skpipe.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mlrun import mlconf, import_function, set_environment, run_local\n",
    "from mlrun.platforms import  mount_pvc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import AutoML Feature Analysis serverless function from MLRun marketplace "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "describe = import_function('hub://describe')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "function: describe\n",
      "describe and visualizes dataset stats\n",
      "default handler: summarize\n",
      "entry points:\n",
      "  summarize: Summarize a table\n",
      "    context(MLClientCtx)  - the function context, default=\n",
      "    table(DataItem)  - MLRun input pointing to pandas dataframe (csv/parquet file path), default=\n",
      "    label_column(str)  - ground truth column label, default=None\n",
      "    class_labels(List[str])  - label for each class in tables and plots, default=[]\n",
      "    plot_hist(bool)  - (True) set this to False for large tables, default=True\n",
      "    plots_dest(str)  - destination folder of summary plots (relative to artifact_path), default=plots\n",
      "    update_dataset  - when the table is a registered dataset update the charts in-place, default=False\n"
     ]
    }
   ],
   "source": [
    "# read the remote function doc, params, usage\n",
    "describe.doc()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run a job to process data and generate plots (saved in <b>`./data/plots`</b>)\n",
    "\n",
    "<b> Run it locally </b>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "> 2020-11-16 00:51:57,066 [info] starting run describe-summarize uid=1015e78baf3d4e47858afdd0da08d911 DB=http://localhost:8080\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>\n",
       ".dictlist {\n",
       "  background-color: #ca8452;\n",
       "  text-align: center;\n",
       "  margin: 4px;\n",
       "  border-radius: 3px; padding: 0px 3px 1px 3px; display: inline-block;}\n",
       ".artifact {\n",
       "  cursor: pointer;\n",
       "  background-color: #ca8452;\n",
       "  text-align: left;\n",
       "  margin: 4px; border-radius: 3px; padding: 0px 3px 1px 3px; display: inline-block;\n",
       "}\n",
       "div.block.hidden {\n",
       "  display: none;\n",
       "}\n",
       ".clickable {\n",
       "  cursor: pointer;\n",
       "}\n",
       ".ellipsis {\n",
       "  display: inline-block;\n",
       "  max-width: 60px;\n",
       "  white-space: nowrap;\n",
       "  overflow: hidden;\n",
       "  text-overflow: ellipsis;\n",
       "}\n",
       ".master-wrapper {\n",
       "  display: flex;\n",
       "  flex-flow: row nowrap;\n",
       "  justify-content: flex-start;\n",
       "  align-items: stretch;\n",
       "}\n",
       ".master-tbl {\n",
       "  flex: 3\n",
       "}\n",
       ".master-wrapper > div {\n",
       "  margin: 4px;\n",
       "  padding: 10px;\n",
       "}\n",
       "iframe.fileview {\n",
       "  border: 0 none;\n",
       "  height: 100%;\n",
       "  width: 100%;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       ".pane-header-title {\n",
       "  width: 80%;\n",
       "  font-weight: 500;\n",
       "}\n",
       ".pane-header {\n",
       "  line-height: 1;\n",
       "  background-color: #ca8452;\n",
       "  padding: 3px;\n",
       "}\n",
       ".pane-header .close {\n",
       "  font-size: 20px;\n",
       "  font-weight: 700;\n",
       "  float: right;\n",
       "  margin-top: -5px;\n",
       "}\n",
       ".master-wrapper .right-pane {\n",
       "  border: 1px inset silver;\n",
       "  width: 40%;\n",
       "  min-height: 300px;\n",
       "  flex: 3\n",
       "  min-width: 500px;\n",
       "}\n",
       ".master-wrapper * {\n",
       "  box-sizing: border-box;\n",
       "}\n",
       "</style><script>\n",
       "function copyToClipboard(fld) {\n",
       "    if (document.queryCommandSupported && document.queryCommandSupported('copy')) {\n",
       "        var textarea = document.createElement('textarea');\n",
       "        textarea.textContent = fld.innerHTML;\n",
       "        textarea.style.position = 'fixed';\n",
       "        document.body.appendChild(textarea);\n",
       "        textarea.select();\n",
       "\n",
       "        try {\n",
       "            return document.execCommand('copy'); // Security exception may be thrown by some browsers.\n",
       "        } catch (ex) {\n",
       "\n",
       "        } finally {\n",
       "            document.body.removeChild(textarea);\n",
       "        }\n",
       "    }\n",
       "}\n",
       "function expandPanel(el) {\n",
       "  const panelName = \"#\" + el.getAttribute('paneName');\n",
       "  console.log(el.title);\n",
       "\n",
       "  document.querySelector(panelName + \"-title\").innerHTML = el.title\n",
       "  iframe = document.querySelector(panelName + \"-body\");\n",
       "\n",
       "  const tblcss = `<style> body { font-family: Arial, Helvetica, sans-serif;}\n",
       "    #csv { margin-bottom: 15px; }\n",
       "    #csv table { border-collapse: collapse;}\n",
       "    #csv table td { padding: 4px 8px; border: 1px solid silver;} </style>`;\n",
       "\n",
       "  function csvToHtmlTable(str) {\n",
       "    return '<div id=\"csv\"><table><tr><td>' +  str.replace(/[\\n\\r]+$/g, '').replace(/[\\n\\r]+/g, '</td></tr><tr><td>')\n",
       "      .replace(/,/g, '</td><td>') + '</td></tr></table></div>';\n",
       "  }\n",
       "\n",
       "  function reqListener () {\n",
       "    if (el.title.endsWith(\".csv\")) {\n",
       "      iframe.setAttribute(\"srcdoc\", tblcss + csvToHtmlTable(this.responseText));\n",
       "    } else {\n",
       "      iframe.setAttribute(\"srcdoc\", this.responseText);\n",
       "    }\n",
       "    console.log(this.responseText);\n",
       "  }\n",
       "\n",
       "  const oReq = new XMLHttpRequest();\n",
       "  oReq.addEventListener(\"load\", reqListener);\n",
       "  oReq.open(\"GET\", el.title);\n",
       "  oReq.send();\n",
       "\n",
       "\n",
       "  //iframe.src = el.title;\n",
       "  const resultPane = document.querySelector(panelName + \"-pane\");\n",
       "  if (resultPane.classList.contains(\"hidden\")) {\n",
       "    resultPane.classList.remove(\"hidden\");\n",
       "  }\n",
       "}\n",
       "function closePanel(el) {\n",
       "  const panelName = \"#\" + el.getAttribute('paneName')\n",
       "  const resultPane = document.querySelector(panelName + \"-pane\");\n",
       "  if (!resultPane.classList.contains(\"hidden\")) {\n",
       "    resultPane.classList.add(\"hidden\");\n",
       "  }\n",
       "}\n",
       "\n",
       "</script>\n",
       "<div class=\"master-wrapper\">\n",
       "  <div class=\"block master-tbl\"><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>project</th>\n",
       "      <th>uid</th>\n",
       "      <th>iter</th>\n",
       "      <th>start</th>\n",
       "      <th>state</th>\n",
       "      <th>name</th>\n",
       "      <th>labels</th>\n",
       "      <th>inputs</th>\n",
       "      <th>parameters</th>\n",
       "      <th>results</th>\n",
       "      <th>artifacts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>default</td>\n",
       "      <td><div title=\"1015e78baf3d4e47858afdd0da08d911\">...08d911</div></td>\n",
       "      <td>0</td>\n",
       "      <td>Nov 16 00:51:57</td>\n",
       "      <td>completed</td>\n",
       "      <td>describe-summarize</td>\n",
       "      <td><div class=\"dictlist\">kind=</div><div class=\"dictlist\">owner=jovyan</div><div class=\"dictlist\">host=8a47298fe0fe</div></td>\n",
       "      <td><div title=\"https://s3.wasabisys.com/iguazio/data/iris/iris_dataset.csv\">table</div></td>\n",
       "      <td><div class=\"dictlist\">label_column=label</div></td>\n",
       "      <td></td>\n",
       "      <td><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result1eb14426\" title=\"/files/data/plots/hist.html\">histograms</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result1eb14426\" title=\"/files/data/plots/violin.html\">violin</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result1eb14426\" title=\"/files/data/plots/imbalance.html\">imbalance</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result1eb14426\" title=\"/files/data/plots/imbalance-weights-vec.csv\">imbalance-weights-vec</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result1eb14426\" title=\"/files/data/plots/correlation-matrix.csv\">correlation-matrix</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result1eb14426\" title=\"/files/data/plots/corr.html\">correlation</div></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div></div>\n",
       "  <div id=\"result1eb14426-pane\" class=\"right-pane block hidden\">\n",
       "    <div class=\"pane-header\">\n",
       "      <span id=\"result1eb14426-title\" class=\"pane-header-title\">Title</span>\n",
       "      <span onclick=\"closePanel(this)\" paneName=\"result1eb14426\" class=\"close clickable\">&times;</span>\n",
       "    </div>\n",
       "    <iframe class=\"fileview\" id=\"result1eb14426-body\"></iframe>\n",
       "  </div>\n",
       "</div>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "to track results use .show() or .logs() or in CLI: \n",
      "!mlrun get run 1015e78baf3d4e47858afdd0da08d911 --project default , !mlrun logs 1015e78baf3d4e47858afdd0da08d911 --project default\n",
      "> 2020-11-16 00:52:01,937 [info] run executed, status=completed\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<mlrun.model.RunObject at 0x7fa2b00e4310>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Run the function against some data, results can be found at ./data/plots\n",
    "\n",
    "run_local(None, describe, params={'label_column': 'label'}, \n",
    "       inputs={\"table\": 'https://s3.wasabisys.com/iguazio/data/iris/iris_dataset.csv'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b> CheckMLRun UI to see the results !</b>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Running the same job remotely\n",
    "\n",
    "The same task can be remotely executed on a kuberneted cluster by using the run method, this require:\n",
    "* Setting MLRun over Kubernetes and pointing the MLRun client/notebook to the MLRun service URL\n",
    "* Specifying a shared storage media and credentials for artifacts (e.g. NFS, S3, V3IO, ..) \n",
    "\n",
    "Read more at [https://mlrun.readthedocs.io/en/latest/install.html](https://mlrun.readthedocs.io/en/latest/install.html)\n",
    "\n",
    "```python\n",
    "# connect our serverless function to the NFS volume \n",
    "describe.apply(mount_pvc('basic', volume_mount_path='/data'))\n",
    "\n",
    "# run the function pointing the input and output paths to the NFS volume\n",
    "describe.run(params={'label_column': 'label'}, \n",
    "             inputs={\"table\": 'https://s3.wasabisys.com/iguazio/data/iris/iris_dataset.csv'}, \n",
    "             artifact_path='/data')\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run Training Job"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = import_function('hub://sklearn_classifier')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "> 2020-11-16 00:52:32,379 [info] starting run sklearn-classifier-train_model uid=4be223546a5b44f3a0afd365d7c08c92 DB=http://localhost:8080\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>\n",
       ".dictlist {\n",
       "  background-color: #ca8452;\n",
       "  text-align: center;\n",
       "  margin: 4px;\n",
       "  border-radius: 3px; padding: 0px 3px 1px 3px; display: inline-block;}\n",
       ".artifact {\n",
       "  cursor: pointer;\n",
       "  background-color: #ca8452;\n",
       "  text-align: left;\n",
       "  margin: 4px; border-radius: 3px; padding: 0px 3px 1px 3px; display: inline-block;\n",
       "}\n",
       "div.block.hidden {\n",
       "  display: none;\n",
       "}\n",
       ".clickable {\n",
       "  cursor: pointer;\n",
       "}\n",
       ".ellipsis {\n",
       "  display: inline-block;\n",
       "  max-width: 60px;\n",
       "  white-space: nowrap;\n",
       "  overflow: hidden;\n",
       "  text-overflow: ellipsis;\n",
       "}\n",
       ".master-wrapper {\n",
       "  display: flex;\n",
       "  flex-flow: row nowrap;\n",
       "  justify-content: flex-start;\n",
       "  align-items: stretch;\n",
       "}\n",
       ".master-tbl {\n",
       "  flex: 3\n",
       "}\n",
       ".master-wrapper > div {\n",
       "  margin: 4px;\n",
       "  padding: 10px;\n",
       "}\n",
       "iframe.fileview {\n",
       "  border: 0 none;\n",
       "  height: 100%;\n",
       "  width: 100%;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       ".pane-header-title {\n",
       "  width: 80%;\n",
       "  font-weight: 500;\n",
       "}\n",
       ".pane-header {\n",
       "  line-height: 1;\n",
       "  background-color: #ca8452;\n",
       "  padding: 3px;\n",
       "}\n",
       ".pane-header .close {\n",
       "  font-size: 20px;\n",
       "  font-weight: 700;\n",
       "  float: right;\n",
       "  margin-top: -5px;\n",
       "}\n",
       ".master-wrapper .right-pane {\n",
       "  border: 1px inset silver;\n",
       "  width: 40%;\n",
       "  min-height: 300px;\n",
       "  flex: 3\n",
       "  min-width: 500px;\n",
       "}\n",
       ".master-wrapper * {\n",
       "  box-sizing: border-box;\n",
       "}\n",
       "</style><script>\n",
       "function copyToClipboard(fld) {\n",
       "    if (document.queryCommandSupported && document.queryCommandSupported('copy')) {\n",
       "        var textarea = document.createElement('textarea');\n",
       "        textarea.textContent = fld.innerHTML;\n",
       "        textarea.style.position = 'fixed';\n",
       "        document.body.appendChild(textarea);\n",
       "        textarea.select();\n",
       "\n",
       "        try {\n",
       "            return document.execCommand('copy'); // Security exception may be thrown by some browsers.\n",
       "        } catch (ex) {\n",
       "\n",
       "        } finally {\n",
       "            document.body.removeChild(textarea);\n",
       "        }\n",
       "    }\n",
       "}\n",
       "function expandPanel(el) {\n",
       "  const panelName = \"#\" + el.getAttribute('paneName');\n",
       "  console.log(el.title);\n",
       "\n",
       "  document.querySelector(panelName + \"-title\").innerHTML = el.title\n",
       "  iframe = document.querySelector(panelName + \"-body\");\n",
       "\n",
       "  const tblcss = `<style> body { font-family: Arial, Helvetica, sans-serif;}\n",
       "    #csv { margin-bottom: 15px; }\n",
       "    #csv table { border-collapse: collapse;}\n",
       "    #csv table td { padding: 4px 8px; border: 1px solid silver;} </style>`;\n",
       "\n",
       "  function csvToHtmlTable(str) {\n",
       "    return '<div id=\"csv\"><table><tr><td>' +  str.replace(/[\\n\\r]+$/g, '').replace(/[\\n\\r]+/g, '</td></tr><tr><td>')\n",
       "      .replace(/,/g, '</td><td>') + '</td></tr></table></div>';\n",
       "  }\n",
       "\n",
       "  function reqListener () {\n",
       "    if (el.title.endsWith(\".csv\")) {\n",
       "      iframe.setAttribute(\"srcdoc\", tblcss + csvToHtmlTable(this.responseText));\n",
       "    } else {\n",
       "      iframe.setAttribute(\"srcdoc\", this.responseText);\n",
       "    }\n",
       "    console.log(this.responseText);\n",
       "  }\n",
       "\n",
       "  const oReq = new XMLHttpRequest();\n",
       "  oReq.addEventListener(\"load\", reqListener);\n",
       "  oReq.open(\"GET\", el.title);\n",
       "  oReq.send();\n",
       "\n",
       "\n",
       "  //iframe.src = el.title;\n",
       "  const resultPane = document.querySelector(panelName + \"-pane\");\n",
       "  if (resultPane.classList.contains(\"hidden\")) {\n",
       "    resultPane.classList.remove(\"hidden\");\n",
       "  }\n",
       "}\n",
       "function closePanel(el) {\n",
       "  const panelName = \"#\" + el.getAttribute('paneName')\n",
       "  const resultPane = document.querySelector(panelName + \"-pane\");\n",
       "  if (!resultPane.classList.contains(\"hidden\")) {\n",
       "    resultPane.classList.add(\"hidden\");\n",
       "  }\n",
       "}\n",
       "\n",
       "</script>\n",
       "<div class=\"master-wrapper\">\n",
       "  <div class=\"block master-tbl\"><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>project</th>\n",
       "      <th>uid</th>\n",
       "      <th>iter</th>\n",
       "      <th>start</th>\n",
       "      <th>state</th>\n",
       "      <th>name</th>\n",
       "      <th>labels</th>\n",
       "      <th>inputs</th>\n",
       "      <th>parameters</th>\n",
       "      <th>results</th>\n",
       "      <th>artifacts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>default</td>\n",
       "      <td><div title=\"4be223546a5b44f3a0afd365d7c08c92\">...c08c92</div></td>\n",
       "      <td>0</td>\n",
       "      <td>Nov 16 00:52:32</td>\n",
       "      <td>completed</td>\n",
       "      <td>sklearn-classifier-train_model</td>\n",
       "      <td><div class=\"dictlist\">kind=</div><div class=\"dictlist\">owner=jovyan</div><div class=\"dictlist\">host=8a47298fe0fe</div><div class=\"dictlist\">class=sklearn.ensemble.RandomForestClassifier</div></td>\n",
       "      <td><div title=\"https://s3.wasabisys.com/iguazio/data/iris/iris_dataset.csv\">dataset</div></td>\n",
       "      <td><div class=\"dictlist\">sample=-1</div><div class=\"dictlist\">label_column=label</div><div class=\"dictlist\">test_size=0.1</div><div class=\"dictlist\">model_pkg_class=sklearn.ensemble.RandomForestClassifier</div></td>\n",
       "      <td><div class=\"dictlist\">accuracy=0.9117647058823529</div><div class=\"dictlist\">test-error=0.08823529411764706</div><div class=\"dictlist\">auc-micro=0.9954584775086505</div><div class=\"dictlist\">auc-weighted=0.9949732620320856</div><div class=\"dictlist\">f1-score=0.9011764705882354</div><div class=\"dictlist\">precision_score=0.907051282051282</div><div class=\"dictlist\">recall_score=0.898148148148148</div></td>\n",
       "      <td><div title=\"/home/jovyan/data/data/test_set.parquet\">test_set</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result5611c6ee\" title=\"/files/data/model/plots/confusion-matrix.html\">confusion-matrix</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result5611c6ee\" title=\"/files/data/model/plots/feature-importances.html\">feature-importances</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result5611c6ee\" title=\"/files/data/model/plots/precision-recall-multiclass.html\">precision-recall-multiclass</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"result5611c6ee\" title=\"/files/data/model/plots/roc-multiclass.html\">roc-multiclass</div><div title=\"/home/jovyan/data/model\">model</div></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div></div>\n",
       "  <div id=\"result5611c6ee-pane\" class=\"right-pane block hidden\">\n",
       "    <div class=\"pane-header\">\n",
       "      <span id=\"result5611c6ee-title\" class=\"pane-header-title\">Title</span>\n",
       "      <span onclick=\"closePanel(this)\" paneName=\"result5611c6ee\" class=\"close clickable\">&times;</span>\n",
       "    </div>\n",
       "    <iframe class=\"fileview\" id=\"result5611c6ee-body\"></iframe>\n",
       "  </div>\n",
       "</div>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "to track results use .show() or .logs() or in CLI: \n",
      "!mlrun get run 4be223546a5b44f3a0afd365d7c08c92 --project default , !mlrun logs 4be223546a5b44f3a0afd365d7c08c92 --project default\n",
      "> 2020-11-16 00:52:34,109 [info] run executed, status=completed\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "# run the function pointing the input and output paths to the NFS volume\n",
    "train_run = run_local(None, train, \n",
    "                      params={\"sample\"          : -1, \n",
    "                              \"label_column\"    : \"label\",\n",
    "                              \"test_size\"       : 0.10,\n",
    "                              'model_pkg_class': \"sklearn.ensemble.RandomForestClassifier\"},\n",
    "                      inputs={\"dataset\": 'https://s3.wasabisys.com/iguazio/data/iris/iris_dataset.csv'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Check the run results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'accuracy': 0.9117647058823529,\n",
       " 'test-error': 0.08823529411764706,\n",
       " 'auc-micro': 0.9954584775086505,\n",
       " 'auc-weighted': 0.9949732620320856,\n",
       " 'f1-score': 0.9011764705882354,\n",
       " 'precision_score': 0.907051282051282,\n",
       " 'recall_score': 0.898148148148148,\n",
       " 'test_set': 'store://default/sklearn-classifier-train_model_test_set#4be223546a5b44f3a0afd365d7c08c92',\n",
       " 'confusion-matrix': '/home/jovyan/data/model/plots/confusion-matrix.html',\n",
       " 'feature-importances': '/home/jovyan/data/model/plots/feature-importances.html',\n",
       " 'precision-recall-multiclass': '/home/jovyan/data/model/plots/precision-recall-multiclass.html',\n",
       " 'roc-multiclass': '/home/jovyan/data/model/plots/roc-multiclass.html',\n",
       " 'model': 'store://default/sklearn-classifier-train_model_model#4be223546a5b44f3a0afd365d7c08c92'}"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_run.outputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run model validation job against the new model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "validate = import_function('hub://test_classifier')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "> 2020-11-16 00:57:33,568 [info] starting run test-classifier-test_classifier uid=61389a64bfd149afa5eaf3851e0019a1 DB=http://localhost:8080\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>\n",
       ".dictlist {\n",
       "  background-color: #ca8452;\n",
       "  text-align: center;\n",
       "  margin: 4px;\n",
       "  border-radius: 3px; padding: 0px 3px 1px 3px; display: inline-block;}\n",
       ".artifact {\n",
       "  cursor: pointer;\n",
       "  background-color: #ca8452;\n",
       "  text-align: left;\n",
       "  margin: 4px; border-radius: 3px; padding: 0px 3px 1px 3px; display: inline-block;\n",
       "}\n",
       "div.block.hidden {\n",
       "  display: none;\n",
       "}\n",
       ".clickable {\n",
       "  cursor: pointer;\n",
       "}\n",
       ".ellipsis {\n",
       "  display: inline-block;\n",
       "  max-width: 60px;\n",
       "  white-space: nowrap;\n",
       "  overflow: hidden;\n",
       "  text-overflow: ellipsis;\n",
       "}\n",
       ".master-wrapper {\n",
       "  display: flex;\n",
       "  flex-flow: row nowrap;\n",
       "  justify-content: flex-start;\n",
       "  align-items: stretch;\n",
       "}\n",
       ".master-tbl {\n",
       "  flex: 3\n",
       "}\n",
       ".master-wrapper > div {\n",
       "  margin: 4px;\n",
       "  padding: 10px;\n",
       "}\n",
       "iframe.fileview {\n",
       "  border: 0 none;\n",
       "  height: 100%;\n",
       "  width: 100%;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       ".pane-header-title {\n",
       "  width: 80%;\n",
       "  font-weight: 500;\n",
       "}\n",
       ".pane-header {\n",
       "  line-height: 1;\n",
       "  background-color: #ca8452;\n",
       "  padding: 3px;\n",
       "}\n",
       ".pane-header .close {\n",
       "  font-size: 20px;\n",
       "  font-weight: 700;\n",
       "  float: right;\n",
       "  margin-top: -5px;\n",
       "}\n",
       ".master-wrapper .right-pane {\n",
       "  border: 1px inset silver;\n",
       "  width: 40%;\n",
       "  min-height: 300px;\n",
       "  flex: 3\n",
       "  min-width: 500px;\n",
       "}\n",
       ".master-wrapper * {\n",
       "  box-sizing: border-box;\n",
       "}\n",
       "</style><script>\n",
       "function copyToClipboard(fld) {\n",
       "    if (document.queryCommandSupported && document.queryCommandSupported('copy')) {\n",
       "        var textarea = document.createElement('textarea');\n",
       "        textarea.textContent = fld.innerHTML;\n",
       "        textarea.style.position = 'fixed';\n",
       "        document.body.appendChild(textarea);\n",
       "        textarea.select();\n",
       "\n",
       "        try {\n",
       "            return document.execCommand('copy'); // Security exception may be thrown by some browsers.\n",
       "        } catch (ex) {\n",
       "\n",
       "        } finally {\n",
       "            document.body.removeChild(textarea);\n",
       "        }\n",
       "    }\n",
       "}\n",
       "function expandPanel(el) {\n",
       "  const panelName = \"#\" + el.getAttribute('paneName');\n",
       "  console.log(el.title);\n",
       "\n",
       "  document.querySelector(panelName + \"-title\").innerHTML = el.title\n",
       "  iframe = document.querySelector(panelName + \"-body\");\n",
       "\n",
       "  const tblcss = `<style> body { font-family: Arial, Helvetica, sans-serif;}\n",
       "    #csv { margin-bottom: 15px; }\n",
       "    #csv table { border-collapse: collapse;}\n",
       "    #csv table td { padding: 4px 8px; border: 1px solid silver;} </style>`;\n",
       "\n",
       "  function csvToHtmlTable(str) {\n",
       "    return '<div id=\"csv\"><table><tr><td>' +  str.replace(/[\\n\\r]+$/g, '').replace(/[\\n\\r]+/g, '</td></tr><tr><td>')\n",
       "      .replace(/,/g, '</td><td>') + '</td></tr></table></div>';\n",
       "  }\n",
       "\n",
       "  function reqListener () {\n",
       "    if (el.title.endsWith(\".csv\")) {\n",
       "      iframe.setAttribute(\"srcdoc\", tblcss + csvToHtmlTable(this.responseText));\n",
       "    } else {\n",
       "      iframe.setAttribute(\"srcdoc\", this.responseText);\n",
       "    }\n",
       "    console.log(this.responseText);\n",
       "  }\n",
       "\n",
       "  const oReq = new XMLHttpRequest();\n",
       "  oReq.addEventListener(\"load\", reqListener);\n",
       "  oReq.open(\"GET\", el.title);\n",
       "  oReq.send();\n",
       "\n",
       "\n",
       "  //iframe.src = el.title;\n",
       "  const resultPane = document.querySelector(panelName + \"-pane\");\n",
       "  if (resultPane.classList.contains(\"hidden\")) {\n",
       "    resultPane.classList.remove(\"hidden\");\n",
       "  }\n",
       "}\n",
       "function closePanel(el) {\n",
       "  const panelName = \"#\" + el.getAttribute('paneName')\n",
       "  const resultPane = document.querySelector(panelName + \"-pane\");\n",
       "  if (!resultPane.classList.contains(\"hidden\")) {\n",
       "    resultPane.classList.add(\"hidden\");\n",
       "  }\n",
       "}\n",
       "\n",
       "</script>\n",
       "<div class=\"master-wrapper\">\n",
       "  <div class=\"block master-tbl\"><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>project</th>\n",
       "      <th>uid</th>\n",
       "      <th>iter</th>\n",
       "      <th>start</th>\n",
       "      <th>state</th>\n",
       "      <th>name</th>\n",
       "      <th>labels</th>\n",
       "      <th>inputs</th>\n",
       "      <th>parameters</th>\n",
       "      <th>results</th>\n",
       "      <th>artifacts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>default</td>\n",
       "      <td><div title=\"61389a64bfd149afa5eaf3851e0019a1\">...0019a1</div></td>\n",
       "      <td>0</td>\n",
       "      <td>Nov 16 00:57:33</td>\n",
       "      <td>completed</td>\n",
       "      <td>test-classifier-test_classifier</td>\n",
       "      <td><div class=\"dictlist\">kind=</div><div class=\"dictlist\">owner=jovyan</div><div class=\"dictlist\">host=8a47298fe0fe</div></td>\n",
       "      <td><div title=\"store://default/sklearn-classifier-train_model_model#4be223546a5b44f3a0afd365d7c08c92\">models_path</div><div title=\"store://default/sklearn-classifier-train_model_test_set#4be223546a5b44f3a0afd365d7c08c92\">test_set</div></td>\n",
       "      <td><div class=\"dictlist\">label_column=label</div></td>\n",
       "      <td><div class=\"dictlist\">accuracy=0.9333333333333333</div><div class=\"dictlist\">test-error=0.06666666666666667</div><div class=\"dictlist\">auc-micro=0.9755555555555555</div><div class=\"dictlist\">auc-weighted=0.9944444444444444</div><div class=\"dictlist\">f1-score=0.9137254901960784</div><div class=\"dictlist\">precision_score=0.8888888888888888</div><div class=\"dictlist\">recall_score=0.9629629629629629</div></td>\n",
       "      <td><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"resultcd7e8cb8\" title=\"/files/data/plots/confusion-matrix.html\">confusion-matrix</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"resultcd7e8cb8\" title=\"/files/data/plots/feature-importances.html\">feature-importances</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"resultcd7e8cb8\" title=\"/files/data/plots/precision-recall-multiclass.html\">precision-recall-multiclass</div><div class=\"artifact\" onclick=\"expandPanel(this)\" paneName=\"resultcd7e8cb8\" title=\"/files/data/plots/roc-multiclass.html\">roc-multiclass</div><div title=\"/home/jovyan/data/test_set_preds.parquet\">test_set_preds</div></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div></div>\n",
       "  <div id=\"resultcd7e8cb8-pane\" class=\"right-pane block hidden\">\n",
       "    <div class=\"pane-header\">\n",
       "      <span id=\"resultcd7e8cb8-title\" class=\"pane-header-title\">Title</span>\n",
       "      <span onclick=\"closePanel(this)\" paneName=\"resultcd7e8cb8\" class=\"close clickable\">&times;</span>\n",
       "    </div>\n",
       "    <iframe class=\"fileview\" id=\"resultcd7e8cb8-body\"></iframe>\n",
       "  </div>\n",
       "</div>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "to track results use .show() or .logs() or in CLI: \n",
      "!mlrun get run 61389a64bfd149afa5eaf3851e0019a1 --project default , !mlrun logs 61389a64bfd149afa5eaf3851e0019a1 --project default\n",
      "> 2020-11-16 00:57:34,442 [info] run executed, status=completed\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<mlrun.model.RunObject at 0x7fa2c813f3d0>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "run_local(None, validate, params={'label_column': 'label'}, \n",
    "          inputs={\"models_path\" : train_run.outputs['model'],\n",
    "                  \"test_set\"    : train_run.outputs['test_set']})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}