.embeddings.huggingface import HuggingFaceEmbeddings
|
6 |
-
from langchain.vectorstores import FAISS
|
7 |
from langchain.chains.question_answering import load_qa_chain
|
8 |
-
from
|
|
|
|
|
|
|
9 |
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
|
10 |
-
|
11 |
|
12 |
# Global variables
|
13 |
knowledge_base = None
|
@@ -38,11 +39,9 @@ def create_knowledge_base(chunks):
|
|
38 |
def load_model():
|
39 |
model_name = "halyn/gemma2-2b-it-finetuned-paperqa"
|
40 |
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
41 |
-
|
42 |
-
model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path)
|
43 |
-
model = PeftModel.from_pretrained(model, model_name)
|
44 |
-
|
45 |
return pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=150, temperature=0.1)
|
|
|
46 |
# QA 체인 설정
|
47 |
def setup_qa_chain():
|
48 |
global qa_chain
|
@@ -99,7 +98,6 @@ def main_page():
|
|
99 |
st.error(f"Failed to process the PDF: {str(e)}")
|
100 |
|
101 |
|
102 |
-
|
103 |
# 채팅 페이지 UI
|
104 |
def chat_page():
|
105 |
st.title(f"Ask anything about {st.session_state.paper_name}")
|
|
|
2 |
import streamlit as st
|
3 |
from PyPDF2 import PdfReader
|
4 |
from langchain.text_splitter import CharacterTextSplitter
|
|
|
|
|
5 |
from langchain.chains.question_answering import load_qa_chain
|
6 |
+
from langchain_community.embeddings import HuggingFaceEmbeddings
|
7 |
+
from langchain_community.vectorstores import FAISS
|
8 |
+
from langchain_community.llms import HuggingFacePipeline
|
9 |
+
|
10 |
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
|
11 |
+
|
12 |
|
13 |
# Global variables
|
14 |
knowledge_base = None
|
|
|
39 |
def load_model():
|
40 |
model_name = "halyn/gemma2-2b-it-finetuned-paperqa"
|
41 |
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
42 |
+
model = AutoModelForCausalLM.from_pretrained(model_name)
|
|
|
|
|
|
|
43 |
return pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=150, temperature=0.1)
|
44 |
+
|
45 |
# QA 체인 설정
|
46 |
def setup_qa_chain():
|
47 |
global qa_chain
|
|
|
98 |
st.error(f"Failed to process the PDF: {str(e)}")
|
99 |
|
100 |
|
|
|
101 |
# 채팅 페이지 UI
|
102 |
def chat_page():
|
103 |
st.title(f"Ask anything about {st.session_state.paper_name}")
|