Spaces:
Sleeping
Sleeping
File size: 1,868 Bytes
4e6a306 82f7e90 9da9b54 82f7e90 15d6c6c 82f7e90 4e6a306 15d6c6c 9da9b54 09c85b0 15d6c6c 9da9b54 4e6a306 82f7e90 15d6c6c 4e6a306 15d6c6c 4e6a306 82f7e90 4e6a306 15d6c6c 4e6a306 9da9b54 4e6a306 15d6c6c 4e6a306 |
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 |
import gradio as gr
from huggingface_hub import InferenceClient
from PyPDF2 import PdfReader
import os
# PDF ํ
์คํธ ๋ฏธ๋ฆฌ ์ฝ์ด์ค๊ธฐ
def extract_pdf_text(pdf_paths):
full_text = ""
for path in pdf_paths:
reader = PdfReader(path)
for page in reader.pages:
text = page.extract_text()
if text:
full_text += text + "\n"
return full_text.strip()
# ๋ ํผ๋ฐ์ค PDF ํ
์คํธ ๋ถ๋ฌ์ค๊ธฐ
pdf_context = extract_pdf_text([
"assets/Programming-Fundamentals-1570222270.pdf",
"assets/1๋ถํ์ด์ฌ_๊ฐ์์๋ฃ_์ ์ฒด.pdf"
])
# ๋ฌด๋ฃ ์ฌ์ฉ ๊ฐ๋ฅํ FLAN-T5 ๋ชจ๋ธ ์ฌ์ฉ
client = InferenceClient(
model="tiiuae/falcon-rw-1b",
token=os.getenv("HUGGINGFACEHUB_API_TOKEN")
)
def respond(message, history, system_message, max_tokens, temperature, top_p):
# ๋จ์ ํ๋กฌํํธ ๊ตฌ์ฑ (flan-t5๋ chat ๊ตฌ์กฐ๊ฐ ์๋)
prompt = f"{system_message}\n\n๋ฌธ์ ์์ฝ:\n{pdf_context}\n\n์ง๋ฌธ: {message}\n๋ต๋ณ:"
result = client.text_generation(
prompt=prompt,
max_new_tokens=max_tokens,
temperature=temperature,
top_p=top_p
)
return result.strip()
demo = gr.ChatInterface(
fn=respond,
additional_inputs=[
gr.Textbox(value="๋น์ ์ ํ์ด์ฌ API ๋ฌธ์์ ๊ธฐ๋ฐํด ๋ต๋ณํ๋ ์ ์ฉํ ์กฐ๊ต์
๋๋ค.", label="System message"),
gr.Slider(minimum=1, maximum=1024, value=512, step=1, label="Max new tokens"),
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p"),
],
title="๐ ํ์ด์ฌ API ๋ ํผ๋ฐ์ค ์ฑ๋ด (FLAN-T5 ๊ธฐ๋ฐ)",
description="ํ๊ตญ๊ณต๋ ์์
์๋ฃ ๊ธฐ๋ฐ์ผ๋ก ์ง๋ฌธ์ ๋ตํ๋ ๋ฌด๋ฃ ์ฑ๋ด์
๋๋ค."
)
if __name__ == "__main__":
demo.launch()
|