limitedonly41's picture
Update app.py
44a6dcd verified
raw
history blame
6.65 kB
import gradio as gr
import asyncio
import requests
from bs4 import BeautifulSoup
import pandas as pd
from tqdm import tqdm
import urllib
from deep_translator import GoogleTranslator
import spaces
# from unsloth import FastLanguageModel
import torch
import re
# Define helper functions
async def fetch_data(url):
headers = {
'Accept': '*/*',
'Accept-Language': 'ru-RU,ru;q=0.9,en-US;q=0.8,en;q=0.7',
'Connection': 'keep-alive',
'Referer': f'{url}',
'Sec-Fetch-Dest': 'empty',
'Sec-Fetch-Mode': 'cors',
'Sec-Fetch-Site': 'cross-site',
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36',
'sec-ch-ua': '"Google Chrome";v="125", "Chromium";v="125", "Not.A/Brand";v="24"',
'sec-ch-ua-mobile': '?0',
'sec-ch-ua-platform': '"macOS"',
}
encoding = 'utf-8'
timeout = 10
try:
def get_content():
req = urllib.request.Request(url, headers=headers)
with urllib.request.urlopen(req, timeout=timeout) as response:
return response.read()
response_content = await asyncio.get_event_loop().run_in_executor(None, get_content)
soup = BeautifulSoup(response_content, 'html.parser', from_encoding=encoding)
title = soup.find('title').text
description = soup.find('meta', attrs={'name': 'description'})
if description and "content" in description.attrs:
description = description.get("content")
else:
description = ""
keywords = soup.find('meta', attrs={'name': 'keywords'})
if keywords and "content" in keywords.attrs:
keywords = keywords.get("content")
else:
keywords = ""
h1_all = " ".join(h.text for h in soup.find_all('h1'))
h2_all = " ".join(h.text for h in soup.find_all('h2'))
h3_all = " ".join(h.text for h in soup.find_all('h3'))
paragraphs_all = " ".join(p.text for p in soup.find_all('p'))
allthecontent = f"{title} {description} {h1_all} {h2_all} {h3_all} {paragraphs_all}"
allthecontent = allthecontent[:4999]
return {
'url': url,
'title': title,
'description': description,
'keywords': keywords,
'h1': h1_all,
'h2': h2_all,
'h3': h3_all,
'paragraphs': paragraphs_all,
'text': allthecontent
}
except Exception as e:
return {
'url': url,
'title': None,
'description': None,
'keywords': None,
'h1': None,
'h2': None,
'h3': None,
'paragraphs': None,
'text': None
}
def concatenate_text(data):
text_parts = [str(data[col]) for col in ['url', 'title', 'description', 'keywords', 'h1', 'h2', 'h3'] if data[col]]
text = ' '.join(text_parts)
text = text.replace(r'\xa0', ' ').replace('\n', ' ').replace('\t', ' ')
text = re.sub(r'\s{2,}', ' ', text)
return text
def translate_text(text):
try:
text = text[:4990]
translated_text = GoogleTranslator(source='auto', target='en').translate(text)
return translated_text
except Exception as e:
print(f"An error occurred during translation: {e}")
return None
model_name = "unsloth/mistral-7b-instruct-v0.3-bnb-4bit"
# Initialize model and tokenizer variables
model = None
tokenizer = None
@spaces.GPU(duration=85)
def summarize_url(url):
global model, tokenizer # Declare model and tokenizer as global variables
# Load the model
max_seq_length = 2048
dtype = None
load_in_4bit = True
if model is None or tokenizer is None:
from unsloth import FastLanguageModel
# Load the model and tokenizer
model, tokenizer = FastLanguageModel.from_pretrained(
model_name=model_name, # YOUR MODEL YOU USED FOR TRAINING
max_seq_length=max_seq_length,
dtype=dtype,
load_in_4bit=load_in_4bit,
)
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
result = asyncio.run(fetch_data(url))
text = concatenate_text(result)
translated_text = translate_text(text)
if len(translated_text) < 100:
return 'not scraped or short text'
alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
### Instruction:
Describe the website text into one word topic:
### Input:
{}
### Response:
"""
prompt = alpaca_prompt.format(translated_text)
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(inputs.input_ids, max_new_tokens=64, use_cache=True)
summary = tokenizer.decode(outputs[0], skip_special_tokens=True)
final_answer = summary.split("### Response:")[1].strip()
return final_answer
# # Create the Gradio interface within a `Blocks` context, like the working example
# with gr.Blocks() as demo:
# # Add title and description to the interface
# gr.HTML("<h1>Website Summary Generator</h1>")
# gr.HTML("<p>Enter a URL to get a one-word topic summary of the website content..</p>")
# # Define input and output elements
# with gr.Row():
# prompt = gr.Textbox(label="Enter Website URL", placeholder="https://example.com")
# output_text = gr.Textbox(label="Topic", interactive=False)
# # Add the button to trigger the function
# submit = gr.Button("Classify")
# # Define the interaction between inputs and outputs
# submit.click(fn=summarize_url, inputs=prompt, outputs=output_text)
# # Add the `if __name__ == "__main__":` block to launch the interface
# if __name__ == "__main__":
# demo.launch()
# with gr as demo:
# # Define Gradio interface
# demo = demo.Interface(
# fn=summarize_url,
# inputs="text",
# outputs="text",
# title="Website Summary Generator",
# description="Enter a URL to get a one-word topic summary of the website content."
# )
# if __name__ == "__main__":
# demo.launch()
# Create a Gradio interface
iface = gr.Interface(
fn=summarize_url,
inputs="text",
outputs="text",
title="Website Summary Generator",
description="Enter a URL to get a one-word topic summary of the website content."
)
# Launch the interface
iface.launch()