Moah tools.

agent_langgraph.py

@@ -1,30 +1,34 @@
 import base64
 import os
 import re
-from typing import Tuple, TypedDict, Annotated, Optional, cast
+import requests
 
-from langchain_core.utils.function_calling import convert_to_openai_tool
-from langchain_core.messages import HumanMessage, SystemMessage, AIMessage
+from typing import Tuple, TypedDict, Annotated, Optional
 
+from langchain_core.messages import HumanMessage, SystemMessage, AIMessage
 from langchain_core.messages import AnyMessage, HumanMessage
 from langchain_openai import ChatOpenAI
-
+from langchain_community.tools.tavily_search import TavilySearchResults
 from langgraph.graph.message import add_messages
-
 from langgraph.graph import START, StateGraph
 from langgraph.prebuilt import ToolNode, tools_condition
-from IPython.display import Image, display
 
-from langchain_community.tools import DuckDuckGoSearchRun
-from langchain_community.tools.tavily_search import TavilySearchResults
-import requests
+from agent_tools import transcribe_youtube_video
+from agent_tools import transcribe_audio_file
+from agent_tools import read_excel_file
+from agent_tools import download_webpage_content
+
+from agent_tools import multiply, add, subtract, divide, modulus, power
+
 
-MODEL_NAME = "gpt-4.1-mini"
 # MODEL_NAME = "gpt-4o"
+MODEL_NAME = "gpt-4.1"
+# MODEL_NAME = "gpt-4.1-mini"
 # MODEL_NAME = "o4-mini"
 
-# VISION_MODEL_NAME = "gpt-4o"
-VISION_MODEL_NAME = "gpt-4.1-mini"
+VISION_MODEL_NAME = "gpt-4o"
+# VISION_MODEL_NAME = "gpt-4.1"
+# VISION_MODEL_NAME = "gpt-4.1-mini"
 # VISION_MODEL_NAME = "o4-mini"
 
 SYSTEM_PROMPT = """\
@@ -33,12 +37,15 @@ I will ask you a question.
 Make sure you understand the question.
 Always think step by step.
 Double check your answer.
+Double check your calculations.
 Report your thoughts, and always finish your answer with the following template: FINAL ANSWER: [YOUR FINAL ANSWER].
 If you didn't arrive at an answer, finish your answer with: FINAL ANSWER: I don't know.
 YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings.
 If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise.
 If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise.
 If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string.
+If you are asked to alphabetize a list, sort the list items alphabetically.
+When the result is a currency number, don't include the currency symbol, just the number itself.
     """.strip()
 
 
@@ -52,10 +59,14 @@ class BasicAgent:
     llm = ChatOpenAI(model=MODEL_NAME, verbose=True)
 
     tools = [
-        # DuckDuckGoSearchRun(),
+        multiply, add, subtract, divide, modulus, power,
         TavilySearchResults(
             tavily_api_key="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"),
-        self.analyze_image,
+        # self.analyze_image,
+        transcribe_youtube_video,
+        transcribe_audio_file,
+        read_excel_file,
+        download_webpage_content,
     ]
 
     self.llm_with_tools = llm.bind_tools(tools)
@@ -109,17 +120,21 @@ class BasicAgent:
         prompt += f"Attached file content:\n{file_content}\n"
       else:
         is_image, mime_type = is_image_file(file_name)
+        is_audio = is_audio_file(file_name)
+        is_excel = is_excel_file(file_name)
+
         if is_image:
           print("Content length:", len(response.content))
           image_data = base64.b64encode(response.content).decode("utf-8")
-          # write the image data to a file
-          with open("dupa-jasia.png", "wb") as f:
-            f.write(response.content)
-          with open("pierdzi-stasia.png.base64", "w") as f:
-            f.write(image_data)
           prompt += f"\nImage file name: {file_name}\n"
           prompt += f"Image file data:\n{image_data}\n"
           prompt += f"Image file image mime type: {mime_type}\n"
+        elif is_audio:
+          audio_url = f"https://agents-course-unit4-scoring.hf.space/files/{task_id}"
+          prompt += f"\nAudio URL: {audio_url}\n"
+        elif is_excel:
+          excel_file_url = f"https://agents-course-unit4-scoring.hf.space/files/{task_id}"
+          prompt += f"\nExcel file URL: {excel_file_url}\n"
 
     input_messages: list[AnyMessage] = [HumanMessage(content=prompt)]
 
@@ -127,6 +142,9 @@ class BasicAgent:
         {
             "messages": input_messages,
             "file_name": file_name
+        },
+        config={
+            "recursion_limit": 100,
         }
     )
 
@@ -227,6 +245,18 @@ def is_image_file(file_name: str) -> Tuple[bool, str]:
   return (ext in image_extensions, mime_type)
 
 
+def is_audio_file(file_name: str) -> bool:
+  plain_text_extensions = {'.mp3'}
+  ext = os.path.splitext(file_name)[1].lower()
+  return ext in plain_text_extensions
+
+
+def is_excel_file(file_name: str) -> bool:
+  excel_extensions = {'.xlsx', '.xls', '.xlsm'}
+  ext = os.path.splitext(file_name)[1].lower()
+  return ext in excel_extensions
+
+
 def ground_truth_answer(question: str) -> str:
   """
   Returns the answer corresponding to the given question,

agent_tools.py

@@ -0,0 +1,154 @@
+import os
+import requests
+import pandas as pd
+from langchain.tools import tool
+from pytube import extract
+from langchain_core.tools import tool
+from openai import OpenAI
+from youtube_transcript_api._api import YouTubeTranscriptApi
+
+
+@tool
+def transcribe_youtube_video(video_url: str) -> str:
+  """Get the transcript of a YouTube video.
+  Args:
+      video_url (str): YouTube URL of the video
+  """
+  video_id = extract.video_id(video_url)
+
+  ytt_api = YouTubeTranscriptApi()
+
+  transcript = ytt_api.fetch(video_id)
+
+  txt = "\n".join([s.text for s in transcript.snippets])
+
+  return txt
+
+
+@tool
+def transcribe_audio_file(audio_url: str) -> str:
+  """Transcribe an audio file to text.
+
+  Args:
+      audio_url (str): The URL of the audio file
+  """
+  response = requests.get(audio_url)
+  response.raise_for_status()
+
+  file_extension = "mp3"
+  file_name = f"audio-file.{file_extension}"
+
+  with open(file_name, 'wb') as file:
+    file.write(response.content)
+
+  openai_client = OpenAI()
+
+  with open(file_name, "rb") as audio_file:
+    transcription = \
+        openai_client.audio.transcriptions.create(
+            model="whisper-1",
+            file=audio_file
+        )
+
+  os.remove(file_name)
+
+  return transcription.text
+
+
+@tool
+def read_excel_file(excel_file_url: str) -> str:
+  """Read an Excel file and return as string.
+  Args:
+      excel_file_url (str): The URL of the Excel file
+  """
+  response = requests.get(excel_file_url)
+  response.raise_for_status()
+
+  file_extension = "xlsx"
+  excel_file_name = f"excel-file.{file_extension}"
+
+  with open(excel_file_name, "wb") as file:
+    file.write(response.content)
+
+  df = pd.read_excel(excel_file_name)
+
+  text = df.to_string()
+
+  os.remove(excel_file_name)
+
+  return text
+
+
+@tool
+def download_webpage_content(page_url: str) -> str:
+  """Load a web page and return its content as text.
+  Args:
+      page_url (str): the URL of web page to get
+  """
+  response = requests.get(page_url)
+
+  response.raise_for_status()
+
+  return response.text
+
+
+@tool
+def multiply(a: float, b: float) -> float:
+  """Multiplies two numbers.
+  Args:
+      a (float): the first number
+      b (float): the second number
+  """
+  return a * b
+
+
+@tool
+def add(a: float, b: float) -> float:
+  """Adds two numbers.
+  Args:
+      a (float): the first number
+      b (float): the second number
+  """
+  return a + b
+
+
+@tool
+def subtract(a: float, b: float) -> float:
+  """Subtracts two numbers.
+  Args:
+      a (float): the first number
+      b (float): the second number
+  """
+  return a - b
+
+
+@tool
+def divide(a: float, b: float) -> float:
+  """Divides two numbers.
+  Args:
+      a (float): the first float number
+      b (float): the second float number
+  """
+  if b == 0:
+    raise ValueError("Cannot divided by zero.")
+  return a / b
+
+
+@tool
+def modulus(a: int, b: int) -> int:
+  """Get the modulus of two numbers.
+  Args:
+      a (int): the first number
+      b (int): the second number
+  """
+  return a % b
+
+
+@tool
+def power(a: float, b: float) -> float:
+  """Get the power of two numbers.
+  Args:
+      a (float): the first number
+      b (float): the second number
+  """
+  return a**b

app_local.py

@@ -5,10 +5,10 @@ import asyncio
 from agent_langgraph import BasicAgent
 
 
-# # # Answer: 3
-# task_id="8e867cd7-cff9-4e6c-867a-ff5ddc2550be"
-# question="How many studio albums were published by Mercedes Sosa between 2000 and 2009 (included)? You can use the latest 2022 version of english wikipedia."
-# file_name=None
+# # Answer: 3
+task_id="8e867cd7-cff9-4e6c-867a-ff5ddc2550be"
+question="How many studio albums were published by Mercedes Sosa between 2000 and 2009 (included)? You can use the latest 2022 version of english wikipedia."
+file_name=None
 
 # # Answer: 3
 # task_id="a1e91b78-d3d8-4675-bb8d-62741b4b68a6"
@@ -21,9 +21,9 @@ from agent_langgraph import BasicAgent
 # file_name = None
 
 # # Answer: Rd5
-task_id="cca530fc-4052-43b2-b130-b30968d8aa44"
-question="Review the chess position provided in the image. It is black's turn. Provide the correct next move for black which guarantees a win. Please provide your response in algebraic notation."
-file_name="cca530fc-4052-43b2-b130-b30968d8aa44.png"
+# task_id="cca530fc-4052-43b2-b130-b30968d8aa44"
+# question="Review the chess position provided in the image. It is black's turn. Provide the correct next move for black which guarantees a win. Please provide your response in algebraic notation."
+# file_name="cca530fc-4052-43b2-b130-b30968d8aa44.png"
 
 # # Answer: FunkMonk
 # task_id="4fc2f1ae-8625-45b5-ab34-ad4433bc21f8"
@@ -35,7 +35,7 @@ file_name="cca530fc-4052-43b2-b130-b30968d8aa44.png"
 # question="Given this table defining * on the set S = {a, b, c, d, e}\n\n|*|a|b|c|d|e|\n|---|---|---|---|---|---|\n|a|a|b|c|b|d|\n|b|b|c|a|e|c|\n|c|c|a|b|b|a|\n|d|b|e|b|e|d|\n|e|d|b|a|d|c|\n\nprovide the subset of S involved in any possible counter-examples that prove * is not commutative. Provide your answer as a comma separated list of the elements in the set in alphabetical order."
 # file_name=None
 
-# # Answer: Extremely
+# Answer: Extremely
 # task_id="9d191bce-651d-4746-be2d-7ef8ecadb9c2"
 # question="Examine the video at https://www.youtube.com/watch?v=1htKBjuUWec.\n\nWhat does Teal'c say in response to the question \"Isn't that hot?\""
 # file_name=None
@@ -80,7 +80,7 @@ file_name="cca530fc-4052-43b2-b130-b30968d8aa44.png"
 # question="On June 6, 2023, an article by Carolyn Collins Petersen was published in Universe Today. This article mentions a team that produced a paper about their observations, linked at the bottom of the article. Find this paper. Under what NASA award number was the work performed by R. G. Arendt supported by?"
 # file_name=None
 
-# # Answer: Saint Petersburg
+# # Answer: Saint Petersburgu
 # task_id="bda648d7-d618-4883-88f4-3466eabd860e"
 # question="Where were the Vietnamese specimens described by Kuznetzov in Nedoshivina's 2010 paper eventually deposited? Just give me the city name without abbreviations."
 # file_name=None

audio_transcribe.py

@@ -0,0 +1,5 @@
+from agent_tools import transcribe_audio_file
+
+audio_url = "https://agents-course-unit4-scoring.hf.space/files/99c9cc74-fdc8-46c6-8f8d-3ce2d3bfeea3"
+
+print(transcribe_audio_file(audio_url))

excel_read.py

@@ -0,0 +1,6 @@
+from agent_tools import read_excel_file
+
+
+excel_file_url = "https://agents-course-unit4-scoring.hf.space/files/7bd855d8-463d-4ed5-93ca-5fe35145f733"
+
+print(read_excel_file(excel_file_url))

requirements.txt

@@ -17,3 +17,8 @@ langchain_core
 langchain-community
 duckduckgo-search
 tavily-python
+
+youtube_transcript_api
+pytube
+pandas
+openpyxl

youtube_transcribe.py

@@ -0,0 +1,6 @@
+from agent_tools import transcribe_youtube_video
+
+
+video_url = "https://www.youtube.com/watch?v=1htKBjuUWec"
+
+print(transcribe_youtube_video(video_url))