Shreyas094 commited on
Commit
f630f04
·
verified ·
1 Parent(s): 5f3ea63

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +14 -5
app.py CHANGED
@@ -260,7 +260,15 @@ def summarize_news_content(content, model):
260
  """
261
  prompt = ChatPromptTemplate.from_template(prompt_template)
262
  formatted_prompt = prompt.format(content=content)
263
- summary = generate_chunked_response(model, formatted_prompt, max_tokens=200)
 
 
 
 
 
 
 
 
264
  return summary
265
 
266
  def process_google_news_rss(query, temperature, top_p, repetition_penalty):
@@ -275,18 +283,19 @@ def process_google_news_rss(query, temperature, top_p, repetition_penalty):
275
 
276
  for article in articles:
277
  try:
278
- summary = summarize_news_content(article["content"], model)
 
 
279
  processed_article = {
280
  "published_date": article["published_date"],
281
  "title": article["title"],
282
  "url": article["url"],
283
- "content": article["content"],
284
  "summary": summary
285
  }
286
  processed_articles.append(processed_article)
287
  except Exception as e:
288
- print(f"Error processing article: {str(e)}")
289
-
290
  if not processed_articles:
291
  return "Failed to process any news articles. Please try a different query or check the summarization process."
292
 
 
260
  """
261
  prompt = ChatPromptTemplate.from_template(prompt_template)
262
  formatted_prompt = prompt.format(content=content)
263
+ full_response = generate_chunked_response(model, formatted_prompt, max_tokens=200)
264
+
265
+ # Extract only the summary part
266
+ summary_parts = full_response.split("Assistant:")
267
+ if len(summary_parts) > 1:
268
+ summary = summary_parts[-1].strip()
269
+ else:
270
+ summary = full_response.strip()
271
+
272
  return summary
273
 
274
  def process_google_news_rss(query, temperature, top_p, repetition_penalty):
 
283
 
284
  for article in articles:
285
  try:
286
+ # Remove HTML tags from content
287
+ clean_content = BeautifulSoup(article["content"], "html.parser").get_text()
288
+ summary = summarize_news_content(clean_content, model)
289
  processed_article = {
290
  "published_date": article["published_date"],
291
  "title": article["title"],
292
  "url": article["url"],
293
+ "content": clean_content,
294
  "summary": summary
295
  }
296
  processed_articles.append(processed_article)
297
  except Exception as e:
298
+ print(f"Error processing article: {str(e)}")
 
299
  if not processed_articles:
300
  return "Failed to process any news articles. Please try a different query or check the summarization process."
301