Refactor code generation examples: simplify Fibonacci and palindrome functions, and enhance flatten function implementation

Author: abossard

notebooks/01_evaluation_and_tuning.ipynb

@@ -1087,47 +1087,71 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
+      "⚠️ Für diesen Test nutzen wir absichtlich ein kleineres Modell: github_copilot/gpt-4o-mini\n",
+      "\n",
       "📋 Task: Code Generation\n",
       "   Metrik: Echte Ausführung mit Testdaten!\n",
       "\n",
       "📝 Aufgabe: Write a function that returns the nth Fibonacci number using iteration.\n",
       "   Generierter Code:\n",
-      "   │ def fibonacci(n: int) -> int:\n",
-      "   │     \"\"\"\n",
-      "   │     Return the nth Fibonacci number using iteration.\n",
-      "   │ \n",
-      "   │     The sequence is defined as:\n",
-      "   │         F(0) = 0\n",
-      "   │         F(1) = 1\n",
-      "   │         F(n) = F(n-1) + F(n-2) for n >= 2\n",
-      "   │ ... (47 Zeilen)\n",
+      "   │ def fibonacci(n):\n",
+      "   │     if n <= 0:\n",
+      "   │         return 0\n",
+      "   │     elif n == 1:\n",
+      "   │         return 1\n",
+      "   │         \n",
+      "   │     a, b = 0, 1\n",
+      "   │     for _ in range(2, n + 1):\n",
+      "   │ ... (10 Zeilen)\n",
       "   ✅ fibonacci(0) → 0\n",
       "   ✅ fibonacci(1) → 1\n",
       "   ✅ fibonacci(10) → 55\n",
       "   ✅ fibonacci(20) → 6765\n",
       "\n",
       "📝 Aufgabe: Write a function that checks if a string is a palindrome, ignoring case and non-\n",
       "   Generierter Code:\n",
-      "   │ def is_palindrome(s: str) -> bool:\n",
-      "   │     \"\"\"\n",
-      "   │     Check if the given string is a palindrome, ignoring case and non-alphanumeric characters.\n",
+      "   │ import re\n",
       "   │ \n",
-      "   │     Args:\n",
-      "   │         s (str): The input string to check.\n",
+      "   │ def is_palindrome(s: str) -> bool:\n",
+      "   │     # Remove non-alphanumeric characters and convert to lower case\n",
+      "   │     cleaned_string = re.sub(r'[^a-zA-Z0-9]', '', s).lower()\n",
+      "   │     # Check if the cleaned string is the same as its reverse\n",
+      "   │     return cleaned_string == cleaned_string[::-1]\n",
       "   │ \n",
-      "   │     Returns:\n",
-      "   │ ... (28 Zeilen)\n",
+      "   │ ... (11 Zeilen)\n",
       "   ✅ is_palindrome('racecar') → True\n",
       "   ✅ is_palindrome('hello') → False\n",
       "   ✅ is_palindrome('A man a plan a canal Panama') → True\n",
       "\n",
-      "📝 Aufgabe: Write a function that flattens a nested list of arbitrary depth into a single li\n"
+      "📝 Aufgabe: Write a function that flattens a nested list of arbitrary depth into a single li\n",
+      "   Generierter Code:\n",
+      "   │ def flatten(nested_list):\n",
+      "   │     flat_list = []\n",
+      "   │     for element in nested_list:\n",
+      "   │         if isinstance(element, list):\n",
+      "   │             flat_list.extend(flatten(element))  # Recursively flatten the sublist\n",
+      "   │         else:\n",
+      "   │             flat_list.append(element)  # Add the non-list element to the flat list\n",
+      "   │     return flat_list\n",
+      "   │ ... (13 Zeilen)\n",
+      "[1, 2, 3, 4, 5, 6, 7, 8]\n",
+      "   ✅ flatten([1, [2], 3]) → [1, 2, 3]\n",
+      "   ✅ flatten([1, [2, [3, [4]]]]) → [1, 2, 3, 4]\n",
+      "\n",
+      "📝 Aufgabe: Write a function that counts the frequency of each word in a string and returns \n"
      ]
     }
    ],
    "source": [
     "import re\n",
     "\n",
+    "# Für Code-Generierung testen wir absichtlich ein KLEINERES Modell\n",
+    "# um zu zeigen dass nicht jedes Modell alles kann\n",
+    "SMALL_MODEL = \"github_copilot/gpt-4o-mini\"\n",
+    "configure_dspy(SMALL_MODEL)\n",
+    "print(f\"⚠️ Für diesen Test nutzen wir absichtlich ein kleineres Modell: {SMALL_MODEL}\")\n",
+    "print()\n",
+    "\n",
     "task = get_task(\"code_generation\")\n",
     "print(f\"📋 Task: {task.name}\")\n",
     "print(f\"   Metrik: Echte Ausführung mit Testdaten!\")\n",
@@ -1265,7 +1289,10 @@
     "print(\"   Ob er die RICHTIGEN ERGEBNISSE liefert — das zählt!\")\n",
     "print()\n",
     "print(\"   Das ist wie bei Software-Tests:\")\n",
-    "print(\"   Ein Programm das startet aber falsch rechnet ist wertlos.\")"
+    "print(\"   Ein Programm das startet aber falsch rechnet ist wertlos.\")\n",
+    "\n",
+    "# Zurück zum Haupt-Modell für die nächsten Zellen\n",
+    "configure_dspy(MODEL)"
    ]
   },
   {