Spaces:
Running
Running
File size: 5,330 Bytes
180ada1 056867b 180ada1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 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 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 |
import gradio as gr
import PyPDF2
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.embeddings import HuggingFaceBgeEmbeddings
from langchain.vectorstores import Chroma
from langchain.memory import ChatMessageHistory, ConversationBufferMemory
from langchain_groq import ChatGroq
from langchain.chains import ConversationalRetrievalChain
from langchain_community.document_loaders import WebBaseLoader
import os
# Function to process text and create ConversationalRetrievalChain
def process_text_and_create_chain(text):
text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
texts = text_splitter.split_text(text)
metadatas = [{"source": f"{i}-pl"} for i in range(len(texts))]
model_name = "BAAI/bge-small-en"
model_kwargs = {"device": "cpu"}
encode_kwargs = {"normalize_embeddings": True}
hf = HuggingFaceBgeEmbeddings(
model_name=model_name, model_kwargs=model_kwargs, encode_kwargs=encode_kwargs
)
db = Chroma.from_texts(texts, hf, metadatas=metadatas)
message_history = ChatMessageHistory()
memory = ConversationBufferMemory(
memory_key="chat_history",
output_key="answer",
chat_memory=message_history,
return_messages=True,
)
llm_groq = ChatGroq(
groq_api_key="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",
model_name='mixtral-8x7b-32768'
)
chain = ConversationalRetrievalChain.from_llm(
llm=llm_groq,
chain_type="stuff",
retriever=db.as_retriever(),
memory=memory,
return_source_documents=True,
)
return chain
# Initialize global variables
global_chain = None
# Function to handle PDF upload
def handle_pdf_upload(file):
if file is None:
return "No file uploaded. Please upload a PDF file.", gr.update(visible=False), gr.update(visible=True)
if not file.name.lower().endswith('.pdf'):
return "Error: Please upload a PDF file.", gr.update(visible=False), gr.update(visible=True)
try:
print(f"Processing file: {file.name}")
pdf_reader = PyPDF2.PdfReader(file.name)
pdf_text = ""
for page in pdf_reader.pages:
pdf_text += page.extract_text()
global global_chain
global_chain = process_text_and_create_chain(pdf_text)
return "PDF processed successfully.", gr.update(visible=True), gr.update(visible=False)
except Exception as e:
print(f"Error processing PDF: {str(e)}")
return f"Error processing PDF: {str(e)}", gr.update(visible=False), gr.update(visible=True)
# Function to handle link input
def handle_link_input(link):
try:
loader = WebBaseLoader(link)
data = loader.load()
doc = "\n".join([doc.page_content for doc in data])
global global_chain
global_chain = process_text_and_create_chain(doc)
return "Link processed successfully.", gr.update(visible=True), gr.update(visible=False)
except Exception as e:
print(f"Error processing link: {str(e)}")
return f"Error processing link: {str(e)}", gr.update(visible=False), gr.update(visible=True)
# Function to handle user query
def handle_query(query, chatbot):
if global_chain is None:
return chatbot + [("Bot", "Please provide input first.")]
try:
result = global_chain({"question": query})
return chatbot + [("You", query), ("System", result['answer'])]
except Exception as e:
print(f"Error processing query: {str(e)}")
return chatbot + [("Bot", f"Error: {str(e)}")]
# Function to toggle input method
def toggle_input_method(input_method):
if input_method == "Upload PDF":
return gr.update(visible=True), gr.update(visible=False)
elif input_method == "Paste Link":
return gr.update(visible=False), gr.update(visible=True)
else:
return gr.update(visible=False), gr.update(visible=False)
# Gradio interface
with gr.Blocks() as demo:
gr.Markdown("# Chat-With-Context")
with gr.Row():
input_method = gr.Radio(["Upload PDF", "Paste Link"], label="Choose Input Method", interactive=True)
with gr.Row(visible=False) as upload_section:
pdf_input = gr.File(label="Upload PDF")
upload_button = gr.Button("Process PDF")
with gr.Row(visible=False) as text_input_section:
text_input = gr.Textbox(label="Paste Link")
submit_text_button = gr.Button("Process Link")
input_status = gr.Textbox(label="Status", interactive=False)
with gr.Row(visible=False) as chat_section:
chatbot = gr.Chatbot(label="Chat")
query_input = gr.Textbox(label="Write Your Question", placeholder="Message Chat-With-Context")
send_button = gr.Button("Send")
input_method.change(toggle_input_method, inputs=input_method, outputs=[upload_section, text_input_section])
upload_button.click(fn=handle_pdf_upload, inputs=pdf_input, outputs=[input_status, chat_section, upload_section])
submit_text_button.click(fn=handle_link_input, inputs=text_input, outputs=[input_status, chat_section, text_input_section])
send_button.click(fn=handle_query, inputs=[query_input, chatbot], outputs=chatbot)
demo.launch(share=True) |