ppt_notes/00_3_Notebook基础操作.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Hi 大家好,本次课程基于AI Studio平台进行，请各位同学先熟悉下Notebook基本操作～\n",
    "\n",
    "如遇问题请查阅以下文档或在课程QQ群（群号618354318）中提问～\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/6e998a2cea474b638559f3a1a85ea9f93aef571570894d88babacce0a50220c0)）\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# 项目启停\n",
    "\n",
    "当进入自己项目的详情页面时, 用户可以选择\"运行\"项目, 也就是准备项目环境. \n",
    "\n",
    "同样的, 当不想继续时, 可以此页面点击\"停止\"以终止项目. \n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/897ace5ff04b49f8896db4e742f61c207e0bd990722640438df5eb75fcb7c264)\n",
    "\n",
    "---\n",
    "\n",
    "当启动项目时, 如个人信息完整, 则可以获得算力奖励\n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/78ce19b45dea432ea08d34b2f69474972fb357b16dc2476384499a63e8e9b5e8)\n",
    "\n",
    "\n",
    "---\n",
    "主动填写个人信息的入口在右上角的个人中心里\n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/fb558d5a532f4c28bbe33fe376c8a2af330ed7437b9542e8ab83c8dc66fb22a7)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# 执行和调试\n",
    "\n",
    "基于Notebook的方案和传统IDE方案略有不同, 请见下图:\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/136c52a305814e20a6f026722f080051d219b216e005432c9a9ac73e079ddbb8)\n",
    "\n",
    "插入断点则需要使用Python自带Debugger: PDB.\n",
    "用法请点击这里: \n",
    "[PDB用法](https://aistudio.baidu.com/aistudio/projectdetail/69987)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# 多文件代码编辑\n",
    "\n",
    "本系统的Notebook支持多文件编辑, 支持.py, .json, .txt, .log等格式的在线编辑, 支持部分图片类型文件的在线预览. \n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/222e72279ad74d1c81fdd250ec61664de4297161cac540e686fae4db2ac942c2)\n",
    "\n",
    "\n",
    "**上传文件的体积是30MB. 如果需要引用更大体积的文件, 请使用数据集功能**. \n",
    "\n",
    "用户也可以使用命令, `!cat <<newfile > newfile.py` 在项目空间内直接创建文件, 之后双击进行编辑. 如下图所示:\n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/6bb509c98c5d4b37859e3fb06287e1e01e04c670f6604de6b2e405ac9e57344a)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# 上传Notebook\n",
    "\n",
    "可以用自己的ipynb文件取代当前默认的Notebook(前提是格式合法)\n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/1c4d746dc96642abac50590eb21e8783688d04c6b07347079a1a5e1887093a76)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Notebook快捷键\n",
    "\n",
    "\n",
    "和原生Notebook快捷键基本一致. \n",
    "\n",
    "快捷键分成两种状态下的. 一个是命令模式, 一个是编辑模式, 掌握之后事半功倍. \n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/3a9f826518cb42e89965641e4eea9a289dbcce74862e4335ba8738d978e37fc2)\n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/968cf800086a47129fdda653891c170a1c9b1248296747aeaa766bb5cef2932c)\n",
    "\n",
    "\n",
    "---\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "#  **Notebook中使用Shell命令**\n",
    "\n",
    "通过在Shell命令前添加! (感叹号), 就可以执行部分Shell命令. 包括诸如 !pip install这样的命令. \n",
    "不过, !apt-get这种可能引发用户进一步操作的命令是不支持的. \n",
    "\n",
    "下面是两个示例."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# 查看当前挂载的数据集目录\n",
    "!ls /home/aistudio/data/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/aistudio/work\r\n"
     ]
    }
   ],
   "source": [
    "#显示当前路径\n",
    "\n",
    "!pwd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "### 使用pip来安装自己需要的package (但不支持apt-get)\n",
    "\n",
    "例如: \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Looking in indexes: https://mirror.baidu.com/pypi/simple/\n",
      "Requirement already satisfied: jupyterthemes in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (0.20.0)\n",
      "Requirement already satisfied: ipython>=5.4.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jupyterthemes) (7.0.1)\n",
      "Requirement already satisfied: jupyter-core in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jupyterthemes) (4.5.0)\n",
      "Requirement already satisfied: matplotlib>=1.4.3 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jupyterthemes) (2.2.3)\n",
      "Requirement already satisfied: lesscpy>=0.11.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jupyterthemes) (0.14.0)\n",
      "Requirement already satisfied: notebook>=5.6.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jupyterthemes) (5.7.0)\n",
      "Requirement already satisfied: setuptools>=18.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (56.2.0)\n",
      "Requirement already satisfied: jedi>=0.10 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (0.17.0)\n",
      "Requirement already satisfied: decorator in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (4.4.2)\n",
      "Requirement already satisfied: pickleshare in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (0.7.5)\n",
      "Requirement already satisfied: simplegeneric>0.8 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (0.8.1)\n",
      "Requirement already satisfied: traitlets>=4.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (4.3.3)\n",
      "Requirement already satisfied: prompt_toolkit<2.1.0,>=2.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (2.0.10)\n",
      "Requirement already satisfied: pygments in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (2.4.2)\n",
      "Requirement already satisfied: backcall in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (0.1.0)\n",
      "Requirement already satisfied: pexpect in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ipython>=5.4.1->jupyterthemes) (4.7.0)\n",
      "Requirement already satisfied: cycler>=0.10 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib>=1.4.3->jupyterthemes) (0.10.0)\n",
      "Requirement already satisfied: pytz in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib>=1.4.3->jupyterthemes) (2019.3)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib>=1.4.3->jupyterthemes) (1.1.0)\n",
      "Requirement already satisfied: numpy>=1.7.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib>=1.4.3->jupyterthemes) (1.20.3)\n",
      "Requirement already satisfied: six>=1.10 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib>=1.4.3->jupyterthemes) (1.15.0)\n",
      "Requirement already satisfied: python-dateutil>=2.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib>=1.4.3->jupyterthemes) (2.8.0)\n",
      "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib>=1.4.3->jupyterthemes) (2.4.2)\n",
      "Requirement already satisfied: ply in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from lesscpy>=0.11.2->jupyterthemes) (3.11)\n",
      "Requirement already satisfied: jinja2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (2.10.1)\n",
      "Requirement already satisfied: tornado>=4 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (5.1.1)\n",
      "Requirement already satisfied: pyzmq>=17 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (18.1.1)\n",
      "Requirement already satisfied: ipython_genutils in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (0.2.0)\n",
      "Requirement already satisfied: jupyter_client>=5.2.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (5.3.3)\n",
      "Requirement already satisfied: nbformat in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (4.4.0)\n",
      "Requirement already satisfied: nbconvert in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (5.3.1)\n",
      "Requirement already satisfied: ipykernel in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (5.1.0)\n",
      "Requirement already satisfied: Send2Trash in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (1.5.0)\n",
      "Requirement already satisfied: terminado>=0.8.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (0.8.2)\n",
      "Requirement already satisfied: prometheus_client in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from notebook>=5.6.0->jupyterthemes) (0.5.0)\n",
      "Requirement already satisfied: parso>=0.7.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jedi>=0.10->ipython>=5.4.1->jupyterthemes) (0.7.0)\n",
      "Requirement already satisfied: wcwidth in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from prompt_toolkit<2.1.0,>=2.0.0->ipython>=5.4.1->jupyterthemes) (0.1.7)\n",
      "Requirement already satisfied: ptyprocess>=0.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pexpect->ipython>=5.4.1->jupyterthemes) (0.6.0)\n",
      "Requirement already satisfied: MarkupSafe>=0.23 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jinja2->notebook>=5.6.0->jupyterthemes) (1.1.1)\n",
      "Requirement already satisfied: jsonschema!=2.5.0,>=2.4 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from nbformat->notebook>=5.6.0->jupyterthemes) (3.1.1)\n",
      "Requirement already satisfied: mistune>=0.7.4 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from nbconvert->notebook>=5.6.0->jupyterthemes) (0.8.4)\n",
      "Requirement already satisfied: pandocfilters>=1.4.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from nbconvert->notebook>=5.6.0->jupyterthemes) (1.4.2)\n",
      "Requirement already satisfied: bleach in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from nbconvert->notebook>=5.6.0->jupyterthemes) (3.1.0)\n",
      "Requirement already satisfied: testpath in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from nbconvert->notebook>=5.6.0->jupyterthemes) (0.4.2)\n",
      "Requirement already satisfied: entrypoints>=0.2.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from nbconvert->notebook>=5.6.0->jupyterthemes) (0.3)\n",
      "Requirement already satisfied: pyrsistent>=0.14.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jsonschema!=2.5.0,>=2.4->nbformat->notebook>=5.6.0->jupyterthemes) (0.15.4)\n",
      "Requirement already satisfied: attrs>=17.4.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jsonschema!=2.5.0,>=2.4->nbformat->notebook>=5.6.0->jupyterthemes) (19.2.0)\n",
      "Requirement already satisfied: importlib-metadata in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from jsonschema!=2.5.0,>=2.4->nbformat->notebook>=5.6.0->jupyterthemes) (0.23)\n",
      "Requirement already satisfied: webencodings in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from bleach->nbconvert->notebook>=5.6.0->jupyterthemes) (0.5.1)\n",
      "Requirement already satisfied: zipp>=0.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from importlib-metadata->jsonschema!=2.5.0,>=2.4->nbformat->notebook>=5.6.0->jupyterthemes) (0.6.0)\n",
      "Requirement already satisfied: more-itertools in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from zipp>=0.5->importlib-metadata->jsonschema!=2.5.0,>=2.4->nbformat->notebook>=5.6.0->jupyterthemes) (7.2.0)\n"
     ]
    }
   ],
   "source": [
    "!pip install jupyterthemes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "### 查看当前环境中安装的package"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Package                Version        \r\n",
      "---------------------- ---------------\r\n",
      "absl-py                0.8.1          \r\n",
      "aspy.yaml              1.3.0          \r\n",
      "astor                  0.8.1          \r\n",
      "astroid                2.4.1          \r\n",
      "attrs                  19.2.0         \r\n",
      "audioread              2.1.8          \r\n",
      "autopep8               1.5.3          \r\n",
      "Babel                  2.8.0          \r\n",
      "backcall               0.1.0          \r\n",
      "bce-python-sdk         0.8.53         \r\n",
      "blackhole              1.0.1          \r\n",
      "bleach                 3.1.0          \r\n",
      "cachetools             4.0.0          \r\n",
      "certifi                2019.9.11      \r\n",
      "cffi                   1.14.0         \r\n",
      "cfgv                   2.0.1          \r\n",
      "chardet                3.0.4          \r\n",
      "Click                  7.0            \r\n",
      "cloudpickle            1.6.0          \r\n",
      "cma                    2.7.0          \r\n",
      "colorama               0.4.4          \r\n",
      "colorlog               4.1.0          \r\n",
      "cycler                 0.10.0         \r\n",
      "Cython                 0.29           \r\n",
      "decorator              4.4.2          \r\n",
      "dill                   0.3.3          \r\n",
      "easydict               1.9            \r\n",
      "entrypoints            0.3            \r\n",
      "et-xmlfile             1.0.1          \r\n",
      "filelock               3.0.12         \r\n",
      "flake8                 3.8.2          \r\n",
      "Flask                  1.1.1          \r\n",
      "Flask-Babel            1.0.0          \r\n",
      "Flask-Cors             3.0.8          \r\n",
      "forbiddenfruit         0.1.3          \r\n",
      "funcsigs               1.0.2          \r\n",
      "future                 0.18.0         \r\n",
      "gast                   0.3.3          \r\n",
      "gitdb                  4.0.5          \r\n",
      "GitPython              3.1.14         \r\n",
      "google-auth            1.10.0         \r\n",
      "google-auth-oauthlib   0.4.1          \r\n",
      "graphviz               0.13           \r\n",
      "grpcio                 1.35.0         \r\n",
      "gunicorn               20.0.4         \r\n",
      "gym                    0.12.1         \r\n",
      "h5py                   2.9.0          \r\n",
      "identify               1.4.10         \r\n",
      "idna                   2.8            \r\n",
      "imageio                2.6.1          \r\n",
      "imageio-ffmpeg         0.3.0          \r\n",
      "importlib-metadata     0.23           \r\n",
      "ipykernel              5.1.0          \r\n",
      "ipython                7.0.1          \r\n",
      "ipython-genutils       0.2.0          \r\n",
      "isort                  4.3.21         \r\n",
      "itsdangerous           1.1.0          \r\n",
      "jdcal                  1.4.1          \r\n",
      "jedi                   0.17.0         \r\n",
      "jieba                  0.42.1         \r\n",
      "Jinja2                 2.10.1         \r\n",
      "joblib                 0.14.1         \r\n",
      "JPype1                 0.7.2          \r\n",
      "json5                  0.9.5          \r\n",
      "jsonschema             3.1.1          \r\n",
      "jupyter-client         5.3.3          \r\n",
      "jupyter-core           4.5.0          \r\n",
      "jupyter-lsp            0.8.0          \r\n",
      "jupyterlab             2.1.3          \r\n",
      "jupyterlab-server      1.1.5          \r\n",
      "jupyterthemes          0.20.0         \r\n",
      "kiwisolver             1.1.0          \r\n",
      "lazy-object-proxy      1.4.3          \r\n",
      "lesscpy                0.14.0         \r\n",
      "librosa                0.7.2          \r\n",
      "lightgbm               3.1.1          \r\n",
      "llvmlite               0.31.0         \r\n",
      "Markdown               3.1.1          \r\n",
      "MarkupSafe             1.1.1          \r\n",
      "matplotlib             2.2.3          \r\n",
      "mccabe                 0.6.1          \r\n",
      "mistune                0.8.4          \r\n",
      "more-itertools         7.2.0          \r\n",
      "moviepy                1.0.1          \r\n",
      "nbconvert              5.3.1          \r\n",
      "nbformat               4.4.0          \r\n",
      "netifaces              0.10.9         \r\n",
      "networkx               2.4            \r\n",
      "nltk                   3.4.5          \r\n",
      "nodeenv                1.3.4          \r\n",
      "notebook               5.7.0          \r\n",
      "numba                  0.48.0         \r\n",
      "numpy                  1.20.3         \r\n",
      "oauthlib               3.1.0          \r\n",
      "objgraph               3.4.1          \r\n",
      "opencv-python          4.1.1.26       \r\n",
      "openpyxl               3.0.5          \r\n",
      "packaging              20.9           \r\n",
      "paddlehub              2.0.4          \r\n",
      "paddlenlp              2.0.0rc7       \r\n",
      "paddlepaddle-gpu       2.1.0.post101  \r\n",
      "pandas                 1.1.5          \r\n",
      "pandocfilters          1.4.2          \r\n",
      "parl                   1.4.1          \r\n",
      "parso                  0.7.0          \r\n",
      "pathlib                1.0.1          \r\n",
      "pexpect                4.7.0          \r\n",
      "pickleshare            0.7.5          \r\n",
      "Pillow                 7.1.2          \r\n",
      "pip                    19.2.3         \r\n",
      "pluggy                 0.13.1         \r\n",
      "ply                    3.11           \r\n",
      "pre-commit             1.21.0         \r\n",
      "prettytable            0.7.2          \r\n",
      "proglog                0.1.9          \r\n",
      "prometheus-client      0.5.0          \r\n",
      "prompt-toolkit         2.0.10         \r\n",
      "protobuf               3.14.0         \r\n",
      "psutil                 5.7.2          \r\n",
      "ptyprocess             0.6.0          \r\n",
      "py4j                   0.10.9.2       \r\n",
      "pyarrow                2.0.0          \r\n",
      "pyasn1                 0.4.8          \r\n",
      "pyasn1-modules         0.2.7          \r\n",
      "pycodestyle            2.6.0          \r\n",
      "pycparser              2.19           \r\n",
      "pycryptodome           3.9.9          \r\n",
      "pydataset              0.2.0          \r\n",
      "pydocstyle             5.0.2          \r\n",
      "pyflakes               2.2.0          \r\n",
      "pyglet                 1.4.5          \r\n",
      "Pygments               2.4.2          \r\n",
      "pylint                 2.5.2          \r\n",
      "pynvml                 8.0.4          \r\n",
      "pyparsing              2.4.2          \r\n",
      "pypmml                 0.9.11         \r\n",
      "pyrsistent             0.15.4         \r\n",
      "python-dateutil        2.8.0          \r\n",
      "python-jsonrpc-server  0.3.4          \r\n",
      "python-language-server 0.33.0         \r\n",
      "pytz                   2019.3         \r\n",
      "PyYAML                 5.1.2          \r\n",
      "pyzmq                  18.1.1         \r\n",
      "rarfile                3.1            \r\n",
      "recordio               0.1.7          \r\n",
      "requests               2.22.0         \r\n",
      "requests-oauthlib      1.3.0          \r\n",
      "resampy                0.2.2          \r\n",
      "rope                   0.17.0         \r\n",
      "rsa                    4.0            \r\n",
      "scikit-learn           0.24.2         \r\n",
      "scipy                  1.6.3          \r\n",
      "seaborn                0.10.0         \r\n",
      "Send2Trash             1.5.0          \r\n",
      "sentencepiece          0.1.85         \r\n",
      "seqeval                1.2.2          \r\n",
      "setuptools             56.2.0         \r\n",
      "shellcheck-py          0.7.1.1        \r\n",
      "simplegeneric          0.8.1          \r\n",
      "six                    1.15.0         \r\n",
      "sklearn                0.0            \r\n",
      "smmap                  3.0.5          \r\n",
      "snowballstemmer        2.0.0          \r\n",
      "SoundFile              0.10.3.post1   \r\n",
      "tabulate               0.8.3          \r\n",
      "tb-nightly             1.15.0a20190801\r\n",
      "tb-paddle              0.3.6          \r\n",
      "tensorboard            2.1.0          \r\n",
      "tensorboardX           1.8            \r\n",
      "termcolor              1.1.0          \r\n",
      "terminado              0.8.2          \r\n",
      "testpath               0.4.2          \r\n",
      "threadpoolctl          2.1.0          \r\n",
      "toml                   0.10.0         \r\n",
      "tornado                5.1.1          \r\n",
      "tqdm                   4.36.1         \r\n",
      "traitlets              4.3.3          \r\n",
      "typed-ast              1.4.1          \r\n",
      "ujson                  1.35           \r\n",
      "urllib3                1.25.6         \r\n",
      "virtualenv             16.7.9         \r\n",
      "visualdl               2.1.1          \r\n",
      "wcwidth                0.1.7          \r\n",
      "webencodings           0.5.1          \r\n",
      "Werkzeug               0.16.0         \r\n",
      "wheel                  0.36.2         \r\n",
      "wrapt                  1.12.1         \r\n",
      "xarray                 0.16.2         \r\n",
      "xgboost                1.3.3          \r\n",
      "xlrd                   1.2.0          \r\n",
      "yapf                   0.26.0         \r\n",
      "zipp                   0.6.0          \r\n"
     ]
    }
   ],
   "source": [
    "!pip list --format=columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## 持久化安装\n",
    "\n",
    "如果需要进行持久化安装, 需要使用持久化路径, 如下方代码示例:\n",
    "```\n",
    "!mkdir /home/aistudio/external-libraries\n",
    "!pip install beautifulsoup4 -t /home/aistudio/external-libraries\n",
    "```\n",
    "\n",
    "同时添加如下代码, 这样每次环境(kernel)启动的时候只要运行下方代码即可:\n",
    "```\n",
    "import sys\n",
    "sys.path.append('/home/aistudio/external-libraries')\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "### 使用git命令来同步代码 \n",
    "\n",
    "例如: \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Errno 2] No such file or directory: 'work/'\n",
      "/home/aistudio/work\n"
     ]
    }
   ],
   "source": [
    "%cd work/\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fatal: destination path 'Paddle' already exists and is not an empty directory.\n",
      "Cloning into 'PaddleNLP'...\n",
      "fatal: unable to access 'https://github.com/PaddlePaddle/PaddleNLP.git/': gnutls_handshake() failed: Error in the pull function.\n"
     ]
    }
   ],
   "source": [
    "!git clone https://github.com/PaddlePaddle/Paddle.git\n",
    "!git clone https://github.com/PaddlePaddle/PaddleNLP.git"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "可以看到同步后work目录下出现一个Models文件夹\n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/fd9217b5f6e74885b2dca27ac6cfb564f3165a5c181f4205a183ce63d7689eb2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    "也可以活用Git命令进行本地代码管理. \n",
    "\n",
    "> 注意: Paddle的Git体积过大, 同步会非常迟缓. 建议使用浏览器插件, 对指定目录进行下载, 然后上传至AI Studio的项目空间中. [Chrome版](https://chrome.google.com/webstore/detail/gitzip-for-github/ffabmkklhbepgcgfonabamgnfafbdlkn)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "## 文件下载\n",
    "\n",
    "文件下载分为\"文件夹打包下载\"和\"单一文件下载\", 两者操作按钮在同一位置:\n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/c95f71b5d89140dc8ce3eac0951fb2a467aa844c070642f5bf2d6a1311ae140c)\n",
    "\n",
    "但文件夹下载可能耗时比较久, 每次只能下载一个文件夹内容, 也可能需要较长时间等待. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Python代码执行与调试\n",
    "\n",
    "Notebook执行Python代码的原理和传统IDE略有不同. \n",
    "\n",
    "1. 传统IDE, 当点击Run按钮时, 编译器/解释器开始构建一个进程. 用户通过单步执行/设置断点进行代码调试和变量监控. 当代码出错, 或用户点击Stop按钮时, 进程被杀死, 资源回收. \n",
    "\n",
    "而Notebook, 一旦启动, 就是开始创建一个\"进程\"(kernel). 每一个Cell, 都是一个天然的断点. 当代码出错, 或用户点击Stop按钮时, \"进程\"通常也不会被杀死. \n",
    "\n",
    "因此如果代码陷入死循环等情况, 需要用户手动关闭并重启该\"进程\". \n",
    "\n",
    "2. 此外, Notebook的Cell是可以随意颠倒顺序来执行的. 这点和传统IDE有很大不同. \n",
    "\n",
    "3. 作为前端的Notebook, 与后端的进程(kernel), 建立有一个Session. 未来本平台将支持terminal功能. 也就可以同时支持多个Session来控制kernel.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# 变量监控\n",
    "\n",
    "因为Notebook的Cell是可以随意颠倒顺序来执行的, 因此本平台自带\"变量监控\"和\"运行历史\". 方便用户了解当前代码运行状态. \n",
    "\n",
    "通过重启环境并清空输出, 可以消除已生成的变量监控. \n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/151f04fbb75d469f9456b80b820b9f700765dfa5816f40a49a05a29643add7bb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "## 2. **Magic命令**\n",
    "\n",
    "Magic命令是Notebook的高级用法了. 可以运行一些特殊的指令. Magic 命令的前面带有一个或两个百分号（% 或 %%），分别代表行 Magic 命令和单元格 Magic 命令。行 Magic 命令仅应用于编写 Magic 命令时所在的行，而单元格 Magic 命令应用于整个单元格。\n",
    "\n",
    "举个例子:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Available line magics:\n",
       "%alias  %alias_magic  %autoawait  %autocall  %automagic  %autosave  %bookmark  %cat  %cd  %clear  %colors  %config  %connect_info  %cp  %debug  %dhist  %dirs  %doctest_mode  %ed  %edit  %env  %gui  %hist  %history  %killbgscripts  %ldir  %less  %lf  %lk  %ll  %load  %load_ext  %loadpy  %logoff  %logon  %logstart  %logstate  %logstop  %ls  %lsmagic  %lx  %macro  %magic  %man  %matplotlib  %mkdir  %more  %mv  %notebook  %page  %pastebin  %pdb  %pdef  %pdoc  %pfile  %pinfo  %pinfo2  %popd  %pprint  %precision  %prun  %psearch  %psource  %pushd  %pwd  %pycat  %pylab  %qtconsole  %quickref  %recall  %rehashx  %reload_ext  %rep  %rerun  %reset  %reset_selective  %rm  %rmdir  %run  %save  %sc  %set_env  %store  %sx  %system  %tb  %time  %timeit  %unalias  %unload_ext  %who  %who_ls  %whos  %xdel  %xmode\n",
       "\n",
       "Available cell magics:\n",
       "%%!  %%HTML  %%SVG  %%bash  %%capture  %%debug  %%file  %%html  %%javascript  %%js  %%latex  %%markdown  %%perl  %%prun  %%pypy  %%python  %%python2  %%python3  %%ruby  %%script  %%sh  %%svg  %%sx  %%system  %%time  %%timeit  %%writefile\n",
       "\n",
       "Automagic is ON, % prefix IS NOT needed for line magics."
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#显示全部可用的Magic命令\n",
    "\n",
    "%lsmagic\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#使用Magic命令来统计运行时长\n",
    "\n",
    "import random"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99.7 µs ± 36.1 ns per loop (mean ± std. dev. of 7 runs, 10000 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "\n",
    "prize = 0 \n",
    "for i in range(100):\n",
    "    roll = random.randint(1, 6)\n",
    "    if roll%2 == 0:\n",
    "        prize += roll\n",
    "    else:\n",
    "        prize -= 1\n",
    "        \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---   \n",
    "\n",
    "**甚至还可以直接嵌入可视化内容, 例如%matplotlib inline:**\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa1190827d0>]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa119082990>]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "x = np.linspace(0,1,300)\n",
    "for w in range(2,6,2):\n",
    "    plt.plot(x, np.sin(np.pi*x)*np.sin(2*w*np.pi*x))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "这里需要注意的是, 当前技术架构局限, 一个Cell里面只能输出**一张图片**. \n",
    "如果输出多张图片, 可能会有显示异常的问题"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "### %env：设置环境变量\n",
    "\n",
    "使用该命令, 可以在不必重启Kernel的情况下管理notebook的环境变量\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "env: OMP_NUM_THREADS=4\n"
     ]
    }
   ],
   "source": [
    "# Running %env without any arguments\n",
    "# lists all environment variables\n",
    "\n",
    "# The line below sets the environment\n",
    "# variable OMP_NUM_THREADS\n",
    "%env OMP_NUM_THREADS=4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "\n",
    "### %run: 运行python代码\n",
    "\n",
    "使用%run 可以运行.py格式的python代码\n",
    "\n",
    "当然是用!python也是可以的\n",
    "\n",
    "我在项目空间中上传了一个.py文件, 里面只有一行print代码. 我们执行一下看看. \n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/5a0d9d3487374946881f7843b40e544caa6ce50aeef14ad88c067abc46c348fb)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:File `'work/SampleOfRun.py'` not found.\n"
     ]
    }
   ],
   "source": [
    "%run work/SampleOfRun.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "python: can't open file 'work/SampleOfRun.py': [Errno 2] No such file or directory\r\n"
     ]
    }
   ],
   "source": [
    "!python work/SampleOfRun.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "### **%%writefile and %pycat: 导出cell内容/显示外部脚本的内容**\n",
    "\n",
    "AI Studio当前支持一定格式文件的预览和处理, 如果您的格式比较特殊, 尚未支持的话, 不妨试试这两个命令. \n",
    "\n",
    "%%writefile magic可以把cell的内容保存到外部文件里。 而%pycat则可把外部文件展示在Cell中\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overwriting SaveToPythonCode.py\n"
     ]
    }
   ],
   "source": [
    "%%writefile SaveToPythonCode.py\n",
    "\n",
    "from math import sqrt\n",
    "for i in range(2,10):\n",
    "    flag=1\n",
    "    k=int(sqrt(i))\n",
    "    for j in range(2,k+1):\n",
    "        if i%j==0:\n",
    "            flag=0\n",
    "        break\n",
    "        if(flag):\n",
    "            print(i)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "因为没有指定路径, 所以文件被保存到了根目录下. 但至少it works.\n",
    "\n",
    "![](https://ai-studio-static-online.cdn.bcebos.com/c4c44796af06452295e018e7b0edc6b59f988337bc534d1fa888fe471c43b439)\n",
    "\n",
    "我们再来尝试从中读文件内容"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%pycat SaveToPythonCode.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "更多Magic命令可以点击这里查询 \n",
    "[Magic命令](https://ipython.readthedocs.io/en/stable/interactive/magics.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "## 关于快速查看某个对象/方法/接口的用法\n",
    "\n",
    "在要查询的对象前输入?或??并执行即可. 单问号是普通信息, 双问号是详细信息. \n",
    "\n",
    "前提: 该对象方法已经被正确导入(import)\n",
    "\n",
    "示例:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import paddle\n",
    "?paddle.nn.LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import paddlenlp\n",
    "?paddlenlp.data.Pad"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "## 关于变量监控\n",
    "\n",
    "你可以通过修改内核选项ast_note_interactivity，使得Jupyter对独占一行的所有变量或者语句都自动显示，这样你就可以马上看到多个语句的运行结果了。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Looking in indexes: https://mirror.baidu.com/pypi/simple/\n",
      "Requirement already satisfied: pydataset in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (0.2.0)\n",
      "Requirement already satisfied: pandas in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pydataset) (1.1.5)\n",
      "Requirement already satisfied: pytz>=2017.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pandas->pydataset) (2019.3)\n",
      "Requirement already satisfied: python-dateutil>=2.7.3 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pandas->pydataset) (2.8.0)\n",
      "Requirement already satisfied: numpy>=1.15.4 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pandas->pydataset) (1.20.3)\n",
      "Requirement already satisfied: six>=1.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from python-dateutil>=2.7.3->pandas->pydataset) (1.15.0)\n"
     ]
    }
   ],
   "source": [
    "!pip install pydataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>depth</th>\n",
       "      <th>mag</th>\n",
       "      <th>stations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-20.42</td>\n",
       "      <td>181.62</td>\n",
       "      <td>562</td>\n",
       "      <td>4.8</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-20.62</td>\n",
       "      <td>181.03</td>\n",
       "      <td>650</td>\n",
       "      <td>4.2</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-26.00</td>\n",
       "      <td>184.10</td>\n",
       "      <td>42</td>\n",
       "      <td>5.4</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-17.97</td>\n",
       "      <td>181.66</td>\n",
       "      <td>626</td>\n",
       "      <td>4.1</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>-20.42</td>\n",
       "      <td>181.96</td>\n",
       "      <td>649</td>\n",
       "      <td>4.0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     lat    long  depth  mag  stations\n",
       "1 -20.42  181.62    562  4.8        41\n",
       "2 -20.62  181.03    650  4.2        15\n",
       "3 -26.00  184.10     42  5.4        43\n",
       "4 -17.97  181.66    626  4.1        19\n",
       "5 -20.42  181.96    649  4.0        11"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>depth</th>\n",
       "      <th>mag</th>\n",
       "      <th>stations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>-25.93</td>\n",
       "      <td>179.54</td>\n",
       "      <td>470</td>\n",
       "      <td>4.4</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>-12.28</td>\n",
       "      <td>167.06</td>\n",
       "      <td>248</td>\n",
       "      <td>4.7</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>-20.13</td>\n",
       "      <td>184.20</td>\n",
       "      <td>244</td>\n",
       "      <td>4.5</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>-17.40</td>\n",
       "      <td>187.80</td>\n",
       "      <td>40</td>\n",
       "      <td>4.5</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1000</th>\n",
       "      <td>-21.59</td>\n",
       "      <td>170.56</td>\n",
       "      <td>165</td>\n",
       "      <td>6.0</td>\n",
       "      <td>119</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        lat    long  depth  mag  stations\n",
       "996  -25.93  179.54    470  4.4        22\n",
       "997  -12.28  167.06    248  4.7        35\n",
       "998  -20.13  184.20    244  4.5        34\n",
       "999  -17.40  187.80     40  4.5        14\n",
       "1000 -21.59  170.56    165  6.0       119"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = \"all\"\n",
    "\n",
    "from pydataset import data\n",
    "\n",
    "quakes = data('quakes')\n",
    "quakes.head()\n",
    "quakes.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "---\n",
    "\n",
    "\n",
    "## 关于调试代码\n",
    "\n",
    "\n",
    "Notebook自带一个调试器, 叫The Python Debugger (pdb)，我们来看看它是如何工作的. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Automatic pdb calling has been turned ON\n",
      "431\n"
     ]
    }
   ],
   "source": [
    "# 对, 它也是依赖Magic命令启动的.\n",
    "# 理论上pdb是可以通过脚本形式来启动. 但是在Notebook中不行, 会造成阻断\n",
    "\n",
    "%pdb\n",
    "\n",
    "\n",
    "def reverse(x: int) -> int:\n",
    "    inputnumber = x.__str__()\n",
    "    reversedStr = inputnumber[::-1]\n",
    "    strOriLen = len(reversedStr)\n",
    "    result = list()\n",
    "    flag = 1\n",
    "    for i in range(0, strOriLen):\n",
    "        if i == (strOriLen -1) and reversedStr[i] == '-' :\n",
    "            flag = -1\n",
    "        else:\n",
    "            result.append(reversedStr[i])\n",
    "    outputs = ''.join(result)\n",
    "    outputInt = int(outputs)        \n",
    "     \n",
    "    outputInt = outputInt * flag\n",
    "        \n",
    "    if outputInt > pow(2,31)-1  or outputInt < -1 * pow(2,31):\n",
    "        outputInt = 0\n",
    "        \n",
    "    return outputInt\n",
    "\n",
    "print(reverse(134))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "下一节将重点讲一下PDB的用法\n",
    "\n",
    "[AI Studio基本操作(二)](https://aistudio.baidu.com/aistudio/projectdetail/69987)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "PaddlePaddle 2.0.0b0 (Python 3.5)",
   "language": "python",
   "name": "py35-paddle1.2.0"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}