diabetes_regression_compute_hyperdrive.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install --upgrade azureml-sdk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Azure ML SDK Version:  1.5.0\n"
     ]
    }
   ],
   "source": [
    "import azureml.core\n",
    "from azureml.core import Workspace\n",
    "\n",
    "# check core SDK version number\n",
    "print(\"Azure ML SDK Version: \", azureml.core.VERSION)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initialize workspace"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Workspace.create(name='eacbmlservicews', subscription_id='f80606e5-788f-4dc3-a9ea-2eb9a7836082', resource_group='adlsgen2')"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load workspace configuration from the config.json file in the current folder.\n",
    "ws = Workspace.from_config()\n",
    "\n",
    "ws\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create an experiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table style=\"width:100%\"><tr><th>Name</th><th>Workspace</th><th>Report Page</th><th>Docs Page</th></tr><tr><td>diabetes_dxd</td><td>eacbmlservicews</td><td><a href=\"https://ml.azure.com/experiments/diabetes_dxd?wsid=/subscriptions/f80606e5-788f-4dc3-a9ea-2eb9a7836082/resourcegroups/adlsgen2/workspaces/eacbmlservicews\" target=\"_blank\" rel=\"noopener\">Link to Azure Machine Learning studio</a></td><td><a href=\"https://docs.microsoft.com/en-us/python/api/azureml-core/azureml.core.experiment.Experiment?view=azure-ml-py\" target=\"_blank\" rel=\"noopener\">Link to Documentation</a></td></tr></table>"
      ],
      "text/plain": [
       "Experiment(Name: diabetes_dxd,\n",
       "Workspace: eacbmlservicews)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "experiment_name = 'diabetes_dxd'\n",
    "\n",
    "from azureml.core import Experiment\n",
    "exp = Experiment(workspace=ws, name=experiment_name)\n",
    "\n",
    "exp\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Upload dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from azureml.core import Dataset\n",
    "\n",
    "diabetes_dataset = Dataset.get_by_name(ws, name='diabetes')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "diabetes = diabetes_dataset.to_pandas_dataframe().drop(\"Path\", axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "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>AGE</th>\n",
       "      <th>SEX</th>\n",
       "      <th>BMI</th>\n",
       "      <th>BP</th>\n",
       "      <th>S1</th>\n",
       "      <th>S2</th>\n",
       "      <th>S3</th>\n",
       "      <th>S4</th>\n",
       "      <th>S5</th>\n",
       "      <th>S6</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>59</td>\n",
       "      <td>2</td>\n",
       "      <td>32.1</td>\n",
       "      <td>101.0</td>\n",
       "      <td>157</td>\n",
       "      <td>93.2</td>\n",
       "      <td>38.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.8598</td>\n",
       "      <td>87</td>\n",
       "      <td>151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>48</td>\n",
       "      <td>1</td>\n",
       "      <td>21.6</td>\n",
       "      <td>87.0</td>\n",
       "      <td>183</td>\n",
       "      <td>103.2</td>\n",
       "      <td>70.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.8918</td>\n",
       "      <td>69</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>72</td>\n",
       "      <td>2</td>\n",
       "      <td>30.5</td>\n",
       "      <td>93.0</td>\n",
       "      <td>156</td>\n",
       "      <td>93.6</td>\n",
       "      <td>41.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.6728</td>\n",
       "      <td>85</td>\n",
       "      <td>141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>24</td>\n",
       "      <td>1</td>\n",
       "      <td>25.3</td>\n",
       "      <td>84.0</td>\n",
       "      <td>198</td>\n",
       "      <td>131.4</td>\n",
       "      <td>40.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.8903</td>\n",
       "      <td>89</td>\n",
       "      <td>206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>23.0</td>\n",
       "      <td>101.0</td>\n",
       "      <td>192</td>\n",
       "      <td>125.4</td>\n",
       "      <td>52.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.2905</td>\n",
       "      <td>80</td>\n",
       "      <td>135</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   AGE  SEX   BMI     BP   S1     S2    S3   S4      S5  S6    Y\n",
       "0   59    2  32.1  101.0  157   93.2  38.0  4.0  4.8598  87  151\n",
       "1   48    1  21.6   87.0  183  103.2  70.0  3.0  3.8918  69   75\n",
       "2   72    2  30.5   93.0  156   93.6  41.0  4.0  4.6728  85  141\n",
       "3   24    1  25.3   84.0  198  131.4  40.0  5.0  4.8903  89  206\n",
       "4   50    1  23.0  101.0  192  125.4  52.0  4.0  4.2905  80  135"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diabetes.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 442 entries, 0 to 441\n",
      "Data columns (total 11 columns):\n",
      "AGE    442 non-null int64\n",
      "SEX    442 non-null int64\n",
      "BMI    442 non-null float64\n",
      "BP     442 non-null float64\n",
      "S1     442 non-null int64\n",
      "S2     442 non-null float64\n",
      "S3     442 non-null float64\n",
      "S4     442 non-null float64\n",
      "S5     442 non-null float64\n",
      "S6     442 non-null int64\n",
      "Y      442 non-null int64\n",
      "dtypes: float64(6), int64(5)\n",
      "memory usage: 38.1 KB\n"
     ]
    }
   ],
   "source": [
    "diabetes.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "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>AGE</th>\n",
       "      <th>SEX</th>\n",
       "      <th>BMI</th>\n",
       "      <th>BP</th>\n",
       "      <th>S1</th>\n",
       "      <th>S2</th>\n",
       "      <th>S3</th>\n",
       "      <th>S4</th>\n",
       "      <th>S5</th>\n",
       "      <th>S6</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "      <td>442.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>48.518100</td>\n",
       "      <td>1.468326</td>\n",
       "      <td>26.375792</td>\n",
       "      <td>94.647014</td>\n",
       "      <td>189.140271</td>\n",
       "      <td>115.439140</td>\n",
       "      <td>49.788462</td>\n",
       "      <td>4.070249</td>\n",
       "      <td>4.641411</td>\n",
       "      <td>91.260181</td>\n",
       "      <td>152.133484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>13.109028</td>\n",
       "      <td>0.499561</td>\n",
       "      <td>4.418122</td>\n",
       "      <td>13.831283</td>\n",
       "      <td>34.608052</td>\n",
       "      <td>30.413081</td>\n",
       "      <td>12.934202</td>\n",
       "      <td>1.290450</td>\n",
       "      <td>0.522391</td>\n",
       "      <td>11.496335</td>\n",
       "      <td>77.093005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>19.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>62.000000</td>\n",
       "      <td>97.000000</td>\n",
       "      <td>41.600000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.258100</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>25.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>38.250000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>23.200000</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>164.250000</td>\n",
       "      <td>96.050000</td>\n",
       "      <td>40.250000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.276700</td>\n",
       "      <td>83.250000</td>\n",
       "      <td>87.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>25.700000</td>\n",
       "      <td>93.000000</td>\n",
       "      <td>186.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.620050</td>\n",
       "      <td>91.000000</td>\n",
       "      <td>140.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>59.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>29.275000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>209.750000</td>\n",
       "      <td>134.500000</td>\n",
       "      <td>57.750000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.997200</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>211.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>79.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>42.200000</td>\n",
       "      <td>133.000000</td>\n",
       "      <td>301.000000</td>\n",
       "      <td>242.400000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>9.090000</td>\n",
       "      <td>6.107000</td>\n",
       "      <td>124.000000</td>\n",
       "      <td>346.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              AGE         SEX         BMI          BP          S1          S2  \\\n",
       "count  442.000000  442.000000  442.000000  442.000000  442.000000  442.000000   \n",
       "mean    48.518100    1.468326   26.375792   94.647014  189.140271  115.439140   \n",
       "std     13.109028    0.499561    4.418122   13.831283   34.608052   30.413081   \n",
       "min     19.000000    1.000000   18.000000   62.000000   97.000000   41.600000   \n",
       "25%     38.250000    1.000000   23.200000   84.000000  164.250000   96.050000   \n",
       "50%     50.000000    1.000000   25.700000   93.000000  186.000000  113.000000   \n",
       "75%     59.000000    2.000000   29.275000  105.000000  209.750000  134.500000   \n",
       "max     79.000000    2.000000   42.200000  133.000000  301.000000  242.400000   \n",
       "\n",
       "               S3          S4          S5          S6           Y  \n",
       "count  442.000000  442.000000  442.000000  442.000000  442.000000  \n",
       "mean    49.788462    4.070249    4.641411   91.260181  152.133484  \n",
       "std     12.934202    1.290450    0.522391   11.496335   77.093005  \n",
       "min     22.000000    2.000000    3.258100   58.000000   25.000000  \n",
       "25%     40.250000    3.000000    4.276700   83.250000   87.000000  \n",
       "50%     48.000000    4.000000    4.620050   91.000000  140.500000  \n",
       "75%     57.750000    5.000000    4.997200   98.000000  211.500000  \n",
       "max     99.000000    9.090000    6.107000  124.000000  346.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diabetes.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "features_names = ['age', 'sex', 'bmi', 'bp', 's1', 's2', 's3', 's4', 's5', 's6']\n",
    "target = \"Y\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create a (remote) target compute"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Cluster already exists\n",
      "Succeeded\n",
      "AmlCompute wait for completion finished\n",
      "\n",
      "Minimum number of nodes requested have been provisioned\n"
     ]
    }
   ],
   "source": [
    "# Compute target creation\n",
    "\n",
    "from azureml.core.compute import ComputeTarget, AmlCompute\n",
    "from azureml.core.compute_target import ComputeTargetException\n",
    "\n",
    "cpu_cluster_name = \"myComputeCluster\"\n",
    "\n",
    "# Verify that cluster does not exist already\n",
    "try:\n",
    "    cpu_cluster = ComputeTarget(workspace=ws, name=cpu_cluster_name)\n",
    "    print(\" Cluster already exists\")\n",
    "except ComputeTargetException:\n",
    "    compute_config = AmlCompute.provisioning_configuration(vm_size='STANDARD_D2_V2',\n",
    "                                                           min_nodes=0, max_nodes=4)\n",
    "    cpu_cluster = ComputeTarget.create(ws, cpu_cluster_name, compute_config)\n",
    "\n",
    "cpu_cluster.wait_for_completion(show_output=True, min_node_count=0, timeout_in_minutes=30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Succeeded\n"
     ]
    }
   ],
   "source": [
    "# Retrieve existing compute target\n",
    "\n",
    "from azureml.core.compute import ComputeTarget\n",
    "\n",
    "\n",
    "compute_target_name = \"ds3-compute-cls\"\n",
    "cpu_cluster_name = ComputeTarget(workspace=ws, name=compute_target_name)\n",
    "\n",
    "print(cpu_cluster_name.provisioning_state)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train the model on a remote target compute"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "script_folder = os.path.join(os.getcwd(), \"remote_train\")\n",
    "os.makedirs(script_folder, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Writing /mnt/batch/tasks/shared/LS_root/mounts/clusters/computeinstance/code/Users/diabetes/remote_train/train.py\n"
     ]
    }
   ],
   "source": [
    "%%writefile $script_folder/train.py\n",
    "\n",
    "import argparse\n",
    "import os\n",
    "import numpy as np\n",
    "import glob\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import Ridge\n",
    "import joblib\n",
    "import pickle\n",
    "\n",
    "from azureml.core import Run\n",
    "from azureml.core import Dataset\n",
    "from utils import load_data\n",
    "\n",
    "# get hold of the current run\n",
    "run = Run.get_context()\n",
    "exp = run.experiment\n",
    "ws = run.experiment.workspace\n",
    "\n",
    "parser = argparse.ArgumentParser()\n",
    "parser.add_argument('--regularization', type=float, dest='reg', default=0.5, help='regularization strength')\n",
    "args = parser.parse_args()\n",
    "\n",
    "# load train and test set into numpy arrays\n",
    "diabetes_dataset = Dataset.get_by_name(ws, name='diabetes')\n",
    "diabetes = diabetes_dataset.to_pandas_dataframe().drop(\"Path\", axis=1)\n",
    "target = \"Y\"\n",
    "X = diabetes.drop(target, axis=1)\n",
    "y = diabetes[\"Y\"].values.reshape(-1,1)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "print('Train a Ridge regression model with regularization strength of', args.reg)\n",
    "model = Ridge(alpha=args.reg, solver=\"auto\", random_state=42)\n",
    "model.fit(X_train, y_train)\n",
    "\n",
    "print('Predict the test set')\n",
    "y_hat = model.predict(X_test)\n",
    "\n",
    "# calculate score on the prediction\n",
    "score = model.score(X_test, y_test)\n",
    "print('Score is ', score)\n",
    "\n",
    "run.log('regularization strength', np.float(args.reg))\n",
    "run.log('score', np.float(score))\n",
    "\n",
    "os.makedirs('outputs', exist_ok=True)\n",
    "# note file saved in the outputs folder is automatically uploaded into experiment record\n",
    "joblib.dump(value=model, filename='outputs/diabetes_HD_remote_model.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/mnt/batch/tasks/shared/LS_root/mounts/clusters/computeinstance/code/Users/diabetes/remote_train/utils.py'"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import shutil\n",
    "shutil.copy('utils.py', script_folder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'diabetes_env.yml'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Set up the (compute target) environnement\n",
    "\n",
    "from azureml.core import Environment\n",
    "from azureml.core.conda_dependencies import CondaDependencies\n",
    "\n",
    "\n",
    "env = Environment(\"diabetes_remote_env\")\n",
    "\n",
    "env.docker.enabled = True\n",
    "env.python.conda_dependencies = CondaDependencies.create(conda_packages=['scikit-learn',\n",
    "                                                                          'pandas',\n",
    "                                                                          'numpy',\n",
    "                                                                          'joblib'\n",
    "                                                                         ])\n",
    "env.python.conda_dependencies.add_pip_package(\"inference-schema[numpy-support]\")\n",
    "\n",
    "env.python.conda_dependencies.save_to_file(\".\", \"diabetes_env.yml\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from azureml.train.estimator import Estimator\n",
    "\n",
    "#script_params = {\n",
    "#    '--regularization': 0.5\n",
    "#}\n",
    "\n",
    "estimator = Estimator(source_directory=script_folder,\n",
    "              #script_params=script_params,\n",
    "              compute_target=cpu_cluster_name,\n",
    "              environment_definition=env,\n",
    "              entry_script='train.py')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from azureml.train.hyperdrive import *\n",
    "\n",
    "# define hyperparameter sampling space\n",
    "ps = RandomParameterSampling(\n",
    "     {\n",
    "         '--regularization': choice(0.01, 0.1, 0.5, 1.0),\n",
    "     }\n",
    ")\n",
    "\n",
    "# define early termination policy\n",
    "early_termination_policy = None #BanditPolicy(slack_factor = 0.15, evaluation_interval=10)\n",
    "\n",
    "# configure the run\n",
    "hyperdrive_run_config = HyperDriveConfig(estimator = estimator,\n",
    "                                         hyperparameter_sampling = ps,\n",
    "                                         policy = early_termination_policy,\n",
    "                                         primary_metric_name = \"score\",\n",
    "                                         primary_metric_goal = PrimaryMetricGoal.MAXIMIZE,\n",
    "                                         max_total_runs = 20,\n",
    "                                         max_concurrent_runs = 2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# start the run\n",
    "hd_run = exp.submit(hyperdrive_run_config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bbe56bb614214b9180f3526a46100af9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "_HyperDriveWidget(widget_settings={'childWidgetDisplay': 'popup', 'send_telemetry': False, 'log_level': 'INFO'…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/aml.mini.widget.v1": "{\"status\": \"Completed\", \"workbench_run_details_uri\": \"https://ml.azure.com/experiments/diabetes_dxd/runs/HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd?wsid=/subscriptions/f80606e5-788f-4dc3-a9ea-2eb9a7836082/resourcegroups/adlsgen2/workspaces/eacbmlservicews\", \"run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\", \"run_properties\": {\"run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\", \"created_utc\": \"2020-06-01T06:48:36.765215Z\", \"properties\": {\"primary_metric_config\": \"{\\\"name\\\": \\\"score\\\", \\\"goal\\\": \\\"maximize\\\"}\", \"resume_from\": \"null\", \"runTemplate\": \"HyperDrive\", \"azureml.runsource\": \"hyperdrive\", \"platform\": \"AML\", \"ContentSnapshotId\": \"5db72d9f-93a1-4ed1-8bc0-c7aa7ed90307\", \"score\": \"0.45259921776197887\", \"best_child_run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_3\", \"best_metric_status\": \"Succeeded\"}, \"tags\": {\"max_concurrent_jobs\": \"2\", \"max_total_jobs\": \"20\", \"max_duration_minutes\": \"10080\", \"policy_config\": \"{\\\"name\\\": \\\"DEFAULT\\\"}\", \"generator_config\": \"{\\\"name\\\": \\\"RANDOM\\\", \\\"parameter_space\\\": {\\\"--regularization\\\": [\\\"choice\\\", [[0.01, 0.1, 0.5, 1.0]]]}}\", \"primary_metric_config\": \"{\\\"name\\\": \\\"score\\\", \\\"goal\\\": \\\"maximize\\\"}\", \"platform_config\": \"{\\\"ServiceAddress\\\": \\\"https://westus2.experiments.azureml.net\\\", \\\"ServiceArmScope\\\": \\\"subscriptions/f80606e5-788f-4dc3-a9ea-2eb9a7836082/resourceGroups/adlsgen2/providers/Microsoft.MachineLearningServices/workspaces/eacbmlservicews/experiments/diabetes_dxd\\\", \\\"SubscriptionId\\\": \\\"f80606e5-788f-4dc3-a9ea-2eb9a7836082\\\", \\\"ResourceGroupName\\\": \\\"adlsgen2\\\", \\\"WorkspaceName\\\": \\\"eacbmlservicews\\\", \\\"ExperimentName\\\": \\\"diabetes_dxd\\\", \\\"Definition\\\": {\\\"Overrides\\\": {\\\"script\\\": \\\"train.py\\\", \\\"arguments\\\": [], \\\"target\\\": \\\"myComputeCluster\\\", \\\"framework\\\": \\\"Python\\\", \\\"communicator\\\": \\\"None\\\", \\\"maxRunDurationSeconds\\\": null, \\\"nodeCount\\\": 1, \\\"environment\\\": {\\\"name\\\": \\\"diabetes_remote_env\\\", \\\"version\\\": null, \\\"environmentVariables\\\": {\\\"EXAMPLE_ENV_VAR\\\": \\\"EXAMPLE_VALUE\\\"}, \\\"python\\\": {\\\"userManagedDependencies\\\": false, \\\"interpreterPath\\\": \\\"python\\\", \\\"condaDependenciesFile\\\": null, \\\"baseCondaEnvironment\\\": null, \\\"condaDependencies\\\": {\\\"name\\\": \\\"project_environment\\\", \\\"dependencies\\\": [\\\"python=3.6.2\\\", {\\\"pip\\\": [\\\"azureml-defaults\\\", \\\"inference-schema[numpy-support]\\\"]}, \\\"scikit-learn\\\", \\\"pandas\\\", \\\"numpy\\\", \\\"joblib\\\"], \\\"channels\\\": [\\\"anaconda\\\", \\\"conda-forge\\\"]}}, \\\"docker\\\": {\\\"enabled\\\": true, \\\"baseImage\\\": \\\"mcr.microsoft.com/azureml/base:intelmpi2018.3-ubuntu16.04\\\", \\\"baseDockerfile\\\": null, \\\"sharedVolumes\\\": true, \\\"shmSize\\\": \\\"2g\\\", \\\"arguments\\\": [], \\\"baseImageRegistry\\\": {\\\"address\\\": null, \\\"username\\\": null, \\\"password\\\": null}}, \\\"spark\\\": {\\\"repositories\\\": [], \\\"packages\\\": [], \\\"precachePackages\\\": true}, \\\"databricks\\\": {\\\"mavenLibraries\\\": [], \\\"pypiLibraries\\\": [], \\\"rcranLibraries\\\": [], \\\"jarLibraries\\\": [], \\\"eggLibraries\\\": []}, \\\"r\\\": null, \\\"inferencingStackVersion\\\": null}, \\\"history\\\": {\\\"outputCollection\\\": true, \\\"snapshotProject\\\": true, \\\"directoriesToWatch\\\": [\\\"logs\\\"]}, \\\"spark\\\": {\\\"configuration\\\": {\\\"spark.app.name\\\": \\\"Azure ML Experiment\\\", \\\"spark.yarn.maxAppAttempts\\\": 1}}, \\\"hdi\\\": {\\\"yarnDeployMode\\\": \\\"cluster\\\"}, \\\"tensorflow\\\": {\\\"workerCount\\\": 1, \\\"parameterServerCount\\\": 1}, \\\"mpi\\\": {\\\"processCountPerNode\\\": 1}, \\\"paralleltask\\\": {\\\"maxRetriesPerWorker\\\": 0, \\\"workerCountPerNode\\\": 1, \\\"terminalExitCodes\\\": null}, \\\"dataReferences\\\": {}, \\\"data\\\": {}, \\\"sourceDirectoryDataStore\\\": null, \\\"amlcompute\\\": {\\\"vmSize\\\": null, \\\"vmPriority\\\": null, \\\"retainCluster\\\": false, \\\"name\\\": null, \\\"clusterMaxNodeCount\\\": 1}}, \\\"TargetDetails\\\": null, \\\"SnapshotId\\\": \\\"5db72d9f-93a1-4ed1-8bc0-c7aa7ed90307\\\", \\\"TelemetryValues\\\": {\\\"amlClientType\\\": \\\"azureml-sdk-train\\\", \\\"amlClientModule\\\": \\\"[Scrubbed]\\\", \\\"amlClientFunction\\\": \\\"[Scrubbed]\\\", \\\"tenantId\\\": \\\"8e2e7c2d-4702-496d-af6c-96e4bfc9f667\\\", \\\"amlClientRequestId\\\": \\\"adf3527c-c167-4a42-9c5f-dc1619474f3c\\\", \\\"amlClientSessionId\\\": \\\"d4cf03bc-ddd3-4080-8b5c-a56162673500\\\", \\\"subscriptionId\\\": \\\"f80606e5-788f-4dc3-a9ea-2eb9a7836082\\\", \\\"estimator\\\": \\\"Estimator\\\", \\\"samplingMethod\\\": \\\"RANDOM\\\", \\\"terminationPolicy\\\": \\\"Default\\\", \\\"primaryMetricGoal\\\": \\\"maximize\\\", \\\"maxTotalRuns\\\": 20, \\\"maxConcurrentRuns\\\": 2, \\\"maxDurationMinutes\\\": 10080, \\\"vmSize\\\": null}}}\", \"resume_child_runs\": \"null\", \"all_jobs_generated\": \"true\", \"cancellation_requested\": \"false\", \"progress_metadata_evaluation_timestamp\": \"\\\"2020-06-01T06:48:37.872916\\\"\", \"progress_metadata_digest\": \"\\\"26753f794ea10819b7e087c2c9caa6a96228cf8a395966cc21f6f28809477681\\\"\", \"progress_metadata_active_timestamp\": \"\\\"2020-06-01T06:48:37.872916\\\"\", \"environment_preparation_status\": \"PREPARED\", \"prepare_run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_preparation\", \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_0\": \"{\\\"--regularization\\\": 0.5}\", \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_1\": \"{\\\"--regularization\\\": 0.1}\", \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_2\": \"{\\\"--regularization\\\": 1.0}\", \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_3\": \"{\\\"--regularization\\\": 0.01}\"}, \"end_time_utc\": \"2020-06-01T07:01:16.674336Z\", \"status\": \"Completed\", \"log_files\": {\"azureml-logs/hyperdrive.txt\": \"https://eacbmlservicew2479569759.blob.core.windows.net/azureml/ExperimentRun/dcid.HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd/azureml-logs/hyperdrive.txt?sv=2019-02-02&sr=b&sig=ABhSIQ6h5z5k32w6cwcqD0KFLoM34JujtGVb6i1%2Fdlo%3D&st=2020-06-01T06%3A55%3A39Z&se=2020-06-01T15%3A05%3A39Z&sp=r\"}, \"log_groups\": [[\"azureml-logs/hyperdrive.txt\"]], \"run_duration\": \"0:12:39\", \"hyper_parameters\": {\"--regularization\": [\"choice\", [[0.01, 0.1, 0.5, 1.0]]]}}, \"child_runs\": [{\"run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_0\", \"run_number\": 3, \"metric\": 0.45238303, \"status\": \"Completed\", \"run_type\": \"azureml.scriptrun\", \"training_percent\": null, \"start_time\": \"2020-06-01T06:57:41.569109Z\", \"end_time\": \"2020-06-01T06:58:26.039296Z\", \"created_time\": \"2020-06-01T06:55:44.910448Z\", \"created_time_dt\": \"2020-06-01T06:55:44.910448Z\", \"duration\": \"0:02:41\", \"hyperdrive_id\": \"c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\", \"arguments\": null, \"param_--regularization\": 0.5, \"best_metric\": 0.45238303}, {\"run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_1\", \"run_number\": 4, \"metric\": 0.4525656, \"status\": \"Completed\", \"run_type\": \"azureml.scriptrun\", \"training_percent\": null, \"start_time\": \"2020-06-01T06:56:06.623964Z\", \"end_time\": \"2020-06-01T06:57:39.889725Z\", \"created_time\": \"2020-06-01T06:55:45.889342Z\", \"created_time_dt\": \"2020-06-01T06:55:45.889342Z\", \"duration\": \"0:01:54\", \"hyperdrive_id\": \"c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\", \"arguments\": null, \"param_--regularization\": 0.1, \"best_metric\": 0.4525656}, {\"run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_2\", \"run_number\": 5, \"metric\": 0.45209455, \"status\": \"Completed\", \"run_type\": \"azureml.scriptrun\", \"training_percent\": null, \"start_time\": \"2020-06-01T06:58:30.621764Z\", \"end_time\": \"2020-06-01T06:59:13.587376Z\", \"created_time\": \"2020-06-01T06:57:50.659513Z\", \"created_time_dt\": \"2020-06-01T06:57:50.659513Z\", \"duration\": \"0:01:22\", \"hyperdrive_id\": \"c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\", \"arguments\": null, \"param_--regularization\": 1.0, \"best_metric\": 0.4525656}, {\"run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_3\", \"run_number\": 6, \"metric\": 0.45259922, \"status\": \"Completed\", \"run_type\": \"azureml.scriptrun\", \"training_percent\": null, \"start_time\": \"2020-06-01T06:59:15.223795Z\", \"end_time\": \"2020-06-01T07:00:57.057709Z\", \"created_time\": \"2020-06-01T06:58:54.626154Z\", \"created_time_dt\": \"2020-06-01T06:58:54.626154Z\", \"duration\": \"0:02:02\", \"hyperdrive_id\": \"c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\", \"arguments\": null, \"param_--regularization\": 0.01, \"best_metric\": 0.45259922}], \"children_metrics\": {\"categories\": [0], \"series\": {\"regularization strength\": [{\"categories\": [3, 4, 5, 6], \"mode\": \"markers\", \"name\": \"regularization strength\", \"stepped\": false, \"type\": \"scatter\", \"data\": [0.5, 0.1, 1.0, 0.01]}, {\"categories\": [3, 4, 5, 6], \"mode\": \"lines\", \"name\": \"regularization strength_max\", \"stepped\": true, \"type\": \"scatter\", \"data\": [0.5, 0.5, 1.0, 1.0]}], \"score\": [{\"categories\": [3, 4, 5, 6], \"mode\": \"markers\", \"name\": \"score\", \"stepped\": false, \"type\": \"scatter\", \"data\": [0.45238302601412694, 0.4525656003848694, 0.45209454516860437, 0.45259921776197887]}, {\"categories\": [3, 4, 5, 6], \"mode\": \"lines\", \"name\": \"score_max\", \"stepped\": true, \"type\": \"scatter\", \"data\": [0.45238302601412694, 0.4525656003848694, 0.4525656003848694, 0.45259921776197887]}]}, \"metricName\": null, \"primaryMetricName\": \"score\", \"showLegend\": false}, \"run_metrics\": [{\"name\": \"best_child_by_primary_metric\", \"run_id\": \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\", \"categories\": [0], \"series\": [{\"data\": [{\"metric_name\": [\"score\", \"score\"], \"timestamp\": [\"2020-06-01 06:58:09.021173+00:00\", \"2020-06-01 07:01:15.811955+00:00\"], \"run_id\": [\"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_1\", \"HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_3\"], \"metric_value\": [0.4525656003848694, 0.45259921776197887], \"final\": [false, true]}]}]}], \"run_logs\": \"[2020-06-01T06:48:37.143894][API][INFO]Experiment created\\r\\n[2020-06-01T06:48:38.4940411Z][SCHEDULER][INFO]The execution environment is being prepared. Please be patient as it can take a few minutes.\\r\\n[2020-06-01T06:48:38.511309][GENERATOR][INFO]Trying to sample '2' jobs from the hyperparameter space\\r\\n[2020-06-01T06:48:41.101161][GENERATOR][INFO]Successfully sampled '2' jobs, they will soon be submitted to the execution target.\\r\\n[2020-06-01T06:55:44.1226518Z][SCHEDULER][INFO]Scheduling job, id='HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_1'\\r\\n[2020-06-01T06:55:44.1084916Z][SCHEDULER][INFO]The execution environment was successfully prepared.\\r\\n[2020-06-01T06:55:44.1095866Z][SCHEDULER][INFO]Scheduling job, id='HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_0'\\r\\n[2020-06-01T06:55:45.0425698Z][SCHEDULER][INFO]Successfully scheduled a job. Id='HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_0'\\r\\n[2020-06-01T06:55:45.9725198Z][SCHEDULER][INFO]Successfully scheduled a job. Id='HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_1'\\r\\n[2020-06-01T06:57:40.970465][GENERATOR][INFO]Trying to sample '1' jobs from the hyperparameter space\\r\\n[2020-06-01T06:57:41.304444][GENERATOR][INFO]Successfully sampled '1' jobs, they will soon be submitted to the execution target.\\r\\n[2020-06-01T06:57:49.4354187Z][SCHEDULER][INFO]Scheduling job, id='HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_2'\\r\\n[2020-06-01T06:57:50.7860499Z][SCHEDULER][INFO]Successfully scheduled a job. Id='HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_2'\\r\\n[2020-06-01T06:58:41.566158][GENERATOR][INFO]Trying to sample '1' jobs from the hyperparameter space\\r\\n[2020-06-01T06:58:41.733358][GENERATOR][INFO]Successfully sampled '1' jobs, they will soon be submitted to the execution target.\\r\\n[2020-06-01T06:58:51.6790251Z][SCHEDULER][INFO]Scheduling job, id='HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_3'\\r\\n[2020-06-01T06:58:54.7544226Z][SCHEDULER][INFO]Successfully scheduled a job. Id='HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_3'\\r\\n[2020-06-01T06:59:41.684939][GENERATOR][INFO]Trying to sample '1' jobs from the hyperparameter space\\r\\n[2020-06-01T06:59:41.698459][GENERATOR][WARNING]Could not sample any more jobs from the space.\\r\\n[2020-06-01T07:01:16.887658][CONTROLLER][INFO]Experiment was 'ExperimentStatus.RUNNING', is 'ExperimentStatus.FINISHED'.\\n\\nRun is completed.\", \"graph\": {}, \"widget_settings\": {\"childWidgetDisplay\": \"popup\", \"send_telemetry\": false, \"log_level\": \"INFO\", \"sdk_version\": \"1.5.0\"}, \"loading\": false}"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# launch the widget to view the progress and results\n",
    "from azureml.widgets import RunDetails\n",
    "\n",
    "RunDetails(hd_run).show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RunId: HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\n",
      "Web View: https://ml.azure.com/experiments/diabetes_dxd/runs/HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd?wsid=/subscriptions/f80606e5-788f-4dc3-a9ea-2eb9a7836082/resourcegroups/adlsgen2/workspaces/eacbmlservicews\n",
      "\n",
      "Streaming azureml-logs/hyperdrive.txt\n",
      "=====================================\n",
      "\n",
      "\"<START>[2020-06-01T06:48:37.143894][API][INFO]Experiment created<END>\\n\"<START>[2020-06-01T06:48:38.4940411Z][SCHEDULER][INFO]The execution environment is being prepared. Please be patient as it can take a few minutes.<END>\"<START>[2020-06-01T06:48:38.511309][GENERATOR][INFO]Trying to sample '2' jobs from the hyperparameter space<END>\\n\"\"<START>[2020-06-01T06:48:41.101161][GENERATOR][INFO]Successfully sampled '2' jobs, they will soon be submitted to the execution target.<END>\\n\"\n",
      "\n",
      "Execution Summary\n",
      "=================\n",
      "RunId: HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd\n",
      "Web View: https://ml.azure.com/experiments/diabetes_dxd/runs/HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd?wsid=/subscriptions/f80606e5-788f-4dc3-a9ea-2eb9a7836082/resourcegroups/adlsgen2/workspaces/eacbmlservicews\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'runId': 'HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd',\n",
       " 'target': 'myComputeCluster',\n",
       " 'status': 'Completed',\n",
       " 'startTimeUtc': '2020-06-01T06:48:36.765215Z',\n",
       " 'endTimeUtc': '2020-06-01T07:01:16.674336Z',\n",
       " 'properties': {'primary_metric_config': '{\"name\": \"score\", \"goal\": \"maximize\"}',\n",
       "  'resume_from': 'null',\n",
       "  'runTemplate': 'HyperDrive',\n",
       "  'azureml.runsource': 'hyperdrive',\n",
       "  'platform': 'AML',\n",
       "  'ContentSnapshotId': '5db72d9f-93a1-4ed1-8bc0-c7aa7ed90307',\n",
       "  'score': '0.45259921776197887',\n",
       "  'best_child_run_id': 'HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_3',\n",
       "  'best_metric_status': 'Succeeded'},\n",
       " 'inputDatasets': [],\n",
       " 'logFiles': {'azureml-logs/hyperdrive.txt': 'https://eacbmlservicew2479569759.blob.core.windows.net/azureml/ExperimentRun/dcid.HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd/azureml-logs/hyperdrive.txt?sv=2019-02-02&sr=b&sig=9U4mUyrWuWu2uhs2LWUMtqyXCNtmtVtmkK8LkbAGV3U%3D&st=2020-06-01T06%3A51%3A17Z&se=2020-06-01T15%3A01%3A17Z&sp=r'}}"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# specify show_output to True for a verbose log\n",
    "hd_run.wait_for_completion(show_output=True) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'script': 'train.py',\n",
       " 'useAbsolutePath': False,\n",
       " 'arguments': ['--regularization', '0.01'],\n",
       " 'sourceDirectoryDataStore': None,\n",
       " 'framework': 'Python',\n",
       " 'communicator': 'None',\n",
       " 'target': 'myComputeCluster',\n",
       " 'dataReferences': {},\n",
       " 'data': {},\n",
       " 'jobName': None,\n",
       " 'maxRunDurationSeconds': None,\n",
       " 'nodeCount': 1,\n",
       " 'environment': {'name': 'diabetes_remote_env',\n",
       "  'version': 'Autosave_2020-06-01T06:48:37Z_2b67a5f8',\n",
       "  'python': {'interpreterPath': 'python',\n",
       "   'userManagedDependencies': False,\n",
       "   'condaDependencies': {'channels': ['anaconda', 'conda-forge'],\n",
       "    'dependencies': ['python=3.6.2',\n",
       "     {'pip': ['azureml-defaults', 'inference-schema[numpy-support]']},\n",
       "     'scikit-learn',\n",
       "     'pandas',\n",
       "     'numpy',\n",
       "     'joblib'],\n",
       "    'name': 'azureml_789e4a31052d218b96e17e2da348e37b'},\n",
       "   'baseCondaEnvironment': None},\n",
       "  'environmentVariables': {'EXAMPLE_ENV_VAR': 'EXAMPLE_VALUE'},\n",
       "  'docker': {'baseImage': 'mcr.microsoft.com/azureml/base:intelmpi2018.3-ubuntu16.04',\n",
       "   'baseDockerfile': None,\n",
       "   'baseImageRegistry': {'address': None, 'username': None, 'password': None},\n",
       "   'enabled': True,\n",
       "   'arguments': []},\n",
       "  'spark': {'repositories': [], 'packages': [], 'precachePackages': True},\n",
       "  'inferencingStackVersion': None},\n",
       " 'history': {'outputCollection': True,\n",
       "  'directoriesToWatch': ['logs'],\n",
       "  'snapshotProject': True},\n",
       " 'spark': {'configuration': {'spark.app.name': 'Azure ML Experiment',\n",
       "   'spark.yarn.maxAppAttempts': '1'}},\n",
       " 'parallelTask': {'maxRetriesPerWorker': 0,\n",
       "  'workerCountPerNode': 1,\n",
       "  'terminalExitCodes': None,\n",
       "  'configuration': {}},\n",
       " 'amlCompute': {'name': None,\n",
       "  'vmSize': None,\n",
       "  'retainCluster': False,\n",
       "  'clusterMaxNodeCount': 1},\n",
       " 'tensorflow': {'workerCount': 1, 'parameterServerCount': 1},\n",
       " 'mpi': {'processCountPerNode': 1},\n",
       " 'hdi': {'yarnDeployMode': 'Cluster'},\n",
       " 'containerInstance': {'region': None, 'cpuCores': 2, 'memoryGb': 3.5},\n",
       " 'exposedPorts': None,\n",
       " 'docker': {'useDocker': True,\n",
       "  'sharedVolumes': True,\n",
       "  'shmSize': '2g',\n",
       "  'arguments': []},\n",
       " 'cmk8sCompute': {'configuration': {}},\n",
       " 'itpCompute': {'configuration': {}}}"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_run = hd_run.get_best_run_by_primary_metric()\n",
    "best_run_metrics = best_run.get_metrics()\n",
    "parameter_values = best_run.get_details()\n",
    "parameter_values['runDefinition']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['--regularization', '0.01']"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameter_values = best_run.get_details()['runDefinition']['arguments']\n",
    "parameter_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Run Id:  HD_c19b511d-c899-4ef2-9f08-e3c6f5f6e7dd_3\n",
      "\n",
      " Score: 0.45259921776197887\n",
      "\n",
      " regularization:  0.01\n"
     ]
    }
   ],
   "source": [
    "print('Best Run Id: ', best_run.id)\n",
    "print('\\n Score:', best_run_metrics['score'])\n",
    "print('\\n regularization: ',parameter_values[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Register the BEST model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['azureml-logs/55_azureml-execution-tvmps_d88853636b2da6c0ad1df16902a52468911b9ff91c9009cea7515e0a7dcdcba5_d.txt', 'azureml-logs/65_job_prep-tvmps_d88853636b2da6c0ad1df16902a52468911b9ff91c9009cea7515e0a7dcdcba5_d.txt', 'azureml-logs/70_driver_log.txt', 'azureml-logs/75_job_post-tvmps_d88853636b2da6c0ad1df16902a52468911b9ff91c9009cea7515e0a7dcdcba5_d.txt', 'azureml-logs/process_info.json', 'azureml-logs/process_status.json', 'logs/azureml/112_azureml.log', 'logs/azureml/job_prep_azureml.log', 'logs/azureml/job_release_azureml.log', 'outputs/diabetes_HD_remote_model.pkl']\n"
     ]
    }
   ],
   "source": [
    "print(best_run.get_file_names())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "diabetes_HD_best_model\tdiabetes_HD_best_model:1\t1\n"
     ]
    }
   ],
   "source": [
    "# register model \n",
    "model = best_run.register_model(model_name='diabetes_HD_best_model', model_path='outputs/diabetes_HD_remote_model.pkl')\n",
    "\n",
    "print(model.name, model.id, model.version, sep='\\t')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create a scoring script"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overwriting score.py\n"
     ]
    }
   ],
   "source": [
    "%%writefile score.py\n",
    "\n",
    "import joblib\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "from inference_schema.schema_decorators import input_schema, output_schema\n",
    "from inference_schema.parameter_types.numpy_parameter_type import NumpyParameterType\n",
    "\n",
    "\n",
    "# The init() method is called once, when the web service starts up.\n",
    "#\n",
    "# Typically you would deserialize the model file, as shown here using joblib,\n",
    "# and store it in a global variable so your run() method can access it later.\n",
    "def init():\n",
    "    global model\n",
    "\n",
    "    # The AZUREML_MODEL_DIR environment variable indicates\n",
    "    # a directory containing the model file you registered.\n",
    "    model_filename = 'diabetes_HD_remote_model.pkl'\n",
    "    model_path = os.path.join(os.environ['AZUREML_MODEL_DIR'], model_filename)\n",
    "#FileNotFoundError: [Errno 2] No such file or directory: 'azureml-models/diabetes_HD_best_model/1/diabetes_hyperdrive_model.pkl'\n",
    "\n",
    "    model = joblib.load(model_path)\n",
    "\n",
    "\n",
    "# The run() method is called each time a request is made to the scoring API.\n",
    "#\n",
    "# Shown here are the optional input_schema and output_schema decorators\n",
    "# from the inference-schema pip package. Using these decorators on your\n",
    "# run() method parses and validates the incoming payload against\n",
    "# the example input you provide here. This will also generate a Swagger\n",
    "# API document for your web service.\n",
    "@input_schema('data', NumpyParameterType(np.array([[59, 2, 32.1, 101.0, 157, 93.2, 38.0, 4.0, 4.8598, 87]])))\n",
    "@output_schema(NumpyParameterType(np.array([151.000])))\n",
    "def run(data):\n",
    "    # Use the model object loaded by init().\n",
    "    result = model.predict(data)\n",
    "\n",
    "    # You can return any JSON-serializable object.\n",
    "    return result.tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define the (inference) environement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "from azureml.core import Environment\n",
    "from azureml.core.conda_dependencies import CondaDependencies\n",
    "\n",
    "\n",
    "environment = Environment('my-sklearn-environment')\n",
    "environment.python.conda_dependencies = CondaDependencies.create(pip_packages=[\n",
    "    'azureml-defaults',\n",
    "    'inference-schema[numpy-support]',\n",
    "    'joblib',\n",
    "    'numpy',\n",
    "    'scikit-learn'\n",
    "])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define a inference configuration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "from azureml.core.model import InferenceConfig\n",
    "\n",
    "\n",
    "inference_config = InferenceConfig(entry_script='score.py', environment=environment)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Deploy in a custom environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running.....................\n",
      "Succeeded\n",
      "ACI service creation operation finished, operation \"Succeeded\"\n"
     ]
    }
   ],
   "source": [
    "from azureml.core import Webservice\n",
    "from azureml.core.webservice import AciWebservice\n",
    "from azureml.exceptions import WebserviceException\n",
    "from azureml.core.model import Model, InferenceConfig # à ajouter ?\n",
    "\n",
    "\n",
    "service_name = 'diabetes-custom-service2hd'\n",
    "\n",
    "# Remove any existing service under the same name.\n",
    "try:\n",
    "    Webservice(ws, service_name).delete()\n",
    "except WebserviceException:\n",
    "    pass\n",
    "\n",
    "aci_config = AciWebservice.deploy_configuration(cpu_cores=1, memory_gb=1)\n",
    "\n",
    "service = Model.deploy(workspace=ws,\n",
    "                       name=service_name,\n",
    "                       models=[model],\n",
    "                       inference_config=inference_config,\n",
    "                       deployment_config=aci_config)\n",
    "\n",
    "service.wait_for_deployment(show_output=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transitioning\n"
     ]
    }
   ],
   "source": [
    "print(service.state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2020-05-27T06:45:44,128953160+00:00 - iot-server/run \n",
      "2020-05-27T06:45:44,134813858+00:00 - gunicorn/run \n",
      "2020-05-27T06:45:44,139708157+00:00 - rsyslog/run \n",
      "2020-05-27T06:45:44,144495856+00:00 - nginx/run \n",
      "/usr/sbin/nginx: /azureml-envs/azureml_fb6ef452c82dd8280844ad10cb54e4a8/lib/libcrypto.so.1.0.0: no version information available (required by /usr/sbin/nginx)\n",
      "/usr/sbin/nginx: /azureml-envs/azureml_fb6ef452c82dd8280844ad10cb54e4a8/lib/libcrypto.so.1.0.0: no version information available (required by /usr/sbin/nginx)\n",
      "/usr/sbin/nginx: /azureml-envs/azureml_fb6ef452c82dd8280844ad10cb54e4a8/lib/libssl.so.1.0.0: no version information available (required by /usr/sbin/nginx)\n",
      "/usr/sbin/nginx: /azureml-envs/azureml_fb6ef452c82dd8280844ad10cb54e4a8/lib/libssl.so.1.0.0: no version information available (required by /usr/sbin/nginx)\n",
      "/usr/sbin/nginx: /azureml-envs/azureml_fb6ef452c82dd8280844ad10cb54e4a8/lib/libssl.so.1.0.0: no version information available (required by /usr/sbin/nginx)\n",
      "EdgeHubConnectionString and IOTEDGE_IOTHUBHOSTNAME are not set. Exiting...\n",
      "2020-05-27T06:45:44,266936129+00:00 - iot-server/finish 1 0\n",
      "2020-05-27T06:45:44,268242328+00:00 - Exit code 1 is normal. Not restarting iot-server.\n",
      "Starting gunicorn 19.9.0\n",
      "Listening at: http://127.0.0.1:31311 (15)\n",
      "Using worker: sync\n",
      "worker timeout is set to 300\n",
      "Booting worker with pid: 41\n",
      "Initialized PySpark session.\n",
      "Initializing logger\n",
      "Starting up app insights client\n",
      "Starting up request id generator\n",
      "Starting up app insight hooks\n",
      "Invoking user's init function\n",
      "Users's init has completed successfully\n",
      "Scoring timeout is found from os.environ: 60000 ms\n",
      "/azureml-envs/azureml_fb6ef452c82dd8280844ad10cb54e4a8/lib/python3.6/site-packages/sklearn/base.py:334: UserWarning: Trying to unpickle estimator Ridge from version 0.22.1 when using version 0.23.1. This might lead to breaking code or invalid results. Use at your own risk.\n",
      "  UserWarning)\n",
      "200\n",
      "127.0.0.1 - - [27/May/2020:06:46:04 +0000] \"GET /swagger.json HTTP/1.0\" 200 2086 \"-\" \"Go-http-client/1.1\"\n",
      "127.0.0.1 - - [27/May/2020:06:47:20 +0000] \"POST /cgi-bin/mainfunction.cgi HTTP/1.0\" 404 232 \"-\" \"XTC\"\n",
      "200\n",
      "127.0.0.1 - - [27/May/2020:06:54:03 +0000] \"GET /swagger.json HTTP/1.0\" 200 2086 \"-\" \"Go-http-client/1.1\"\n",
      "200\n",
      "127.0.0.1 - - [27/May/2020:06:54:10 +0000] \"GET /swagger.json HTTP/1.0\" 200 2086 \"-\" \"Go-http-client/1.1\"\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(service.get_logs())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "http://1ab3d401-9a51-4305-8298-c01d4ce44faf.westeurope.azurecontainer.io/score\n"
     ]
    }
   ],
   "source": [
    "print(service.scoring_uri)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "diabetes-custom-service2hd\thttp://1ab3d401-9a51-4305-8298-c01d4ce44faf.westeurope.azurecontainer.io/score\n"
     ]
    }
   ],
   "source": [
    "# reload the model\n",
    "\n",
    "from azureml.core import Webservice\n",
    "\n",
    "\n",
    "service_name = 'diabetes-custom-service2hd'\n",
    "service = Webservice(ws, service_name)\n",
    "\n",
    "print(service.name, service.scoring_uri, sep='\\t')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test the service"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[210.73488558853785], [212.11233401556842]]\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "\n",
    "input_payload = json.dumps({\n",
    "    'data': [\n",
    "        [59, 2, 32.1, 101.0, 157, 93.2, 38.0, 4.0, 4.8598, 87],\n",
    "        [69, 2, 32.1, 101.0, 157, 93.2, 38.0, 4.0, 4.8598, 87]\n",
    "    ],\n",
    "    'method': 'predict'  # If you have a classification model, you can get probabilities by changing this to 'predict_proba'.\n",
    "})\n",
    "\n",
    "output = service.run(input_payload)\n",
    "\n",
    "print(output)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#service.delete()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.6 - AzureML",
   "language": "python",
   "name": "python3-azureml"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}