Spaces:
Sleeping
Sleeping
File size: 1,447 Bytes
d712281 b80349f d712281 0dec546 b80349f d712281 b80349f d712281 0dec546 d712281 0dec546 b80349f d712281 0dec546 d712281 b80349f 0dec546 b80349f |
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 |
from transformers import pipeline, AutoTokenizer
import gradio as gr
# Load tokenizer for truncation
tokenizer = AutoTokenizer.from_pretrained("sshleifer/distilbart-cnn-6-6")
# Load summarization pipeline
summarizer = pipeline(
"summarization",
model="sshleifer/distilbart-cnn-6-6",
)
# Truncation helper 🫶
def safe_truncate(text, max_tokens=1024):
tokens = tokenizer.encode(text, truncation=True, max_length=max_tokens)
return tokenizer.decode(tokens, skip_special_tokens=True)
# Summary function with truncation applied
def summarize_article(text):
short_text = safe_truncate(text) # prevent model from breaking!
summary = summarizer(
short_text,
max_length=250,
min_length=100,
do_sample=False,
)
return summary[0]['summary_text']
# Default example
default_article = """New York (CNN)When Liana Barrientos was 23 years old, she got married...""" # [shortened for demo]
default_summary = summarize_article(default_article)
# Gradio Interface
iface = gr.Interface(
fn=summarize_article,
inputs=gr.Textbox(lines=20, label="Article (Read Only)", value=default_article, interactive=False),
outputs=gr.Textbox(label="Summary (Read Only)", value=default_summary, interactive=False),
title="⚡ Fast Article Summarizer (CPU Optimized)",
description="Fast summarization with longer output using CPU only. Inputs and outputs are read-only."
)
iface.launch()
|