Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -1,63 +1,78 @@
|
|
1 |
import gradio as gr
|
2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
"""
|
5 |
-
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
|
6 |
-
"""
|
7 |
-
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
|
8 |
|
|
|
9 |
|
10 |
-
def
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
max_tokens,
|
15 |
-
temperature,
|
16 |
-
top_p,
|
17 |
-
):
|
18 |
-
messages = [{"role": "system", "content": system_message}]
|
19 |
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
if val[1]:
|
24 |
-
messages.append({"role": "assistant", "content": val[1]})
|
25 |
|
26 |
-
|
|
|
|
|
27 |
|
28 |
-
|
|
|
29 |
|
30 |
-
for
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
)
|
37 |
-
token = message.choices[0].delta.content
|
38 |
|
39 |
-
|
40 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
41 |
|
42 |
-
|
43 |
-
|
44 |
-
""
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
),
|
58 |
-
],
|
59 |
-
)
|
60 |
-
|
61 |
-
|
62 |
-
if __name__ == "__main__":
|
63 |
-
demo.launch()
|
|
|
1 |
import gradio as gr
|
2 |
+
import chainlit as cl
|
3 |
+
from langchain_openai import ChatOpenAI
|
4 |
+
from langchain_community.embeddings import HuggingFaceEmbeddings
|
5 |
+
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
6 |
+
from langchain_community.vectorstores import Chroma
|
7 |
+
from langchain_community.document_loaders import PyPDFLoader
|
8 |
+
from langchain.chains import ConversationalRetrievalChain
|
9 |
+
from langchain_community.chat_message_histories import ChatMessageHistory
|
10 |
+
from langchain.memory import ConversationBufferMemory
|
11 |
+
from langchain_core.prompts import PromptTemplate
|
12 |
|
13 |
+
# Your API key
|
14 |
+
open_ai_key = "sk-proj-DQPYy7NQXLkgtJDLzrijT3BlbkFJuPuWnU33xKyKxgLQauKO"
|
15 |
+
|
16 |
+
llm = ChatOpenAI(api_key=open_ai_key)
|
17 |
+
|
18 |
+
template = """Use the following pieces of information to answer the user's question.
|
19 |
+
If you don't know the answer, just say that you don't know, don't try to make up an answer.
|
20 |
+
|
21 |
+
Context: {context}
|
22 |
+
Question: {question}
|
23 |
+
|
24 |
+
Only return the helpful answer below and nothing else.
|
25 |
+
Helpful answer:
|
26 |
"""
|
|
|
|
|
|
|
27 |
|
28 |
+
prompt = PromptTemplate(template=template, input_variables=["context", "question"])
|
29 |
|
30 |
+
def process_pdf_and_ask_question(pdf_file, question):
|
31 |
+
# Load and process the PDF
|
32 |
+
loader = PyPDFLoader(pdf_file.name)
|
33 |
+
pdf_data = loader.load()
|
|
|
|
|
|
|
|
|
|
|
34 |
|
35 |
+
# Split the text into chunks
|
36 |
+
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
|
37 |
+
docs = text_splitter.split_documents(pdf_data)
|
|
|
|
|
38 |
|
39 |
+
# Create a Chroma vector store
|
40 |
+
embeddings = HuggingFaceEmbeddings(model_name="embaas/sentence-transformers-multilingual-e5-base")
|
41 |
+
db = Chroma.from_documents(docs, embeddings)
|
42 |
|
43 |
+
# Initialize message history for conversation
|
44 |
+
message_history = ChatMessageHistory()
|
45 |
|
46 |
+
# Memory for conversational context
|
47 |
+
memory = ConversationBufferMemory(
|
48 |
+
memory_key="chat_history",
|
49 |
+
output_key="answer",
|
50 |
+
chat_memory=message_history,
|
51 |
+
return_messages=True,
|
52 |
+
)
|
|
|
53 |
|
54 |
+
# Create a chain that uses the Chroma vector store
|
55 |
+
chain = ConversationalRetrievalChain.from_llm(
|
56 |
+
llm=llm,
|
57 |
+
chain_type="stuff",
|
58 |
+
retriever=db.as_retriever(),
|
59 |
+
memory=memory,
|
60 |
+
return_source_documents=False,
|
61 |
+
combine_docs_chain_kwargs={'prompt': prompt}
|
62 |
+
)
|
63 |
|
64 |
+
# Process the question
|
65 |
+
res = chain({"input": question})
|
66 |
+
return res["answer"]
|
67 |
+
|
68 |
+
def gradio_interface(pdf, question):
|
69 |
+
return process_pdf_and_ask_question(pdf, question)
|
70 |
+
|
71 |
+
# Gradio interface
|
72 |
+
gr.Interface(
|
73 |
+
fn=gradio_interface,
|
74 |
+
inputs=[gr.inputs.File(file_count="single", type="file"), gr.inputs.Textbox(lines=2, placeholder="Ask a question...")],
|
75 |
+
outputs="text",
|
76 |
+
title="PDF Q&A",
|
77 |
+
description="Upload a PDF and ask questions about it.",
|
78 |
+
).launch()
|
|
|
|
|
|
|
|
|
|
|
|
|
|