Spaces:
Sleeping
Sleeping
siddhartharya
commited on
Commit
•
fe49b51
1
Parent(s):
dcf746e
Update app.py
Browse files
app.py
CHANGED
@@ -5,14 +5,13 @@ from bs4 import BeautifulSoup
|
|
5 |
from sentence_transformers import SentenceTransformer
|
6 |
import faiss
|
7 |
import numpy as np
|
8 |
-
import
|
9 |
-
import
|
10 |
import re
|
11 |
import base64
|
12 |
import logging
|
13 |
import os
|
14 |
import sys
|
15 |
-
import time
|
16 |
|
17 |
# Import OpenAI library
|
18 |
import openai
|
@@ -71,23 +70,9 @@ GROQ_API_KEY = os.getenv('GROQ_API_KEY')
|
|
71 |
if not GROQ_API_KEY:
|
72 |
logger.error("GROQ_API_KEY environment variable not set.")
|
73 |
|
74 |
-
# Set OpenAI API key and base URL to use Groq Cloud API
|
75 |
openai.api_key = GROQ_API_KEY
|
76 |
openai.api_base = "https://api.groq.com/openai/v1"
|
77 |
|
78 |
-
def extract_retry_after(error_message):
|
79 |
-
"""
|
80 |
-
Extract the retry-after time from the rate limit error message.
|
81 |
-
"""
|
82 |
-
match = re.search(r'Please try again in (\d+\.?\d*)s', error_message)
|
83 |
-
if match:
|
84 |
-
return float(match.group(1)) + 1 # Add a buffer of 1 second
|
85 |
-
else:
|
86 |
-
return 5 # Default retry after 5 seconds
|
87 |
-
|
88 |
-
def exponential_backoff(retries):
|
89 |
-
return min(60, (2 ** retries)) # Cap the wait time at 60 seconds
|
90 |
-
|
91 |
def extract_main_content(soup):
|
92 |
"""
|
93 |
Extract the main content from a webpage while filtering out boilerplate content.
|
@@ -157,10 +142,6 @@ def get_page_metadata(soup):
|
|
157 |
|
158 |
return metadata
|
159 |
|
160 |
-
async def generate_summary_async(bookmark):
|
161 |
-
async with llm_semaphore:
|
162 |
-
await asyncio.get_event_loop().run_in_executor(None, generate_summary, bookmark)
|
163 |
-
|
164 |
def generate_summary(bookmark):
|
165 |
"""
|
166 |
Generate a concise summary for a bookmark using available content and LLM via the Groq Cloud API.
|
@@ -204,11 +185,11 @@ def generate_summary(bookmark):
|
|
204 |
if use_prior_knowledge:
|
205 |
# Construct prompt to use prior knowledge
|
206 |
prompt = f"""
|
207 |
-
You are a knowledgeable assistant.
|
208 |
|
209 |
The user provided a URL: {bookmark.get('url')}
|
210 |
|
211 |
-
Please provide a concise summary in **no more than two sentences** about this website
|
212 |
|
213 |
Focus on:
|
214 |
- The main purpose or topic of the website.
|
@@ -233,44 +214,24 @@ Be concise and objective.
|
|
233 |
"""
|
234 |
|
235 |
# Call the LLM via Groq Cloud API
|
236 |
-
|
237 |
-
|
238 |
-
|
239 |
-
|
240 |
-
|
241 |
-
|
242 |
-
|
243 |
-
|
244 |
-
],
|
245 |
-
max_tokens=100, # Reduced max tokens
|
246 |
-
temperature=0.5,
|
247 |
-
)
|
248 |
-
break # Exit loop if successful
|
249 |
-
except openai.error.RateLimitError as e:
|
250 |
-
retry_after = extract_retry_after(str(e)) or exponential_backoff(retries)
|
251 |
-
logger.warning(f"Rate limit exceeded. Retrying after {retry_after} seconds.")
|
252 |
-
time.sleep(retry_after)
|
253 |
-
retries += 1
|
254 |
-
except Exception as e:
|
255 |
-
logger.error(f"Error generating summary: {e}", exc_info=True)
|
256 |
-
bookmark['summary'] = 'No summary available.'
|
257 |
-
return bookmark
|
258 |
-
|
259 |
summary = response['choices'][0]['message']['content'].strip()
|
260 |
if not summary:
|
261 |
raise ValueError("Empty summary received from the model.")
|
262 |
logger.info("Successfully generated LLM summary")
|
263 |
bookmark['summary'] = summary
|
264 |
-
|
265 |
|
266 |
except Exception as e:
|
267 |
logger.error(f"Error generating summary: {e}", exc_info=True)
|
268 |
bookmark['summary'] = 'No summary available.'
|
269 |
-
return bookmark
|
270 |
-
|
271 |
-
async def assign_category_async(bookmark):
|
272 |
-
async with llm_semaphore:
|
273 |
-
await asyncio.get_event_loop().run_in_executor(None, assign_category, bookmark)
|
274 |
|
275 |
def assign_category(bookmark):
|
276 |
"""
|
@@ -279,12 +240,12 @@ def assign_category(bookmark):
|
|
279 |
if bookmark.get('dead_link'):
|
280 |
bookmark['category'] = 'Dead Link'
|
281 |
logger.info(f"Assigned category 'Dead Link' to bookmark: {bookmark.get('url')}")
|
282 |
-
return
|
283 |
|
284 |
summary = bookmark.get('summary', '')
|
285 |
if not summary:
|
286 |
bookmark['category'] = 'Uncategorized'
|
287 |
-
return
|
288 |
|
289 |
# Prepare the prompt
|
290 |
categories_str = ', '.join([f'"{cat}"' for cat in CATEGORIES if cat != 'Dead Link'])
|
@@ -302,40 +263,28 @@ Categories:
|
|
302 |
Respond with only the category name.
|
303 |
"""
|
304 |
|
305 |
-
|
306 |
-
|
307 |
-
|
308 |
-
|
309 |
-
|
310 |
-
|
311 |
-
|
312 |
-
|
313 |
-
|
314 |
-
|
315 |
-
|
316 |
-
|
317 |
-
|
318 |
-
|
319 |
-
|
320 |
-
logger.warning(f"Rate limit exceeded. Retrying after {retry_after} seconds.")
|
321 |
-
time.sleep(retry_after)
|
322 |
-
retries += 1
|
323 |
-
except Exception as e:
|
324 |
-
logger.error(f"Error assigning category: {e}", exc_info=True)
|
325 |
bookmark['category'] = 'Uncategorized'
|
326 |
-
|
327 |
-
|
328 |
-
category = response['choices'][0]['message']['content'].strip().strip('"')
|
329 |
|
330 |
-
|
331 |
-
|
332 |
-
bookmark['category'] = category
|
333 |
-
logger.info(f"Assigned category '{category}' to bookmark: {bookmark.get('url')}")
|
334 |
-
else:
|
335 |
bookmark['category'] = 'Uncategorized'
|
336 |
-
logger.warning(f"Invalid category '{category}' returned by LLM for bookmark: {bookmark.get('url')}")
|
337 |
-
|
338 |
-
return bookmark
|
339 |
|
340 |
def parse_bookmarks(file_content):
|
341 |
"""
|
@@ -356,109 +305,65 @@ def parse_bookmarks(file_content):
|
|
356 |
logger.error("Error parsing bookmarks: %s", e, exc_info=True)
|
357 |
raise
|
358 |
|
359 |
-
|
360 |
"""
|
361 |
-
Fetch information about a URL
|
362 |
"""
|
363 |
url = bookmark['url']
|
364 |
if url in fetch_cache:
|
365 |
bookmark.update(fetch_cache[url])
|
366 |
-
return
|
367 |
-
|
368 |
-
|
369 |
-
|
370 |
-
|
371 |
-
|
372 |
-
|
373 |
-
|
374 |
-
|
375 |
-
|
376 |
-
|
377 |
-
|
378 |
-
|
379 |
-
|
380 |
-
|
381 |
-
|
382 |
-
|
383 |
-
|
384 |
-
logger.info(f"Fetched content length for {url}: {len(content)} characters")
|
385 |
-
|
386 |
-
# Handle status codes
|
387 |
-
if response.status >= 500:
|
388 |
-
# Server error, consider as dead link
|
389 |
-
bookmark['dead_link'] = True
|
390 |
-
bookmark['description'] = ''
|
391 |
-
bookmark['html_content'] = ''
|
392 |
-
logger.warning(f"Dead link detected: {url} with status {response.status}")
|
393 |
-
else:
|
394 |
-
bookmark['dead_link'] = False
|
395 |
-
bookmark['html_content'] = content
|
396 |
-
bookmark['description'] = ''
|
397 |
-
logger.info(f"Fetched information for {url}")
|
398 |
-
break # Exit loop if successful
|
399 |
-
|
400 |
-
except asyncio.exceptions.TimeoutError:
|
401 |
-
bookmark['dead_link'] = False # Mark as 'Unknown' instead of 'Dead'
|
402 |
-
bookmark['etag'] = 'N/A'
|
403 |
-
bookmark['status_code'] = 'Timeout'
|
404 |
-
bookmark['description'] = ''
|
405 |
-
bookmark['html_content'] = ''
|
406 |
-
bookmark['slow_link'] = True # Custom flag to indicate slow response
|
407 |
-
logger.warning(f"Timeout while fetching {url}. Marking as 'Slow'.")
|
408 |
-
break # Exit loop after timeout
|
409 |
-
except Exception as e:
|
410 |
bookmark['dead_link'] = True
|
411 |
-
bookmark['etag'] = 'N/A'
|
412 |
-
bookmark['status_code'] = 'Error'
|
413 |
bookmark['description'] = ''
|
414 |
bookmark['html_content'] = ''
|
415 |
-
logger.
|
416 |
-
|
417 |
-
|
418 |
-
|
419 |
-
|
420 |
-
|
421 |
-
|
422 |
-
|
423 |
-
|
424 |
-
|
425 |
-
|
426 |
-
|
427 |
-
|
428 |
-
|
429 |
-
|
430 |
-
Fetch all bookmarks asynchronously.
|
431 |
-
"""
|
432 |
-
logger.info("Processing bookmarks asynchronously")
|
433 |
-
try:
|
434 |
-
connector = aiohttp.TCPConnector(limit=10) # Increase limit if necessary
|
435 |
-
timeout = aiohttp.ClientTimeout(total=60) # Set timeout
|
436 |
-
async with aiohttp.ClientSession(connector=connector, timeout=timeout) as session:
|
437 |
-
tasks = []
|
438 |
-
for bookmark in bookmarks_list:
|
439 |
-
task = asyncio.ensure_future(fetch_url_info(session, bookmark))
|
440 |
-
tasks.append(task)
|
441 |
-
await asyncio.gather(*tasks)
|
442 |
-
logger.info("Completed processing bookmarks asynchronously")
|
443 |
except Exception as e:
|
444 |
-
|
445 |
-
|
446 |
-
|
447 |
-
|
448 |
-
|
449 |
-
|
450 |
-
|
451 |
-
|
452 |
-
|
453 |
-
|
454 |
-
|
455 |
-
|
456 |
-
|
457 |
-
|
458 |
-
|
459 |
-
tasks.append(assign_category_async(bookmark))
|
460 |
-
await asyncio.gather(*tasks)
|
461 |
-
logger.info("Completed LLM processing of bookmarks")
|
462 |
|
463 |
def vectorize_and_index(bookmarks_list):
|
464 |
"""
|
@@ -489,16 +394,16 @@ def display_bookmarks():
|
|
489 |
index = i + 1
|
490 |
if bookmark.get('dead_link'):
|
491 |
status = "❌ Dead Link"
|
492 |
-
card_style = "border: 2px solid
|
493 |
-
text_style = "color:
|
494 |
elif bookmark.get('slow_link'):
|
495 |
status = "⏳ Slow Response"
|
496 |
card_style = "border: 2px solid orange;"
|
497 |
text_style = "color: orange;"
|
498 |
else:
|
499 |
status = "✅ Active"
|
500 |
-
card_style = "border: 2px solid
|
501 |
-
text_style = "color:
|
502 |
|
503 |
title = bookmark['title']
|
504 |
url = bookmark['url']
|
@@ -559,19 +464,14 @@ def process_uploaded_file(file):
|
|
559 |
for idx, bookmark in enumerate(bookmarks):
|
560 |
bookmark['id'] = idx
|
561 |
|
562 |
-
#
|
563 |
-
|
564 |
-
|
565 |
-
except Exception as e:
|
566 |
-
logger.error(f"Error processing bookmarks asynchronously: {e}", exc_info=True)
|
567 |
-
return "Error processing bookmarks.", '', gr.update(choices=[]), display_bookmarks()
|
568 |
|
569 |
-
#
|
570 |
-
|
571 |
-
|
572 |
-
|
573 |
-
logger.error(f"Error processing bookmarks with LLM: {e}", exc_info=True)
|
574 |
-
return "Error processing bookmarks with LLM.", '', gr.update(choices=[]), display_bookmarks()
|
575 |
|
576 |
try:
|
577 |
faiss_index = vectorize_and_index(bookmarks)
|
@@ -642,7 +542,7 @@ def edit_selected_bookmarks_category(selected_indices, new_category):
|
|
642 |
|
643 |
# Update choices and display
|
644 |
choices = [f"{i+1}. {bookmark['title']} (Category: {bookmark['category']})"
|
645 |
-
|
646 |
|
647 |
return message, gr.update(choices=choices), display_bookmarks()
|
648 |
|
@@ -718,31 +618,17 @@ Bookmarks:
|
|
718 |
Provide a concise and helpful response.
|
719 |
"""
|
720 |
|
721 |
-
|
722 |
-
|
723 |
-
|
724 |
-
|
725 |
-
|
726 |
-
|
727 |
-
|
728 |
-
|
729 |
-
],
|
730 |
-
max_tokens=500,
|
731 |
-
temperature=0.7,
|
732 |
-
)
|
733 |
-
break # Exit loop if successful
|
734 |
-
except openai.error.RateLimitError as e:
|
735 |
-
retry_after = extract_retry_after(str(e)) or exponential_backoff(retries)
|
736 |
-
logger.warning(f"Rate limit exceeded. Retrying after {retry_after} seconds.")
|
737 |
-
time.sleep(retry_after)
|
738 |
-
retries += 1
|
739 |
-
except Exception as e:
|
740 |
-
error_message = f"⚠️ Error processing your query: {str(e)}"
|
741 |
-
logger.error(error_message, exc_info=True)
|
742 |
-
return error_message
|
743 |
-
|
744 |
answer = response['choices'][0]['message']['content'].strip()
|
745 |
-
logger.info("Chatbot response generated
|
|
|
746 |
return answer
|
747 |
|
748 |
except Exception as e:
|
@@ -868,6 +754,4 @@ def build_app():
|
|
868 |
print(f"Error building the app: {e}")
|
869 |
|
870 |
if __name__ == "__main__":
|
871 |
-
# Define a semaphore to limit concurrent LLM API calls
|
872 |
-
llm_semaphore = asyncio.Semaphore(3) # Adjust based on allowed concurrency
|
873 |
build_app()
|
|
|
5 |
from sentence_transformers import SentenceTransformer
|
6 |
import faiss
|
7 |
import numpy as np
|
8 |
+
import requests
|
9 |
+
import time
|
10 |
import re
|
11 |
import base64
|
12 |
import logging
|
13 |
import os
|
14 |
import sys
|
|
|
15 |
|
16 |
# Import OpenAI library
|
17 |
import openai
|
|
|
70 |
if not GROQ_API_KEY:
|
71 |
logger.error("GROQ_API_KEY environment variable not set.")
|
72 |
|
|
|
73 |
openai.api_key = GROQ_API_KEY
|
74 |
openai.api_base = "https://api.groq.com/openai/v1"
|
75 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
76 |
def extract_main_content(soup):
|
77 |
"""
|
78 |
Extract the main content from a webpage while filtering out boilerplate content.
|
|
|
142 |
|
143 |
return metadata
|
144 |
|
|
|
|
|
|
|
|
|
145 |
def generate_summary(bookmark):
|
146 |
"""
|
147 |
Generate a concise summary for a bookmark using available content and LLM via the Groq Cloud API.
|
|
|
185 |
if use_prior_knowledge:
|
186 |
# Construct prompt to use prior knowledge
|
187 |
prompt = f"""
|
188 |
+
You are a knowledgeable assistant with up-to-date information as of 2023.
|
189 |
|
190 |
The user provided a URL: {bookmark.get('url')}
|
191 |
|
192 |
+
Please provide a concise summary in **no more than two sentences** about this website.
|
193 |
|
194 |
Focus on:
|
195 |
- The main purpose or topic of the website.
|
|
|
214 |
"""
|
215 |
|
216 |
# Call the LLM via Groq Cloud API
|
217 |
+
response = openai.ChatCompletion.create(
|
218 |
+
model='llama-3.1-70b-versatile',
|
219 |
+
messages=[
|
220 |
+
{"role": "user", "content": prompt}
|
221 |
+
],
|
222 |
+
max_tokens=100,
|
223 |
+
temperature=0.5,
|
224 |
+
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
225 |
summary = response['choices'][0]['message']['content'].strip()
|
226 |
if not summary:
|
227 |
raise ValueError("Empty summary received from the model.")
|
228 |
logger.info("Successfully generated LLM summary")
|
229 |
bookmark['summary'] = summary
|
230 |
+
time.sleep(3) # Wait to respect rate limits
|
231 |
|
232 |
except Exception as e:
|
233 |
logger.error(f"Error generating summary: {e}", exc_info=True)
|
234 |
bookmark['summary'] = 'No summary available.'
|
|
|
|
|
|
|
|
|
|
|
235 |
|
236 |
def assign_category(bookmark):
|
237 |
"""
|
|
|
240 |
if bookmark.get('dead_link'):
|
241 |
bookmark['category'] = 'Dead Link'
|
242 |
logger.info(f"Assigned category 'Dead Link' to bookmark: {bookmark.get('url')}")
|
243 |
+
return
|
244 |
|
245 |
summary = bookmark.get('summary', '')
|
246 |
if not summary:
|
247 |
bookmark['category'] = 'Uncategorized'
|
248 |
+
return
|
249 |
|
250 |
# Prepare the prompt
|
251 |
categories_str = ', '.join([f'"{cat}"' for cat in CATEGORIES if cat != 'Dead Link'])
|
|
|
263 |
Respond with only the category name.
|
264 |
"""
|
265 |
|
266 |
+
try:
|
267 |
+
response = openai.ChatCompletion.create(
|
268 |
+
model='llama-3.1-70b-versatile',
|
269 |
+
messages=[
|
270 |
+
{"role": "user", "content": prompt}
|
271 |
+
],
|
272 |
+
max_tokens=10,
|
273 |
+
temperature=0,
|
274 |
+
)
|
275 |
+
category = response['choices'][0]['message']['content'].strip().strip('"')
|
276 |
+
# Validate the category
|
277 |
+
if category in CATEGORIES:
|
278 |
+
bookmark['category'] = category
|
279 |
+
logger.info(f"Assigned category '{category}' to bookmark: {bookmark.get('url')}")
|
280 |
+
else:
|
|
|
|
|
|
|
|
|
|
|
281 |
bookmark['category'] = 'Uncategorized'
|
282 |
+
logger.warning(f"Invalid category '{category}' returned by LLM for bookmark: {bookmark.get('url')}")
|
283 |
+
time.sleep(3) # Wait to respect rate limits
|
|
|
284 |
|
285 |
+
except Exception as e:
|
286 |
+
logger.error(f"Error assigning category: {e}", exc_info=True)
|
|
|
|
|
|
|
287 |
bookmark['category'] = 'Uncategorized'
|
|
|
|
|
|
|
288 |
|
289 |
def parse_bookmarks(file_content):
|
290 |
"""
|
|
|
305 |
logger.error("Error parsing bookmarks: %s", e, exc_info=True)
|
306 |
raise
|
307 |
|
308 |
+
def fetch_url_info(bookmark):
|
309 |
"""
|
310 |
+
Fetch information about a URL.
|
311 |
"""
|
312 |
url = bookmark['url']
|
313 |
if url in fetch_cache:
|
314 |
bookmark.update(fetch_cache[url])
|
315 |
+
return
|
316 |
+
|
317 |
+
try:
|
318 |
+
logger.info(f"Fetching URL info for: {url}")
|
319 |
+
headers = {
|
320 |
+
'User-Agent': 'Mozilla/5.0',
|
321 |
+
'Accept-Language': 'en-US,en;q=0.9',
|
322 |
+
}
|
323 |
+
response = requests.get(url, headers=headers, timeout=5, verify=False, allow_redirects=True)
|
324 |
+
bookmark['etag'] = response.headers.get('ETag', 'N/A')
|
325 |
+
bookmark['status_code'] = response.status_code
|
326 |
+
|
327 |
+
content = response.text
|
328 |
+
logger.info(f"Fetched content length for {url}: {len(content)} characters")
|
329 |
+
|
330 |
+
# Handle status codes
|
331 |
+
if response.status_code >= 500:
|
332 |
+
# Server error, consider as dead link
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
333 |
bookmark['dead_link'] = True
|
|
|
|
|
334 |
bookmark['description'] = ''
|
335 |
bookmark['html_content'] = ''
|
336 |
+
logger.warning(f"Dead link detected: {url} with status {response.status_code}")
|
337 |
+
else:
|
338 |
+
bookmark['dead_link'] = False
|
339 |
+
bookmark['html_content'] = content
|
340 |
+
bookmark['description'] = ''
|
341 |
+
logger.info(f"Fetched information for {url}")
|
342 |
+
|
343 |
+
except requests.exceptions.Timeout:
|
344 |
+
bookmark['dead_link'] = False # Mark as 'Unknown' instead of 'Dead'
|
345 |
+
bookmark['etag'] = 'N/A'
|
346 |
+
bookmark['status_code'] = 'Timeout'
|
347 |
+
bookmark['description'] = ''
|
348 |
+
bookmark['html_content'] = ''
|
349 |
+
bookmark['slow_link'] = True # Custom flag to indicate slow response
|
350 |
+
logger.warning(f"Timeout while fetching {url}. Marking as 'Slow'.")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
351 |
except Exception as e:
|
352 |
+
bookmark['dead_link'] = True
|
353 |
+
bookmark['etag'] = 'N/A'
|
354 |
+
bookmark['status_code'] = 'Error'
|
355 |
+
bookmark['description'] = ''
|
356 |
+
bookmark['html_content'] = ''
|
357 |
+
logger.error(f"Error fetching URL info for {url}: {e}", exc_info=True)
|
358 |
+
finally:
|
359 |
+
fetch_cache[url] = {
|
360 |
+
'etag': bookmark.get('etag'),
|
361 |
+
'status_code': bookmark.get('status_code'),
|
362 |
+
'dead_link': bookmark.get('dead_link'),
|
363 |
+
'description': bookmark.get('description'),
|
364 |
+
'html_content': bookmark.get('html_content', ''),
|
365 |
+
'slow_link': bookmark.get('slow_link', False),
|
366 |
+
}
|
|
|
|
|
|
|
367 |
|
368 |
def vectorize_and_index(bookmarks_list):
|
369 |
"""
|
|
|
394 |
index = i + 1
|
395 |
if bookmark.get('dead_link'):
|
396 |
status = "❌ Dead Link"
|
397 |
+
card_style = "border: 2px solid red;"
|
398 |
+
text_style = "color: red;"
|
399 |
elif bookmark.get('slow_link'):
|
400 |
status = "⏳ Slow Response"
|
401 |
card_style = "border: 2px solid orange;"
|
402 |
text_style = "color: orange;"
|
403 |
else:
|
404 |
status = "✅ Active"
|
405 |
+
card_style = "border: 2px solid green;"
|
406 |
+
text_style = "color: black;"
|
407 |
|
408 |
title = bookmark['title']
|
409 |
url = bookmark['url']
|
|
|
464 |
for idx, bookmark in enumerate(bookmarks):
|
465 |
bookmark['id'] = idx
|
466 |
|
467 |
+
# Fetch bookmark info sequentially
|
468 |
+
for bookmark in bookmarks:
|
469 |
+
fetch_url_info(bookmark)
|
|
|
|
|
|
|
470 |
|
471 |
+
# Process bookmarks sequentially with LLM
|
472 |
+
for bookmark in bookmarks:
|
473 |
+
generate_summary(bookmark)
|
474 |
+
assign_category(bookmark)
|
|
|
|
|
475 |
|
476 |
try:
|
477 |
faiss_index = vectorize_and_index(bookmarks)
|
|
|
542 |
|
543 |
# Update choices and display
|
544 |
choices = [f"{i+1}. {bookmark['title']} (Category: {bookmark['category']})"
|
545 |
+
for i, bookmark in enumerate(bookmarks)]
|
546 |
|
547 |
return message, gr.update(choices=choices), display_bookmarks()
|
548 |
|
|
|
618 |
Provide a concise and helpful response.
|
619 |
"""
|
620 |
|
621 |
+
response = openai.ChatCompletion.create(
|
622 |
+
model='llama-3.1-70b-versatile',
|
623 |
+
messages=[
|
624 |
+
{"role": "user", "content": prompt}
|
625 |
+
],
|
626 |
+
max_tokens=500,
|
627 |
+
temperature=0.7,
|
628 |
+
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
629 |
answer = response['choices'][0]['message']['content'].strip()
|
630 |
+
logger.info("Chatbot response generated")
|
631 |
+
time.sleep(3) # Wait to respect rate limits
|
632 |
return answer
|
633 |
|
634 |
except Exception as e:
|
|
|
754 |
print(f"Error building the app: {e}")
|
755 |
|
756 |
if __name__ == "__main__":
|
|
|
|
|
757 |
build_app()
|