{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1.模型保存与加载\n",
    "\n",
    "Q:pytorch中的序列化与反序列化\n",
    "- `torch.save(obj, f, pickle_module=<module 'pickle' from '/home/v-yuega/miniconda/envs/torch_doc/lib/python3.7/pickle.py'>, pickle_protocol=2, _use_new_zipfile_serialization=True)`\n",
    "- obj:对象，f:输出路径\n",
    "- `torch.load(f, map_location=None, pickle_module=<module 'pickle' from '/home/v-yuega/miniconda/envs/torch_doc/lib/python3.7/pickle.py'>, **pickle_load_args)`\n",
    "- f：文件路径，map_location：指定存放位置,cpu or gpu\n",
    "\n",
    "Q:模型保存的两种方式是什么？\n",
    "- 1.保存整个Module：torch.save(net, path)\n",
    "- 2.保存模型参数：torch.save(net.state_dict(), path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2.模型finetune\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3.GPU的使用\n",
    "\n",
    "Q:to函数如何转换数据类型和设备？\n",
    "- ```python\n",
    "x = torch.ones((3, 3))\n",
    "x = x.to(torch.float64)\n",
    "\n",
    "x = torch.ones((3, 3))\n",
    "x = x.to(\"cuda\")\n",
    "\n",
    "linear = nn.Linear(2, 2)\n",
    "linear.to(torch.double)\n",
    "\n",
    "gpu1 = torch.device(\"cuda\")\n",
    "linear.to(gpu1)\n",
    "```\n",
    "- 注意：张量不执行inplace，模型执行inplace"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}