{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6bfa65e2-6aa7-45d3-a52d-8d6339f75501",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "* Running on local URL:  http://127.0.0.1:7860\n",
      "* To create a public link, set `share=True` in `launch()`.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div><iframe src=\"http://127.0.0.1:7860/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import gradio as gr\n",
    "from transformers import AutoTokenizer, AutoModelForCausalLM\n",
    "import torch\n",
    "\n",
    "# بارگذاری مدل و توکنایزر\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"HooshvareLab/gpt2-fa\")\n",
    "model = AutoModelForCausalLM.from_pretrained(\"HooshvareLab/gpt2-fa\")\n",
    "\n",
    "# تابع تولید متن با دستور\n",
    "def generate_with_instruction(instruction):\n",
    "    prompt = f\"دستور: {instruction}\\nپاسخ:\"\n",
    "    input_ids = tokenizer.encode(prompt, return_tensors=\"pt\")\n",
    "    output = model.generate(\n",
    "        input_ids,\n",
    "        max_new_tokens=900,\n",
    "        do_sample=True,\n",
    "        temperature=0.7,\n",
    "        top_k=50,\n",
    "        top_p=0.92,\n",
    "        repetition_penalty=1.3,\n",
    "        no_repeat_ngram_size=3,\n",
    "        pad_token_id=tokenizer.eos_token_id,\n",
    "        eos_token_id=tokenizer.eos_token_id\n",
    "    )\n",
    "    return tokenizer.decode(output[0], skip_special_tokens=True)\n",
    "\n",
    "# رابط گرافیکی\n",
    "gr.Interface(\n",
    "    fn=generate_with_instruction,\n",
    "    inputs=gr.Textbox(label=\"📝 دستور وارد کنید\", placeholder=\"مثلاً: درباره‌ی تاثیر ورزش بر ذهن بنویس\", lines=3),\n",
    "    outputs=gr.Textbox(label=\"📄 پاسخ مدل\"),\n",
    "    title=\"💬 دستور به مدل GPT2 فارسی\",\n",
    "    description=\"با وارد کردن دستور در قالب فارسی، مدل شروع به تولید متن می‌کند. مثل: نوشتن، خلاصه‌سازی یا ترجمه.\",\n",
    "    theme=\"soft\"\n",
    ").launch()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b693482a-21c0-483f-bb46-deb050ce82ee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "* Running on local URL:  http://127.0.0.1:7860\n",
      "* To create a public link, set `share=True` in `launch()`.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div><iframe src=\"http://127.0.0.1:7860/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import gradio as gr\n",
    "from transformers import GPT2LMHeadModel, GPT2Tokenizer\n",
    "import torch\n",
    "\n",
    "# بارگذاری مدل و توکنایزر\n",
    "model_name = \"HooshvareLab/gpt2-fa\"\n",
    "model = GPT2LMHeadModel.from_pretrained(model_name)\n",
    "tokenizer = GPT2Tokenizer.from_pretrained(model_name)\n",
    "\n",
    "# تنظیم مدل برای تولید متن\n",
    "model.eval()\n",
    "\n",
    "# تابعی برای تولید متن\n",
    "def generate_text(prompt):\n",
    "    input_ids = tokenizer.encode(prompt, return_tensors='pt')\n",
    "    \n",
    "    # تولید متن\n",
    "    with torch.no_grad():\n",
    "        output = model.generate(input_ids, max_length=100, num_return_sequences=1, no_repeat_ngram_size=2, temperature=0.7)\n",
    "\n",
    "    # تبدیل توکن‌های خروجی به متن\n",
    "    generated_text = tokenizer.decode(output[0], skip_special_tokens=True)\n",
    "    \n",
    "    return generated_text\n",
    "\n",
    "# تعریف رابط کاربری با Gradio\n",
    "iface = gr.Interface(fn=generate_text, inputs=\"text\", outputs=\"text\", live=True, title=\"تولید متن فارسی با مدل GPT-2\")\n",
    "\n",
    "# اجرا کردن رابط\n",
    "iface.launch()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fc890196-c4bf-4b01-ab3c-0796836f9355",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:base] *",
   "language": "python",
   "name": "conda-base-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}