Delete ai_agents_course_submission.py

ai_agents_course_submission.py

@@ -1,213 +0,0 @@
-# -*- coding: utf-8 -*-
-"""AI Agents Course_Submission.ipynb
-
-Automatically generated by Colab.
-
-Original file is located at
-    https://colab.research.google.com/drive/138KHbO69DpJpUvuTVAvsZ53EyXZAdr6K
-"""
-
-!pip install smolagents -U
-
-pip install duckduckgo-search
-
-!pip install requests
-
-#login to hugging face
-from huggingface_hub import notebook_login
-
-notebook_login()
-
-from smolagents import CodeAgent, tool, DuckDuckGoSearchTool, InferenceClientModel
-import requests
-
-@tool
-def suggest_outfit(occasion: str) -> str:
-    """
-    Suggests an outfit based on the occasion.
-    Args:
-        occasion (str): The type of occasion for the event. Allowed values are:
-                        - "casual": Outfit for casual event.
-                        - "formal": Outfit for formal event.
-                        - "active": Outfit for doing sport.
-                        - "custom": Custom outfit.
-    """
-    if occasion == "casual":
-        return "T-shirt, Jeans and sneakers."
-    elif occasion == "formal":
-        return "White shirt, tie, blue suit and Oxford shoes."
-    elif occasion == "active":
-        return "Jogging trousers, T-shirt and trainers."
-    else:
-        return "Custom outfit for the fashion advisor."
-
-@tool
-def define_coordinates(city: str) -> dict:
-    """
-    Returns a dictionary with the geographical coordinates of a city.
-    Args:
-        city (str): A city name.
-    Returns:
-        A dictionary with the latitude, longitude and timezone of the city
-    """
-    city = city.strip()
-    if not city:
-        return {"error": "City name cannot be empty."}
-
-    url = f"https://geocoding-api.open-meteo.com/v1/search?name={city}&count=1"
-
-    try:
-        response = requests.get(url)
-        data = response.json()
-
-        if "results" not in data or not data["results"]:
-            return {"error": f"No location found for city: {city}"}
-
-        result = data["results"][0]
-        return {
-            "latitude": result["latitude"],
-            "longitude": result["longitude"],
-            "timezone": result["timezone"]
-        }
-
-    except Exception as e:
-        return {"error": str(e)}
-
-@tool
-def get_weather(city: str) -> dict:
-    """
-    Returns a dictionary with the current weather conditions in a city.
-    Args:
-        city (str): A city name.
-    Returns:
-        A dictionary with: temperature, wind_speed, weather_code, weather_description and a summary of the current weather.
-    """
-    coords = define_coordinates(city)
-    if "error" in coords:
-     return {"error": coords["error"]}
-
-    url = f'https://api.open-meteo.com/v1/forecast?latitude={coords["latitude"]}&longitude={coords["longitude"]}&current_weather=true&timezone={coords["timezone"]}'
-
-    try:
-        response = requests.get(url)
-        data = response.json()
-        temperature = data["current_weather"]["temperature"]
-        wind_speed = data["current_weather"]["windspeed"]
-        weather_code = data["current_weather"]["weathercode"]
-
-        return {
-                "temperature": temperature,
-                "wind_speed": wind_speed,
-                "weather_code": weather_code,
-           }
-
-
-    except Exception as e:
-        return {"error": str(e)}
-
-@tool
-def classify_weather(city: str) -> str:
-    """
-    Classifies the weather in the city based on a conversion table.
-    Args:
-      city (str): A city name.
-    Return:
-        Weather description based on the following conversion table:
-        0	Clear sky
-        1–3	Mainly clear to partly cloudy
-        45, 48	Fog and depositing rime fog
-        51–57	Drizzle (light to dense)
-        61–67	Rain (light to heavy)
-        71–77	Snow fall (light to heavy)
-        80–82	Rain showers
-        85–86	Snow showers
-        95	Thunderstorm
-        96–99	Thunderstorm with hail
-    """
-    weather = get_weather(city)
-    if "error" in weather:
-        return f"Unable to classify weather due to error: {weather['error']}"
-
-    temperature = weather["temperature"]
-    wind_speed = weather["wind_speed"]
-    weather_code = weather["weather_code"]
-
-    if weather_code == 0:
-        weather_description = "clear sky"
-    elif 1 <= weather_code <= 3:
-        weather_description = "mainly clear to partly cloudy"
-    elif weather_code in [45, 48]:
-        weather_description = "fog or depositing rime fog"
-    elif 51 <= weather_code <= 57:
-        weather_description = "drizzle (light to dense)"
-    elif 61 <= weather_code <= 67:
-        weather_description = "rain (light to heavy)"
-    elif 71 <= weather_code <= 77:
-        weather_description = "snow fall (light to heavy)"
-    elif 80 <= weather_code <= 82:
-        weather_description = "rain showers"
-    elif 85 <= weather_code <= 86:
-        weather_description = "snow showers"
-    elif weather_code == 95:
-        weather_description = "thunderstorm"
-    elif 96 <= weather_code <= 99:
-        weather_description = "thunderstorm with hail"
-    else:
-        weather_description = "unknown weather conditions"
-
-    return (
-        f"The current temperature in {city} is {temperature}°C, "
-        f"with a wind speed of {wind_speed} km/h. "
-        f"Weather description: {weather_description}."
-    )
-
-agent = CodeAgent(tools=[DuckDuckGoSearchTool(), suggest_outfit, define_coordinates, get_weather, classify_weather], model=InferenceClientModel())
-
-response = agent.run("""
-First, find the current weather in Rome.
-Then, classify the weather based on standard weather codes (e.g. clear, rain, snow, etc).
-Third, assess what kind of outfit would be comfortable for a park run given this weather.
-Fourth, call the appropriate tool to suggest an outfit for that context.
-Finally, summarize both the weather and outfit recommendations clearly in one paragraph.
-
-""")
-print(response)
-
-import json
-
-entry = {
-    "task_id": "task_id_rome_weather_outfit",
-    "model_answer": "The current weather in Rome is clear with a temperature of 29.5°C and a wind speed of 13.0 km/h. A comfortable outfit for a park run in this weather would be jogging trousers, a T-shirt, and trainers.",
-    "reasoning_trace": [
-        "Step 1: Called get_weather('Rome') to fetch current conditions.",
-        "Step 2: Used classify_weather('Rome') to interpret weather code (code 0 = clear sky).",
-        "Step 3: Determined that the task involves running in a park, which corresponds to the 'active' occasion.",
-        "Step 4: Called suggest_outfit('active') to retrieve clothing suitable for physical activity.",
-        "Step 5: Composed a final answer combining weather data and outfit suggestion into one summary."
-    ]
-}
-
-with open("submission.jsonl", "w") as f:
-    f.write(json.dumps(entry) + "\n")
-
-from google.colab import files
-files.download("submission.jsonl")
-
-more_entries = [
-    {
-        "task_id": "task_id_milan_weather_outfit",
-        "model_answer": "The current weather in Milan is mainly clear with a temperature of 22.3°C and wind speed of 9.5 km/h. A suitable outfit for a walk would be a light jacket, casual trousers, and comfortable shoes.",
-        "reasoning_trace": [
-            "Step 1: Fetch Milan weather using get_weather('Milan').",
-            "Step 2: Interpret code via classify_weather — partly cloudy.",
-            "Step 3: Occasion inferred as casual walk.",
-            "Step 4: Called suggest_outfit('casual').",
-            "Step 5: Merged weather and outfit into a final summary."
-        ]
-    },
-    # Add more entries here as needed
-]
-
-with open("submission.jsonl", "a") as f:
-    for entry in more_entries:
-        f.write(json.dumps(entry) + "\n")