File size: 2,028 Bytes
231a99e 269ee13 231a99e 269ee13 231a99e 269ee13 231a99e 902bbe5 231a99e 269ee13 902bbe5 231a99e |
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 |
import torch
import gradio as gr
from transformers import PegasusTokenizer, PegasusForConditionalGeneration
# Define the PEGASUS model and tokenizer
MODEL_NAME = 'VishnuPottabatthini/PEGASUS_Large' # Change this to the PEGASUS model
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
# Load the tokenizer and model
tokenizer = PegasusTokenizer.from_pretrained(MODEL_NAME)
model = PegasusForConditionalGeneration.from_pretrained(MODEL_NAME).to(device)
# Define the summarization function
def summarize(text, state):
try:
# Tokenize the input text
inputs = tokenizer(
text,
return_tensors="pt",
truncation=True,
max_length=1024 # Adjust max length according to your model's capabilities
).to(device)
# Generate the summary
summary_ids = model.generate(
inputs['input_ids'],
attention_mask=inputs['attention_mask'],
max_length=150, # Maximum length of the summary
min_length=30, # Minimum length of the summary
num_beams=4, # Beam search to improve the quality of generated text
early_stopping=True
)
# Decode the summary
summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
return state + "\n" + summary, state + "\n" + summary
except Exception as e:
return str(e), state
# Create the Gradio interface
mf_summarize = gr.Interface(
fn=summarize,
inputs=[
gr.Textbox(placeholder="Enter text to summarize...", lines=10),
gr.State(value="")
],
outputs=[
gr.Textbox(lines=15, label="Summary"),
gr.State()
],
theme="huggingface",
title="Article Summarization",
live=True,
description=(
"Enter a long piece of text to generate a concise summary using a PEGASUS model. "
"This demo uses a custom PEGASUS model from 🤗 Transformers."
)
)
# Launch the Gradio interface
mf_summarize.launch() |