AmineLemsih's picture
Ajout de l'agent résumé YouTube
fc5a09d verified
raw
history blame
3.15 kB
from smolagents import CodeAgent,DuckDuckGoSearchTool, HfApiModel,load_tool, toolInferenceClientModel
from youtube_transcript_api import YouTubeTranscriptApi # pip install youtube-transcript-api
import datetime
import requests
import pytz
import yaml, textwrap, re
from tools.final_answer import FinalAnswerTool
from Gradio_UI import GradioUI
# ---------- TOOLS ----------
@tool
def get_transcript(url: str) -> str:
"""
Télécharge la transcription (FR ou EN) d’une vidéo YouTube.
Args:
url: lien complet YouTube
"""
# extraction robuste de l’ID
video_id = re.search(r"(?:v=|youtu\.be/)([^&\n?#]+)", url)
if not video_id:
return "Impossible de détecter l’ID vidéo."
video_id = video_id.group(1)
try:
transcript = YouTubeTranscriptApi.get_transcript(
video_id, languages=['fr', 'en'])
return " ".join(seg["text"] for seg in transcript)
except Exception as e:
return f"Erreur transcript : {e}"
@tool
def summarize(text: str, max_chars: int = 1500) -> str:
"""
Résume le texte en ~10 bullet points (Markdown).
Args:
text: transcription brute
max_chars: longueur max passée au LLM
"""
# coupe si trop long pour le contexte
snippet = text[:max_chars]
prompt = (
"Résume le texte suivant en 8‑10 bullet points clairs :\n\n"
f"{snippet}\n\n# Résumé :"
)
return prompt # on renvoie le prompt ; l’agent fera l’appel LLM
@tool
def get_current_time_in_timezone(timezone: str) -> str:
"""A tool that fetches the current local time in a specified timezone.
Args:
timezone: A string representing a valid timezone (e.g., 'America/New_York').
"""
try:
# Create timezone object
tz = pytz.timezone(timezone)
# Get current time in that timezone
local_time = datetime.datetime.now(tz).strftime("%Y-%m-%d %H:%M:%S")
return f"The current local time in {timezone} is: {local_time}"
except Exception as e:
return f"Error fetching time for timezone '{timezone}': {str(e)}"
# ---------- AGENT ----------
final_answer = FinalAnswerTool()
# If the agent does not answer, the model is overloaded, please use another model or the following Hugging Face Endpoint that also contains qwen2.5 coder:
# model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud'
model = HfApiModel(
max_tokens=2096,
temperature=0.5,
model_id='Qwen/Qwen2.5-Coder-32B-Instruct',# it is possible that this model may be overloaded
custom_role_conversions=None,
)
# Import tool from Hub
image_generation_tool = load_tool("agents-course/text-to-image", trust_remote_code=True)
with open("prompts.yaml", 'r') as stream:
prompt_templates = yaml.safe_load(stream)
agent = CodeAgent(
model=model,
tools=tools=[final_answer, get_transcript, summarize], ## add your tools here (don't remove final answer)
max_steps=6,
verbosity_level=1,
grammar=None,
planning_interval=None,
name=None,
description=None,
prompt_templates=prompt_templates
)
GradioUI(agent).launch()