{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 08. 异步通信并行策略"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "UltraOpt的异步通信并行策略代码基本借鉴和使用了`HpBandSter`的代码，您可以参考[HpBandSter文档](https://automl.github.io/HpBandSter/build/html/index.html)来理解这部分教程的内容。\n",
    "\n",
    "---\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "UltraOpt支持以下3种并行策略：\n",
    "\n",
    "|并行策略|策略名|是否支持多保真度|\n",
    "|-------|-----|-------------|\n",
    "|串行| Serial| No |\n",
    "|MapReduce| MapReduce| No |\n",
    "|异步通信|AsyncComm| **Yes** |"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "UltraOpt的并行策略默认为 `AsyncComm`， 即 `异步通信`。如果您设置了 `n_jobs=1` 并且 `multi_fidelity_iter_generator=None` ，UltraOpt自动推断您的并行策略为 `Serial` \n",
    "\n",
    "如果您想关闭自动推断，可以在`fmin`中设置`auto_identify_serial_strategy=False`\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们之前介绍过UltraOpt的设计哲学是以优化器为中心，其中优化器实现了`ask`(样本推荐)和`tell`(结果反馈)两个接口。现在，我们通过一套远程过程调用框架(Pyro4)异步地实现整个过程：\n",
    "\n",
    "![AsyncComm](https://img-blog.csdnimg.cn/20210102124133367.jpg)\n",
    "\n",
    "在这个异步通信框架中，有以下几个重要组件：\n",
    "\n",
    "- **NameServer** 提供按名寻址的命名服务 | [HpBandSter对应文档](https://automl.github.io/HpBandSter/build/html/core/nameserver.html)\n",
    "- **Master** 整合了优化器与多保真度迭代生成器，只负责任务的调度和优化器更新，不负责配置的评价(如AutoML场景的训练与预测)\n",
    "- **Worker** 可以部署在集群的不同计算结点上，受Master的调度， 负责对配置进行评价(如AutoML场景的训练与预测)，并将结果返回给`Dispacher`\n",
    "- **Dispacher** 调度器，负责将`Master`的计算请求发送给`Worker`，和返回`Worker`结果给`Master`\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "下面我们通过一个实例，自定义Worker并绕开`ultraopt.fmin`中启动`Workers`和`Master`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from ultraopt.async_comm.master import Master\n",
    "from ultraopt.async_comm.worker import Worker\n",
    "from ultraopt.async_comm.nameserver import NameServer\n",
    "from ultraopt.utils.net import get_a_free_port\n",
    "import logging\n",
    "logging.basicConfig(level=logging.INFO)  # verbose print"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 启动 NameServer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "ns_host = \"0.0.0.0\"\n",
    "port = get_a_free_port(9090, ns_host)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "NS = NameServer(run_id=\"learn_ultraopt\", host=ns_host, port=port)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('0.0.0.0', 9090)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "NS.start()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 启动 Worker"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "启动用于AutoML场景的`Worker` (仅作为参考，您可以根据自己的需求进行设计)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 方案1. 采用默认`Worker`，用`评价函数`对其初始化"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 引入automl评价函数\n",
    "from ultraopt.tests.automl import evaluator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "worker1 = Worker(\n",
    "    run_id=\"learn_ultraopt\",\n",
    "    nameserver=ns_host,\n",
    "    nameserver_port=port,\n",
    "    worker_id=0\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "用`评价函数`对其初始化"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "worker1.initialize(evaluator) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "采用**线程**在**后台**运行Worker"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:Worker[0]:WORKER(worker_id='ultraopt.run_learn_ultraopt.worker.tqc-PC.13698.0_139720144627456'): start listening for jobs\n"
     ]
    }
   ],
   "source": [
    "worker1.run(\n",
    "    background=True,\n",
    "    concurrent_type=\"thread\" # 也可以采用 process ，但可能会在Python3.7+版本启动失败，故建议用线程\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 方案2. 继承默认`Worker`为专用`Worker`，重写`compute`函数"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们在 `05. Implement a Simple AutoML System` 教程中学习过， AutoML评价器依赖 `训练数据` ，在实际应用中，`训练数据` 往往都很大。上文演示的用**本地函数**简单地启动多个`Worker`的方法就不能在**集群环境**下使用了。\n",
    "\n",
    "在**集群环境**下，我们应该在每个计算结点单独启动`Worker`。这就需要我们继承默认`Worker`为**专用**`Worker`，并重写`compute`函数。\n",
    "\n",
    "我们参考教程 `06. Combine Multi-Fidelity Optimization` 或者 `ultraopt.tests.automl.Evaluator` 关于评价器的实现，开发 `AutoMLWorker` 。我们需要注意3点：\n",
    "\n",
    "- note 1 : 记得调用基类的构造函数\n",
    "- note 2 : 记得重载 `compute` 函数\n",
    "- note 3 : 记得 `compute` 函数的返回值为字典类型，且必须含有 `loss` 键"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "import sklearn.metrics\n",
    "from sklearn.datasets import load_digits\n",
    "from sklearn.model_selection import StratifiedKFold  \n",
    "import numpy as np\n",
    "\n",
    "from ultraopt.hdl import layering_config\n",
    "\n",
    "default_cv = StratifiedKFold(n_splits=3, shuffle=True, random_state=0)\n",
    "\n",
    "class AutoMLWorker(Worker):\n",
    "    def __init__(\n",
    "        self,\n",
    "        X, y,\n",
    "        metric=\"accuracy\",\n",
    "        cv=default_cv,\n",
    "        **kwargs\n",
    "    ):\n",
    "        self.X = X\n",
    "        self.y = y\n",
    "        self.metric = metric\n",
    "        self.cv = cv\n",
    "        # note 1 : 记得调用基类的构造函数\n",
    "        super(AutoMLWorker, self).__init__(**kwargs)\n",
    "    \n",
    "    # note 2 : 记得重载 `compute` 函数\n",
    "    def compute(self, config_id, config, config_info, budget, working_directory):\n",
    "        '''重写基类的 compute 函数'''\n",
    "        layered_dict = layering_config(config)\n",
    "        AS_HP = layered_dict['classifier'].copy()\n",
    "        AS, HP = AS_HP.popitem()\n",
    "        ML_model = eval(AS)(**HP)\n",
    "        sample_ratio  = budget\n",
    "        scores = []\n",
    "        for i, (train_ix, valid_ix) in enumerate(self.cv.split(X, y)):\n",
    "            rng = np.random.RandomState(i)\n",
    "            size = int(train_ix.size * sample_ratio)\n",
    "            train_ix = rng.choice(train_ix, size, replace=False)\n",
    "            X_train = X[train_ix, :]\n",
    "            y_train = y[train_ix]\n",
    "            X_valid = X[valid_ix, :]\n",
    "            y_valid = y[valid_ix]\n",
    "            ML_model.fit(X_train, y_train)\n",
    "            y_pred = ML_model.predict(X_valid)\n",
    "            score = eval(f\"sklearn.metrics.{self.metric}_score\")(y_valid, y_pred)\n",
    "            scores.append(score)\n",
    "        score = np.mean(scores)\n",
    "        # note 3 : 记得 `compute` 函数的返回值为字典类型，且必须含有 `loss` 键\n",
    "        return {\n",
    "            \"loss\": 1 - score  \n",
    "        }"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "模拟在计算结点中加载本地数据并实例化一个worker"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, y = load_digits(return_X_y=True)\n",
    "worker2 = AutoMLWorker(\n",
    "    X, y, \n",
    "    # 与异步通信有关的配置保持不变\n",
    "    run_id=\"learn_ultraopt\",\n",
    "    nameserver=ns_host,\n",
    "    nameserver_port=port,\n",
    "    worker_id=1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:Worker[1]:WORKER(worker_id='ultraopt.run_learn_ultraopt.worker.tqc-PC.13698.1_139720144627456'): start listening for jobs\n"
     ]
    }
   ],
   "source": [
    "worker2.run(background=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 启动 Master"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from ultraopt.optimizer import ETPEOptimizer\n",
    "from ultraopt.multi_fidelity import HyperBandIterGenerator\n",
    "from ultraopt.utils.progress import default_callback\n",
    "from ultraopt.tests.automl import config_space\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 控制tabular_nn 日志等级\n",
    "logging.getLogger(\"tabular_nn.component.embedding_encoder.EmbeddingEncoder\").setLevel(logging.WARNING)\n",
    "logging.getLogger(\"tabular_nn.entity_embedding_nn.TrainEntityEmbeddingNN\").setLevel(logging.WARNING)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer = ETPEOptimizer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "iter_generator = HyperBandIterGenerator(min_budget=1/4, max_budget=1, eta=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer.initialize(config_space=config_space, budgets=iter_generator.get_budgets())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "master = Master(\n",
    "    run_id=\"learn_ultraopt\", \n",
    "    optimizer=optimizer, \n",
    "    iter_generator=iter_generator, \n",
    "    progress_callback=default_callback,\n",
    "    checkpoint_file=None, # 不保存检查点\n",
    "    checkpoint_freq=1,\n",
    "    nameserver=ns_host, \n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  3%|▎         | 4/130 [00:00<00:18,  6.71trial/s, max budget: 0.25, best loss: 0.048]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:incumbent_trajectory:Challenger (0.0328) is better than incumbent (0.0774) when budget is (1/2).\n",
      "INFO:incumbent_trajectory:Changes in incumbent:\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:n_neighbors : 52 -> 5\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:p : '2:int' -> '1:int'\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  4%|▍         | 5/130 [00:00<00:19,  6.33trial/s, max budget: 0.25, best loss: 0.048]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:incumbent_trajectory:Challenger (0.0289) is better than incumbent (0.0328) when budget is (1/2).\n",
      "INFO:incumbent_trajectory:Changes in incumbent:\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:n_neighbors : 5 -> 1\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  7%|▋         | 9/130 [00:01<00:16,  7.24trial/s, max budget: 0.25, best loss: 0.048]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:incumbent_trajectory:Challenger (0.0184) is better than incumbent (0.0189) when budget is (1).\n",
      "INFO:incumbent_trajectory:Changes in incumbent:\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:n_neighbors : 5 -> 1\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 44%|████▍     | 57/130 [00:13<00:21,  3.40trial/s, max budget: 1.0, best loss: 0.018] "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:incumbent_trajectory:Challenger (0.0217) is better than incumbent (0.0289) when budget is (1/2).\n",
      "INFO:incumbent_trajectory:Changes in incumbent:\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:p : '1:int' -> '2:int'\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 45%|████▌     | 59/130 [00:14<00:16,  4.28trial/s, max budget: 1.0, best loss: 0.018]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:incumbent_trajectory:Challenger (0.0139) is better than incumbent (0.0184) when budget is (1).\n",
      "INFO:incumbent_trajectory:Changes in incumbent:\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:p : '1:int' -> '2:int'\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 48%|████▊     | 62/130 [00:14<00:14,  4.57trial/s, max budget: 1.0, best loss: 0.014]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:incumbent_trajectory:Challenger (0.0128) is better than incumbent (0.0139) when budget is (1).\n",
      "INFO:incumbent_trajectory:Changes in incumbent:\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:n_neighbors : 1 -> 3\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 58%|█████▊    | 76/130 [00:16<00:07,  6.76trial/s, max budget: 1.0, best loss: 0.013]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:incumbent_trajectory:Challenger (0.0117) is better than incumbent (0.0128) when budget is (1).\n",
      "INFO:incumbent_trajectory:Changes in incumbent:\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:n_neighbors : 3 -> 4\n",
      "INFO:incumbent_trajectory:Challenger (0.0345) is better than incumbent (0.0479) when budget is (1/4).\n",
      "INFO:incumbent_trajectory:Changes in incumbent:\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:n_neighbors : 1 -> 5\n",
      "INFO:incumbent_trajectory:  classifier:KNeighborsClassifier:p : '1:int' -> '2:int'\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100%|██████████| 130/130 [00:21<00:00,  5.95trial/s, max budget: 1.0, best loss: 0.012]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<ultraopt.result.Result at 0x7f123c6fff60>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "master.run(n_iterations=30)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们可以对**优化过程**和**优化结果**进行可视化分析："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from ultraopt import FMinResult"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "+--------------------------------------------------------------------------------------------------------------------------+\n",
       "| HyperParameters                                     | Optimal Value                                                      |\n",
       "+-----------------------------------------------------+----------------------+----------------------+----------------------+\n",
       "| classifier:__choice__                               | KNeighborsClassifier | KNeighborsClassifier | KNeighborsClassifier |\n",
       "| classifier:KNeighborsClassifier:n_neighbors         | 5                    | 1                    | 4                    |\n",
       "| classifier:KNeighborsClassifier:p                   | 2:int                | 2:int                | 2:int                |\n",
       "| classifier:KNeighborsClassifier:weights             | distance             | distance             | distance             |\n",
       "| classifier:LinearSVC:C                              | -                    | -                    | -                    |\n",
       "| classifier:LinearSVC:dual                           | -                    | -                    | -                    |\n",
       "| classifier:LinearSVC:loss                           | -                    | -                    | -                    |\n",
       "| classifier:LinearSVC:max_iter                       | -                    | -                    | -                    |\n",
       "| classifier:LinearSVC:multi_class                    | -                    | -                    | -                    |\n",
       "| classifier:LinearSVC:penalty                        | -                    | -                    | -                    |\n",
       "| classifier:LinearSVC:random_state                   | -                    | -                    | -                    |\n",
       "| classifier:RandomForestClassifier:bootstrap         | -                    | -                    | -                    |\n",
       "| classifier:RandomForestClassifier:criterion         | -                    | -                    | -                    |\n",
       "| classifier:RandomForestClassifier:max_features      | -                    | -                    | -                    |\n",
       "| classifier:RandomForestClassifier:min_samples_leaf  | -                    | -                    | -                    |\n",
       "| classifier:RandomForestClassifier:min_samples_split | -                    | -                    | -                    |\n",
       "| classifier:RandomForestClassifier:n_estimators      | -                    | -                    | -                    |\n",
       "| classifier:RandomForestClassifier:random_state      | -                    | -                    | -                    |\n",
       "+-----------------------------------------------------+----------------------+----------------------+----------------------+\n",
       "| Budgets                                             | 1/4                  | 1/2                  | 1 (max)              |\n",
       "+-----------------------------------------------------+----------------------+----------------------+----------------------+\n",
       "| Optimal Loss                                        | 0.0345               | 0.0217               | 0.0117               |\n",
       "+-----------------------------------------------------+----------------------+----------------------+----------------------+\n",
       "| Num Configs                                         | 40                   | 40                   | 50                   |\n",
       "+-----------------------------------------------------+----------------------+----------------------+----------------------+"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = FMinResult(optimizer)\n",
    "result"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我们之前在后台启动了`2`个`Worker`， 所以**并行图**显示时序中的最大并行数为`2`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "result.plot_concurrent_over_time(num_points=200);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "绘制之前学习过的其他图片："
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pylab as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x864 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = (16, 12)\n",
    "plt.subplot(2, 2, 1)\n",
    "result.plot_convergence_over_time();\n",
    "plt.subplot(2, 2, 2)\n",
    "ax = result.plot_convergence_over_iter(budget=1, name=\"budget = 1\")\n",
    "plt.subplot(2, 2, 3)\n",
    "result.plot_finished_over_time();\n",
    "plt.subplot(2, 2, 4)\n",
    "result.plot_correlation_across_budgets();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 释放资源"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "完成计算后，我们需要将后台的`Worker`等线程关闭，避免对端口等资源的长时间占用。\n",
    "\n",
    "我们首先关闭 `master` , 关闭master时会自动关闭所有被连接的 `worker`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:ultraopt.async_comm.master.Master:HBMASTER: shutdown initiated, shutdown_workers = True\n"
     ]
    }
   ],
   "source": [
    "master.shutdown(shutdown_workers=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "然后关闭 `NameServer`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "NS.shutdown()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "auto-sklearn",
   "language": "python",
   "name": "auto-sklearn"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}