{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7abd1d8d",
   "metadata": {},
   "source": [
    "# 🎭 Three Character Text Generators\n",
    "## Mickey Mouse 🐭 | Yoda 🟢 | Spider-Man 🕷️\n",
    "\n",
    "This notebook fine-tunes 3 separate LoRA adapters for each character and creates an interactive chat interface.\n",
    "\n",
    "**Characters:**\n",
    "- 🐭 **Mickey Mouse** - Cheerful and optimistic mouse from Toontown\n",
    "- 🟢 **Yoda** - Wise Jedi Master with unique backward speech pattern\n",
    "- 🕷️ **Spider-Man** - Friendly neighborhood web-slinging superhero"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ac3f3d3",
   "metadata": {},
   "source": [
    "---\n",
    "## 🚀 Google Colab Setup (FREE GPU!)\n",
    "\n",
    "**Follow these steps to run this notebook on Google Colab with free GPU:**\n",
    "\n",
    "1. **Upload to Google Drive:**\n",
    "   - Go to [Google Drive](https://drive.google.com)\n",
    "   - Upload this notebook: `CharacterTextGenerators.ipynb`\n",
    "   - Upload the 3 training data files: `mickeymouse.txt`, `yoda.txt`, `spiderman.txt`\n",
    "   - Keep all files in the same folder\n",
    "\n",
    "2. **Open in Colab:**\n",
    "   - Right-click `CharacterTextGenerators.ipynb` in Google Drive\n",
    "   - Select \"Open with\" → \"Google Colaboratory\"\n",
    "\n",
    "3. **Enable GPU:**\n",
    "   - In Colab, click \"Runtime\" → \"Change runtime type\"\n",
    "   - Set \"Hardware accelerator\" to \"T4 GPU\" (free tier)\n",
    "   - Click \"Save\"\n",
    "\n",
    "4. **Run the notebook:**\n",
    "   - The GPU check cell will confirm GPU is available\n",
    "   - Training will take ~10-15 minutes (instead of hours on CPU!)\n",
    "\n",
    "**Colab will automatically handle all library installations - no setup needed!**\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad548564",
   "metadata": {},
   "source": [
    "---\n",
    "## 1️⃣ Setup: Install Libraries and Load Training Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "776810af",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Install required libraries\n",
    "%pip install transformers peft datasets accelerate ipywidgets\n",
    "\n",
    "import torch\n",
    "from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments, Trainer\n",
    "from peft import LoraConfig, get_peft_model, TaskType, PeftModel\n",
    "from datasets import Dataset\n",
    "\n",
    "print(\"✅ Libraries imported successfully!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "29a56176",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check if GPU is available\n",
    "import torch\n",
    "\n",
    "print(\"🔍 Checking hardware availability...\\n\")\n",
    "\n",
    "if torch.cuda.is_available():\n",
    "    print(f\"✅ GPU Available: {torch.cuda.get_device_name(0)}\")\n",
    "    print(f\"   GPU Memory: {torch.cuda.get_device_properties(0).total_memory / 1024**3:.2f} GB\")\n",
    "    print(f\"   CUDA Version: {torch.version.cuda}\")\n",
    "    print(f\"   Current Device: cuda:{torch.cuda.current_device()}\")\n",
    "    print(\"\\n🎮 Training will use GPU acceleration!\")\n",
    "else:\n",
    "    print(\"⚠️ WARNING: No GPU detected!\")\n",
    "    print(\"   Training will use CPU (much slower)\")\n",
    "    print(\"   Consider using Google Colab or a system with GPU\")\n",
    "\n",
    "print(f\"\\n📊 PyTorch Version: {torch.__version__}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c305f195",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the character training data from text files\n",
    "exec(open('mickeymouse.txt').read())\n",
    "exec(open('yoda.txt').read())\n",
    "exec(open('spiderman.txt').read())\n",
    "\n",
    "print(f\"✅ Mickey Mouse: {len(train_data_mickey)} training examples\")\n",
    "print(f\"✅ Yoda: {len(train_data_yoda)} training examples\")\n",
    "print(f\"✅ Spider-Man: {len(train_data_spider_man)} training examples\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6e87ec7",
   "metadata": {},
   "source": [
    "---\n",
    "## 2️⃣ Load Base Model and Tokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8bcba967",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Choose a small, efficient model\n",
    "model_name = \"Qwen/Qwen2-0.5B-Instruct\"\n",
    "\n",
    "# Load tokenizer\n",
    "tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)\n",
    "tokenizer.pad_token = tokenizer.eos_token\n",
    "tokenizer.padding_side = \"right\"\n",
    "\n",
    "# Load base model\n",
    "base_model = AutoModelForCausalLM.from_pretrained(\n",
    "    model_name,\n",
    "    torch_dtype=torch.float16,\n",
    "    device_map=\"auto\",\n",
    "    trust_remote_code=True\n",
    ")\n",
    "\n",
    "print(f\"🔹 Base model loaded: {model_name}\")\n",
    "print(f\"🔹 Parameters: {base_model.num_parameters():,}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83f6f8f7",
   "metadata": {},
   "source": [
    "---\n",
    "## 3️⃣ Training Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8f1fe22c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_character(character_name, train_data, output_dir, epochs=3):\n",
    "    \"\"\"\n",
    "    Fine-tune a LoRA adapter for a specific character\n",
    "    \n",
    "    Args:\n",
    "        character_name: Name of the character\n",
    "        train_data: List of training examples\n",
    "        output_dir: Directory to save the adapter\n",
    "        epochs: Number of training epochs\n",
    "    \"\"\"\n",
    "    print(f\"\\n{'='*70}\")\n",
    "    print(f\"🎭 TRAINING {character_name.upper()}\")\n",
    "    print(f\"{'='*70}\\n\")\n",
    "    \n",
    "    # Reload base model fresh for each character\n",
    "    model = AutoModelForCausalLM.from_pretrained(\n",
    "        model_name,\n",
    "        torch_dtype=torch.float16,\n",
    "        device_map=\"auto\",\n",
    "        trust_remote_code=True\n",
    "    )\n",
    "    \n",
    "    # Configure LoRA\n",
    "    lora_config = LoraConfig(\n",
    "        r=16,\n",
    "        lora_alpha=32,\n",
    "        target_modules=[\"q_proj\", \"v_proj\", \"k_proj\", \"o_proj\"],\n",
    "        lora_dropout=0.05,\n",
    "        bias=\"none\",\n",
    "        task_type=TaskType.CAUSAL_LM\n",
    "    )\n",
    "    \n",
    "    # Apply LoRA\n",
    "    model = get_peft_model(model, lora_config)\n",
    "    print(f\"✅ LoRA adapter applied\")\n",
    "    model.print_trainable_parameters()\n",
    "    \n",
    "    # Format dataset using chat template\n",
    "    def format_chat(example):\n",
    "        text = tokenizer.apply_chat_template(\n",
    "            example[\"messages\"],\n",
    "            tokenize=False,\n",
    "            add_generation_prompt=False\n",
    "        )\n",
    "        return {\"text\": text}\n",
    "    \n",
    "    dataset = Dataset.from_list(train_data)\n",
    "    dataset = dataset.map(format_chat)\n",
    "    \n",
    "    # Tokenize\n",
    "    def tokenize_function(examples):\n",
    "        result = tokenizer(\n",
    "            examples[\"text\"],\n",
    "            truncation=True,\n",
    "            padding=\"max_length\",\n",
    "            max_length=256\n",
    "        )\n",
    "        result[\"labels\"] = result[\"input_ids\"].copy()\n",
    "        return result\n",
    "    \n",
    "    tokenized_dataset = dataset.map(tokenize_function, batched=True, remove_columns=[\"text\", \"messages\"])\n",
    "    \n",
    "    # Training arguments\n",
    "    training_args = TrainingArguments(\n",
    "        output_dir=output_dir,\n",
    "        num_train_epochs=epochs,\n",
    "        per_device_train_batch_size=2,\n",
    "        gradient_accumulation_steps=2,\n",
    "        learning_rate=2e-4,\n",
    "        logging_steps=5,  # Log more frequently for better progress visibility\n",
    "        save_strategy=\"epoch\",\n",
    "        report_to=\"none\",\n",
    "        fp16=True,\n",
    "        use_cpu=False,  # Force GPU usage\n",
    "        no_cuda=False,  # Enable CUDA\n",
    "        disable_tqdm=False,  # Enable progress bars\n",
    "        logging_first_step=True,  # Show initial progress\n",
    "    )\n",
    "    \n",
    "    # Train\n",
    "    trainer = Trainer(\n",
    "        model=model,\n",
    "        args=training_args,\n",
    "        train_dataset=tokenized_dataset,\n",
    "        tokenizer=tokenizer,\n",
    "    )\n",
    "    \n",
    "    print(f\"\\n🔥 Starting training for {character_name}...\\n\")\n",
    "    trainer.train()\n",
    "    \n",
    "    # Save\n",
    "    model.save_pretrained(output_dir)\n",
    "    tokenizer.save_pretrained(output_dir)\n",
    "    \n",
    "    print(f\"\\n✅ {character_name} adapter saved to {output_dir}\\n\")\n",
    "    \n",
    "    # Clear memory\n",
    "    del model\n",
    "    del trainer\n",
    "    torch.cuda.empty_cache()\n",
    "    \n",
    "    return output_dir\n",
    "\n",
    "print(\"✅ Training function defined!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e520227",
   "metadata": {},
   "source": [
    "---\n",
    "## 4️⃣ Train All Three Characters\n",
    "\n",
    "⚠️ **Note:** This will take approximately 10-15 minutes on a GPU (T4 or better recommended)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e04031a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train Mickey Mouse\n",
    "# Note: Press the stop button (■) in the notebook toolbar to interrupt training if needed\n",
    "\n",
    "import signal\n",
    "import sys\n",
    "\n",
    "def signal_handler(sig, frame):\n",
    "    print('\\n⚠️ Training interrupted by user!')\n",
    "    sys.exit(0)\n",
    "\n",
    "signal.signal(signal.SIGINT, signal_handler)\n",
    "\n",
    "print(\"🐭 Training Mickey Mouse...\")\n",
    "print(\"💡 Tip: Click the stop button (■) to interrupt training\\n\")\n",
    "\n",
    "mickey_adapter = train_character(\n",
    "    character_name=\"Mickey Mouse\",\n",
    "    train_data=train_data_mickey,\n",
    "    output_dir=\"./mickey-lora-adapter\",\n",
    "    epochs=3\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d4fa5595",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train Yoda\n",
    "# Note: Press the stop button (■) in the notebook toolbar to interrupt training if needed\n",
    "\n",
    "import signal\n",
    "import sys\n",
    "\n",
    "def signal_handler(sig, frame):\n",
    "    print('\\n⚠️ Training interrupted by user!')\n",
    "    sys.exit(0)\n",
    "\n",
    "signal.signal(signal.SIGINT, signal_handler)\n",
    "\n",
    "print(\"🟢 Training Yoda...\")\n",
    "print(\"💡 Tip: Click the stop button (■) to interrupt training\\n\")\n",
    "\n",
    "yoda_adapter = train_character(\n",
    "    character_name=\"Yoda\",\n",
    "    train_data=train_data_yoda,\n",
    "    output_dir=\"./yoda-lora-adapter\",\n",
    "    epochs=3\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b703cf4e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train Spider-Man\n",
    "# Note: Press the stop button (■) in the notebook toolbar to interrupt training if needed\n",
    "\n",
    "import signal\n",
    "import sys\n",
    "\n",
    "def signal_handler(sig, frame):\n",
    "    print('\\n⚠️ Training interrupted by user!')\n",
    "    sys.exit(0)\n",
    "\n",
    "signal.signal(signal.SIGINT, signal_handler)\n",
    "\n",
    "print(\"🕷️ Training Spider-Man...\")\n",
    "print(\"💡 Tip: Click the stop button (■) to interrupt training\\n\")\n",
    "\n",
    "spiderman_adapter = train_character(\n",
    "    character_name=\"Spider-Man\",\n",
    "    train_data=train_data_spider_man,\n",
    "    output_dir=\"./spiderman-lora-adapter\",\n",
    "    epochs=3\n",
    ")\n",
    "\n",
    "print(\"\\n\" + \"=\"*70)\n",
    "print(\"🎉 ALL THREE CHARACTERS TRAINED SUCCESSFULLY!\")\n",
    "print(\"=\"*70)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f315ffab",
   "metadata": {},
   "source": [
    "---\n",
    "## 5️⃣ Character Chat System\n",
    "\n",
    "Create an interactive chat interface to talk with any of the three characters!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "08114e83",
   "metadata": {},
   "outputs": [],
   "source": [
    "class CharacterChat:\n",
    "    \"\"\"Interactive chat with multiple character personalities\"\"\"\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.characters = {\n",
    "            \"Mickey Mouse\": {\n",
    "                \"adapter\": \"./mickey-lora-adapter\",\n",
    "                \"emoji\": \"🐭\",\n",
    "                \"description\": \"Cheerful and optimistic mouse from Toontown\"\n",
    "            },\n",
    "            \"Yoda\": {\n",
    "                \"adapter\": \"./yoda-lora-adapter\",\n",
    "                \"emoji\": \"🟢\",\n",
    "                \"description\": \"Wise Jedi Master who speaks in unique way\"\n",
    "            },\n",
    "            \"Spider-Man\": {\n",
    "                \"adapter\": \"./spiderman-lora-adapter\",\n",
    "                \"emoji\": \"🕷️\",\n",
    "                \"description\": \"Friendly neighborhood web-slinging hero\"\n",
    "            }\n",
    "        }\n",
    "        \n",
    "        self.current_character = None\n",
    "        self.current_model = None\n",
    "        self.base_model = None\n",
    "        self.tokenizer = None\n",
    "        \n",
    "    def load_character(self, character_name):\n",
    "        \"\"\"Load a specific character's LoRA adapter\"\"\"\n",
    "        if character_name not in self.characters:\n",
    "            print(f\"❌ Character '{character_name}' not found!\")\n",
    "            return False\n",
    "        \n",
    "        print(f\"\\n🔄 Loading {character_name}...\")\n",
    "        \n",
    "        # Clear previous model if exists\n",
    "        if self.current_model is not None:\n",
    "            del self.current_model\n",
    "            torch.cuda.empty_cache()\n",
    "        \n",
    "        # Load base model if not loaded\n",
    "        if self.base_model is None:\n",
    "            self.base_model = AutoModelForCausalLM.from_pretrained(\n",
    "                model_name,\n",
    "                torch_dtype=torch.float16,\n",
    "                device_map=\"auto\",\n",
    "                trust_remote_code=True\n",
    "            )\n",
    "        \n",
    "        # Load tokenizer if not loaded\n",
    "        if self.tokenizer is None:\n",
    "            adapter_path = self.characters[character_name][\"adapter\"]\n",
    "            self.tokenizer = AutoTokenizer.from_pretrained(adapter_path)\n",
    "        \n",
    "        # Load character's LoRA adapter\n",
    "        adapter_path = self.characters[character_name][\"adapter\"]\n",
    "        self.current_model = PeftModel.from_pretrained(self.base_model, adapter_path)\n",
    "        self.current_model.eval()\n",
    "        \n",
    "        self.current_character = character_name\n",
    "        \n",
    "        emoji = self.characters[character_name][\"emoji\"]\n",
    "        print(f\"✅ {emoji} {character_name} loaded and ready to chat!\\n\")\n",
    "        return True\n",
    "    \n",
    "    def chat(self, user_message, max_tokens=50, temperature=0.7):\n",
    "        \"\"\"Generate a response from the current character\"\"\"\n",
    "        if self.current_model is None:\n",
    "            return \"❌ No character loaded! Please load a character first.\"\n",
    "        \n",
    "        # Format message using chat template\n",
    "        messages = [{\"role\": \"user\", \"content\": user_message}]\n",
    "        prompt = self.tokenizer.apply_chat_template(\n",
    "            messages, \n",
    "            tokenize=False, \n",
    "            add_generation_prompt=True\n",
    "        )\n",
    "        \n",
    "        # Generate response\n",
    "        inputs = self.tokenizer(prompt, return_tensors=\"pt\").to(self.current_model.device)\n",
    "        \n",
    "        with torch.no_grad():\n",
    "            outputs = self.current_model.generate(\n",
    "                **inputs,\n",
    "                max_new_tokens=max_tokens,\n",
    "                do_sample=True,\n",
    "                temperature=temperature,\n",
    "                top_p=0.9,\n",
    "                repetition_penalty=1.1,\n",
    "                pad_token_id=self.tokenizer.eos_token_id\n",
    "            )\n",
    "        \n",
    "        # Decode response\n",
    "        response = self.tokenizer.decode(\n",
    "            outputs[0][len(inputs['input_ids'][0]):], \n",
    "            skip_special_tokens=True\n",
    "        )\n",
    "        \n",
    "        return response\n",
    "    \n",
    "    def show_characters(self):\n",
    "        \"\"\"Display available characters\"\"\"\n",
    "        print(\"\\n\" + \"=\"*70)\n",
    "        print(\"🎭 AVAILABLE CHARACTERS:\")\n",
    "        print(\"=\"*70)\n",
    "        for i, (name, info) in enumerate(self.characters.items(), 1):\n",
    "            emoji = info[\"emoji\"]\n",
    "            desc = info[\"description\"]\n",
    "            current = \" ⭐ (CURRENT)\" if name == self.current_character else \"\"\n",
    "            print(f\"\\n{i}. {emoji} {name}{current}\")\n",
    "            print(f\"   {desc}\")\n",
    "        print(\"\\n\" + \"=\"*70 + \"\\n\")\n",
    "\n",
    "# Create the chat instance\n",
    "chat = CharacterChat()\n",
    "print(\"✅ Character Chat System initialized!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9365b754",
   "metadata": {},
   "source": [
    "---\n",
    "## 6️⃣ View Available Characters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "96b9749c",
   "metadata": {},
   "outputs": [],
   "source": [
    "chat.show_characters()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7f6d0a1",
   "metadata": {},
   "source": [
    "---\n",
    "## 7️⃣ Chat with Mickey Mouse 🐭"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d1098b5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "chat.load_character(\"Mickey Mouse\")\n",
    "\n",
    "# Test with different prompts\n",
    "test_prompts = [\n",
    "    \"Hello! How are you today?\",\n",
    "    \"What's your favorite thing to do?\",\n",
    "    \"Tell me about your friends\",\n",
    "]\n",
    "\n",
    "for prompt in test_prompts:\n",
    "    response = chat.chat(prompt)\n",
    "    print(f\"👤 You: {prompt}\")\n",
    "    print(f\"🐭 Mickey: {response}\")\n",
    "    print(\"-\" * 70 + \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "785e2f85",
   "metadata": {},
   "source": [
    "---\n",
    "## 8️⃣ Chat with Yoda 🟢"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "679a53f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "chat.load_character(\"Yoda\")\n",
    "\n",
    "test_prompts = [\n",
    "    \"Hello! How are you today?\",\n",
    "    \"What is the Force?\",\n",
    "    \"Can you teach me to be a Jedi?\",\n",
    "]\n",
    "\n",
    "for prompt in test_prompts:\n",
    "    response = chat.chat(prompt)\n",
    "    print(f\"👤 You: {prompt}\")\n",
    "    print(f\"🟢 Yoda: {response}\")\n",
    "    print(\"-\" * 70 + \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03881450",
   "metadata": {},
   "source": [
    "---\n",
    "## 9️⃣ Chat with Spider-Man 🕷️"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77515949",
   "metadata": {},
   "outputs": [],
   "source": [
    "chat.load_character(\"Spider-Man\")\n",
    "\n",
    "test_prompts = [\n",
    "    \"Hello! How are you today?\",\n",
    "    \"How do you swing between buildings?\",\n",
    "    \"What's it like being a hero?\",\n",
    "]\n",
    "\n",
    "for prompt in test_prompts:\n",
    "    response = chat.chat(prompt)\n",
    "    print(f\"👤 You: {prompt}\")\n",
    "    print(f\"🕷️ Spider-Man: {response}\")\n",
    "    print(\"-\" * 70 + \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a3ac6fea",
   "metadata": {},
   "source": [
    "---\n",
    "## 🔟 Interactive Chat Interface\n",
    "\n",
    "Chat freely with any character! Switch between them anytime."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "350d85ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "def interactive_chat():\n",
    "    \"\"\"\n",
    "    Interactive chat interface - chat with any character!\n",
    "    \n",
    "    Commands:\n",
    "    - Type your message to chat with the current character\n",
    "    - Type 'switch' to change character\n",
    "    - Type 'list' to see available characters\n",
    "    - Type 'quit' to exit\n",
    "    \"\"\"\n",
    "    \n",
    "    print(\"\\n\" + \"=\"*70)\n",
    "    print(\"🎭 INTERACTIVE CHARACTER CHAT\")\n",
    "    print(\"=\"*70)\n",
    "    print(\"\\nCommands:\")\n",
    "    print(\"  - Type your message to chat with the current character\")\n",
    "    print(\"  - Type 'switch' to change character\")\n",
    "    print(\"  - Type 'list' to see available characters\")\n",
    "    print(\"  - Type 'quit' to exit\")\n",
    "    print(\"\\n\" + \"=\"*70 + \"\\n\")\n",
    "    \n",
    "    # Load first character if none loaded\n",
    "    if chat.current_character is None:\n",
    "        chat.load_character(\"Mickey Mouse\")\n",
    "    \n",
    "    while True:\n",
    "        # Get current character info\n",
    "        emoji = chat.characters[chat.current_character][\"emoji\"]\n",
    "        \n",
    "        # Get user input\n",
    "        user_input = input(f\"\\n👤 You: \").strip()\n",
    "        \n",
    "        if not user_input:\n",
    "            continue\n",
    "        \n",
    "        # Handle commands\n",
    "        if user_input.lower() == 'quit':\n",
    "            print(\"\\n👋 Goodbye! Thanks for chatting!\\n\")\n",
    "            break\n",
    "        \n",
    "        elif user_input.lower() == 'list':\n",
    "            chat.show_characters()\n",
    "            continue\n",
    "        \n",
    "        elif user_input.lower() == 'switch':\n",
    "            chat.show_characters()\n",
    "            \n",
    "            character_choice = input(\"Enter character name (or press Enter to cancel): \").strip()\n",
    "            \n",
    "            if character_choice and character_choice in chat.characters:\n",
    "                chat.load_character(character_choice)\n",
    "            elif character_choice:\n",
    "                print(f\"❌ '{character_choice}' not found. Staying with {chat.current_character}\")\n",
    "            \n",
    "            continue\n",
    "        \n",
    "        # Generate response\n",
    "        response = chat.chat(user_input)\n",
    "        print(f\"{emoji} {chat.current_character}: {response}\")\n",
    "\n",
    "# Ready to start\n",
    "print(\"💬 Ready to start interactive chat!\")\n",
    "print(\"Run: interactive_chat()\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ecacfb59",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Uncomment and run to start chatting!\n",
    "# interactive_chat()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b8a32fc",
   "metadata": {},
   "source": [
    "---\n",
    "## 🎪 BONUS: Character Comparison\n",
    "\n",
    "Ask the same question to all three characters and compare their responses!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "12927d9c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compare_characters(question):\n",
    "    \"\"\"Ask the same question to all three characters\"\"\"\n",
    "    \n",
    "    print(\"\\n\" + \"=\"*70)\n",
    "    print(f\"❓ QUESTION: {question}\")\n",
    "    print(\"=\"*70 + \"\\n\")\n",
    "    \n",
    "    for character_name in [\"Mickey Mouse\", \"Yoda\", \"Spider-Man\"]:\n",
    "        chat.load_character(character_name)\n",
    "        response = chat.chat(question, max_tokens=60)\n",
    "        \n",
    "        emoji = chat.characters[character_name][\"emoji\"]\n",
    "        print(f\"{emoji} {character_name}:\")\n",
    "        print(f\"   {response}\\n\")\n",
    "        print(\"-\" * 70 + \"\\n\")\n",
    "\n",
    "print(\"✅ Comparison function ready!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b0832fbc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Test with various questions\n",
    "questions = [\n",
    "    \"What makes you happy?\",\n",
    "    \"What's your biggest challenge?\",\n",
    "    \"Do you have any advice for me?\",\n",
    "]\n",
    "\n",
    "for q in questions:\n",
    "    compare_characters(q)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5971ef5d",
   "metadata": {},
   "source": [
    "---\n",
    "## 🎯 Summary\n",
    "\n",
    "You've successfully created 3 character text generators!\n",
    "\n",
    "**What you can do:**\n",
    "1. ✅ Chat with Mickey Mouse, Yoda, or Spider-Man\n",
    "2. ✅ Switch between characters anytime\n",
    "3. ✅ Compare how each character responds to the same question\n",
    "4. ✅ Use the interactive chat for natural conversations\n",
    "\n",
    "**Each character maintains their unique personality:**\n",
    "- 🐭 Mickey Mouse: Cheerful, optimistic, uses \"Oh boy!\" and \"Haha!\"\n",
    "- 🟢 Yoda: Wise, speaks in reverse syntax, uses \"Hmm\" and \"Yes\"\n",
    "- 🕷️ Spider-Man: Witty, heroic, makes jokes about swinging and web-slinging\n",
    "\n",
    "**Try your own prompts and have fun chatting!** 🎉"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}