genai-rss-space / agent.py
8shanrahan's picture
Update agent.py
8ef1932 verified
from transformers import pipeline
import time
# Load once at startup
classifier = pipeline(
"zero-shot-classification",
model="valhalla/distilbart-mnli-12-3"
)
summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")
def analyze_article(text, title, link):
t0 = time.time()
result = classifier(
text,
candidate_labels=["analytics", "data science", "business insight"],
multi_label=True
)
print(f"[DEBUG] Classifier Model inference time: {time.time() - t0:.2f}s")
# Summarize article text
summary = summarizer(text[:1024], max_length=150, min_length=40, do_sample=False)[0]['summary_text']
print(f"[DEBUG] Summarizer Model inference time: {time.time() - t0:.2f}s")
is_useful = any(label in ["analytics", "data science"] for label in result['labels'][:2])
return {
"title": title,
"link": link,
"summary": summary,
"top_label": result['labels'][0],
"relevance": "Useful" if is_useful else "Not useful"
}