{
 "cells": [
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "# 快速入门\n",
    "本章为TorchAir图模式实操入门章节，承接上一章理论原理内容，章节从环境校验起步，提供标准化 SimpleModel 示例工程，完整演示`npugraph_ex` 图模式代码的编写与运行流程。\n",
    "\n",
    "---\n",
    "\n",
    "## 前置要求\n",
    "\n",
    "在开始本教程之前，建议你具备以下硬件环境：\n",
    "- 确认有可用的NPU环境，推荐使用[CANNLab](https://devstation.connect.huaweicloud.com/devdesktop/webide)一站式开发平台\n",
    "- CANN 版本、torch 与 torch_npu 版本配套，[版本配套关系参考](https://gitcode.com/Ascend/pytorch#ascend-extension-for-pytorch%E7%89%88%E6%9C%AC%E9%85%8D%E5%A5%97%E8%A1%A8)\n",
    "---\n",
    "\n",
    "## 章节目标\n",
    "完成本课程后，你将能够：\n",
    "\n",
    "- 能够独立完成昇腾NPU环境校验，熟练编写适配`npugraph_ex`后端的完整推理代码，掌握torch.compile关键参数配置规范；\n",
    "- 区分原生 `Eager` 逐算子执行模式与 `npugraph_ex` 图模式的代码写法，理解两种模式在执行流程、推理时延上的直观差异；\n",
    "- 掌握昇腾 Profiler 性能采集代码编写方法，可分别采集 `Eager`、`npugraph_ex` 两种模式的性能 Trace 文件，并分析两者的性能收益。\n",
    "\n",
    "## 章节内容\n",
    "* [2.1 章节介绍](02.01_chapter_intro.ipynb)\n",
    "* [2.2 快速入门](02.02_quick_start.ipynb)\n",
    "* [2.3 章节练习](02.03_chapter_practice.ipynb)\n",
    "\n",
    "---\n",
    "\n",
    "本章课程请从 [快速入门](02.02_quick_start.ipynb) 开始学习。\n"
   ],
   "id": "c57dd7a60cf798ae"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}