File size: 3,076 Bytes
8b07697 7244d4f 8b07697 ea77eb1 61b1afa 8b07697 61b1afa 8b07697 61b1afa 8b07697 61b1afa 8b07697 61b1afa 8b07697 61b1afa 7244d4f 61b1afa 3b57b61 61b1afa 7244d4f 61b1afa 8b07697 61b1afa 3b57b61 61b1afa |
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 |
from transformers import pipeline
import requests
from bs4 import BeautifulSoup
from time import sleep
# Load the model once
try:
hf_pipe = pipeline("text2text-generation", model="google/flan-t5-small", max_length=512, temperature=0.7)
except Exception as e:
print(f"Error loading model: {e}")
hf_pipe = None
def safe_invoke(prompt: str) -> str:
if not hf_pipe:
return "Error: Model not loaded"
try:
outputs = hf_pipe(prompt)
if outputs and isinstance(outputs, list):
return outputs[0]['generated_text'].strip()
return "Error"
except Exception as e:
print(f"Error during generation: {e}")
return "Error"
def scrape_startpage(query: str, max_results: int = 3):
url = f"https://www.startpage.com/sp/search?query={query.replace(' ', '+')}"
headers = {"User-Agent": "Mozilla/5.0"}
for attempt in range(3):
try:
res = requests.get(url, headers=headers, timeout=10)
res.raise_for_status()
soup = BeautifulSoup(res.text, "html.parser")
results = []
# Find divs with class "result" (Startpage search results)
for r in soup.find_all("div", class_="result")[:max_results]:
title = r.find("h3")
desc = r.find("p", class_="desc")
title_text = title.get_text(strip=True) if title else "No title"
desc_text = desc.get_text(strip=True) if desc else "No description"
results.append(f"{title_text}: {desc_text}")
return results
except Exception as e:
print(f"Scrape error (attempt {attempt+1}): {e}")
sleep(2 ** attempt)
return []
def generate_post(topic, platform, search_results):
base_prompt = f"""You are a social media expert. Write a professional {platform} post about "{topic}".
Use this information to help you: {search_results}
Make the post clear, engaging, and suitable for corporate clients.
Output only the post text."""
return safe_invoke(base_prompt)
def score_post(post, platform, score_type):
prompt = f"""Rate the following post on {score_type} from 1 to 10 (just give a number):
Platform: {platform}
Post: {post}
"""
return safe_invoke(prompt)
def workflow(topic, platform):
# Step 1: Web search results
search_results = scrape_startpage(topic)
combined_results = " | ".join(search_results) if search_results else "No additional info."
# Step 2: Generate post
post = generate_post(topic, platform, combined_results)
if post == "Error":
return post, "Error", "Error", "Error"
# Step 3: Scores
engagement = score_post(post, platform, "engagement")
tone = score_post(post, platform, "tone")
clarity = score_post(post, platform, "clarity")
# Validate scores (should be digits)
def valid_score(s):
return s and s.strip().isdigit()
if not all(map(valid_score, (engagement, tone, clarity))):
return post, "Error", "Error", "Error"
return post, engagement, tone, clarity
|