Merge pull request #835 from carl-graup/carl-grp/week1_day2

Add carl-grp's homework for week1 day2 (cleared outputs)

community-contributions/carl-grp/week1_day2_homework.ipynb

@@ -0,0 +1,156 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "7316c8d3",
+   "metadata": {},
+   "source": [
+    "# How well can LLMs advice on tech purchases?"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "6acdeb48",
+   "metadata": {},
+   "source": [
+    "Just execute the two tries after setting everything up and take a look how both models perform in helping you choose the best new iPhone.\n",
+    "\n",
+    "I was suprised how much better ChatGPT is!\n",
+    "\n",
+    "P.S.: Since I'm German, the output and the websites are German. Just change the URLs if you want another language."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8ae52614",
+   "metadata": {},
+   "source": [
+    "### General setup"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d7f15f25",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import sys\n",
+    "import os\n",
+    "sys.path.append(os.path.join('..', '..', 'week1'))\n",
+    "from scraper import fetch_website_contents\n",
+    "from operator import concat\n",
+    "from openai import OpenAI\n",
+    "from IPython.display import Markdown, display\n",
+    "from dotenv import load_dotenv"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "bbd7dc61",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "system_prompt_homework_day2 = \"\"\"\n",
+    "You are tech advisor that knows about how newest mobiles like the iPhone work and what different user groups need, \n",
+    "like best cameras for Pro users or battery life for average consumers. \n",
+    "You advise fact-based and thorough on the best deals for each of your customers.\n",
+    "You deal with scraped websites given to you by ignoring random text from headers and footers of websites \n",
+    "and focusing on the content that actually presents products. \n",
+    "IMPORTANT: Only use facts that you got by the user via prompting.\n",
+    "Try hard to really identify which text is noise and which makes the product descriptions.\n",
+    "\"\"\"\n",
+    "user_prompt_prefix_homework_day2 = \"\"\"\n",
+    "Hi, as a student that still has the iPhone 13 Pro but wants a newer iPhone, \n",
+    "i need some advice on which iPhone to buy. Can you help me choose from apples website? \n",
+    "Here are the contents of apple's iphone websites that i obtained using a scraper.\n",
+    "Provide a short summary of the text from this website-scraper and afterwards help me choose the best option. \n",
+    "If it includes news or announcements, then summarize these too. \n",
+    "\"\"\"\n",
+    "\n",
+    "url_1 = \"https://www.apple.com/de/iphone-17-pro/\"\n",
+    "url_2 = \"https://www.apple.com/de/iphone-air/\"\n",
+    "url_3 = \"https://www.apple.com/de/iphone-17/\"\n",
+    "\n",
+    "urls = [url_1, url_2, url_3]\n",
+    "website_content_homework_day2 = \"\"\n",
+    "\n",
+    "for url in urls:\n",
+    "    website_content_homework_day2 += fetch_website_contents(url)\n",
+    "\n",
+    "homework_day2_message = [\n",
+    "    {\"role\":\"system\", \"content\":system_prompt_homework_day2},\n",
+    "    {\"role\":\"user\", \"content\": user_prompt_prefix_homework_day2 + website_content_homework_day2}\n",
+    "]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "cc854a00",
+   "metadata": {},
+   "source": [
+    "### Try 1: Local run with Ollama 3.2 "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7fae9b5b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!ollama pull llama3.2\n",
+    "\n",
+    "OLLAMA_BASE_URL = \"http://localhost:11434/v1\"\n",
+    "ollama = OpenAI(base_url=OLLAMA_BASE_URL, api_key='ollama')\n",
+    "\n",
+    "response = ollama.chat.completions.create(model=\"llama3.2\", messages=homework_day2_message,)\n",
+    "display(Markdown(response.choices[0].message.content))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e7563157",
+   "metadata": {},
+   "source": [
+    "### Try 2: Online with ChatGPT-5-nano"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d764cc76",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "load_dotenv(override=True)\n",
+    "\n",
+    "openai = OpenAI()\n",
+    "response = openai.chat.completions.create(model=\"gpt-5-nano\", messages=homework_day2_message,)\n",
+    "display(Markdown(response.choices[0].message.content))"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}