conv1d-generic-bank-marketing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:17.694159Z",
     "start_time": "2020-09-19T17:11:17.610135Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 1;\n",
       "                var nbb_unformatted_code = \"%load_ext nb_black\\n%load_ext autoreload\\n\\n%autoreload 2\";\n",
       "                var nbb_formatted_code = \"%load_ext nb_black\\n%load_ext autoreload\\n\\n%autoreload 2\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%load_ext nb_black\n",
    "%load_ext autoreload\n",
    "\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:18.871796Z",
     "start_time": "2020-09-19T17:11:17.695143Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 2;\n",
       "                var nbb_unformatted_code = \"import os\\nfrom pathlib import Path\\n\\nfrom requests import get\\nimport pandas as pd\\nimport numpy as np\\n\\nfrom sklearn.metrics import roc_auc_score, log_loss\\nfrom sklearn.preprocessing import LabelEncoder\\nfrom tensorflow.keras.utils import plot_model\\nfrom tensorflow.keras.callbacks import EarlyStopping\\n\\nimport logging\\n\\nlogging.basicConfig(level=logging.WARN)\";\n",
       "                var nbb_formatted_code = \"import os\\nfrom pathlib import Path\\n\\nfrom requests import get\\nimport pandas as pd\\nimport numpy as np\\n\\nfrom sklearn.metrics import roc_auc_score, log_loss\\nfrom sklearn.preprocessing import LabelEncoder\\nfrom tensorflow.keras.utils import plot_model\\nfrom tensorflow.keras.callbacks import EarlyStopping\\n\\nimport logging\\n\\nlogging.basicConfig(level=logging.WARN)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "from requests import get\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "from sklearn.metrics import roc_auc_score, log_loss\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from tensorflow.keras.utils import plot_model\n",
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "\n",
    "import logging\n",
    "\n",
    "logging.basicConfig(level=logging.WARN)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:19.051396Z",
     "start_time": "2020-09-19T17:11:18.873200Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 3;\n",
       "                var nbb_unformatted_code = \"from xplainet.input_utils import preproc_dataset\\nfrom xplainet.model import build_model\\nfrom xplainet.random_utils import setup_seed, SEED\\nfrom sklearn.model_selection import StratifiedShuffleSplit\\n\\nimport matplotlib.pyplot as plt\\nfrom matplotlib.pyplot import imshow\\n\\n%matplotlib inline\\n\\nfrom zipfile import ZipFile\";\n",
       "                var nbb_formatted_code = \"from xplainet.input_utils import preproc_dataset\\nfrom xplainet.model import build_model\\nfrom xplainet.random_utils import setup_seed, SEED\\nfrom sklearn.model_selection import StratifiedShuffleSplit\\n\\nimport matplotlib.pyplot as plt\\nfrom matplotlib.pyplot import imshow\\n\\n%matplotlib inline\\n\\nfrom zipfile import ZipFile\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from xplainet.input_utils import preproc_dataset\n",
    "from xplainet.model import build_model\n",
    "from xplainet.random_utils import setup_seed, SEED\n",
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.pyplot import imshow\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "from zipfile import ZipFile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:19.074232Z",
     "start_time": "2020-09-19T17:11:19.052482Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 4;\n",
       "                var nbb_unformatted_code = \"setup_seed()\";\n",
       "                var nbb_formatted_code = \"setup_seed()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "setup_seed()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:19.156922Z",
     "start_time": "2020-09-19T17:11:19.075218Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 5;\n",
       "                var nbb_unformatted_code = \"def download(url, out, force=False, verify=True):\\n    out.parent.mkdir(parents=True, exist_ok=True)\\n    if force and out.exists():\\n        print(f\\\"Removing file at {str(out)}\\\")\\n        out.unlink()\\n\\n    if out.exists():\\n        print(\\\"File already exists.\\\")\\n        return\\n    print(f\\\"Downloading {url} at {str(out)} ...\\\")\\n    # open in binary mode\\n    with out.open(mode=\\\"wb\\\") as file:\\n        # get request\\n        response = get(url, verify=verify)\\n        for chunk in response.iter_content(100000):\\n            # write to file\\n            file.write(chunk)\";\n",
       "                var nbb_formatted_code = \"def download(url, out, force=False, verify=True):\\n    out.parent.mkdir(parents=True, exist_ok=True)\\n    if force and out.exists():\\n        print(f\\\"Removing file at {str(out)}\\\")\\n        out.unlink()\\n\\n    if out.exists():\\n        print(\\\"File already exists.\\\")\\n        return\\n    print(f\\\"Downloading {url} at {str(out)} ...\\\")\\n    # open in binary mode\\n    with out.open(mode=\\\"wb\\\") as file:\\n        # get request\\n        response = get(url, verify=verify)\\n        for chunk in response.iter_content(100000):\\n            # write to file\\n            file.write(chunk)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def download(url, out, force=False, verify=True):\n",
    "    out.parent.mkdir(parents=True, exist_ok=True)\n",
    "    if force and out.exists():\n",
    "        print(f\"Removing file at {str(out)}\")\n",
    "        out.unlink()\n",
    "\n",
    "    if out.exists():\n",
    "        print(\"File already exists.\")\n",
    "        return\n",
    "    print(f\"Downloading {url} at {str(out)} ...\")\n",
    "    # open in binary mode\n",
    "    with out.open(mode=\"wb\") as file:\n",
    "        # get request\n",
    "        response = get(url, verify=verify)\n",
    "        for chunk in response.iter_content(100000):\n",
    "            # write to file\n",
    "            file.write(chunk)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:19.201097Z",
     "start_time": "2020-09-19T17:11:19.157942Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 6;\n",
       "                var nbb_unformatted_code = \"def plot_history(history):\\n    loss_list = [s for s in history.history.keys() if \\\"loss\\\" in s and \\\"val\\\" not in s]\\n    val_loss_list = [s for s in history.history.keys() if \\\"loss\\\" in s and \\\"val\\\" in s]\\n    acc_list = [s for s in history.history.keys() if \\\"AUC\\\" in s and \\\"val\\\" not in s]\\n    val_acc_list = [s for s in history.history.keys() if \\\"AUC\\\" in s and \\\"val\\\" in s]\\n\\n    if len(loss_list) == 0:\\n        print(\\\"Loss is missing in history\\\")\\n        return\\n\\n    ## As loss always exists\\n    epochs = range(1, len(history.history[loss_list[0]]) + 1)\\n\\n    ## Loss\\n    plt.figure(1)\\n    for l in loss_list:\\n        plt.plot(\\n            epochs,\\n            history.history[l],\\n            \\\"b\\\",\\n            label=\\\"Training loss (\\\"\\n            + str(str(format(history.history[l][-1], \\\".5f\\\")) + \\\")\\\"),\\n        )\\n    for l in val_loss_list:\\n        plt.plot(\\n            epochs,\\n            history.history[l],\\n            \\\"g\\\",\\n            label=\\\"Validation loss (\\\"\\n            + str(str(format(history.history[l][-1], \\\".5f\\\")) + \\\")\\\"),\\n        )\\n\\n    plt.title(\\\"Loss\\\")\\n    plt.xlabel(\\\"Epochs\\\")\\n    plt.ylabel(\\\"Loss\\\")\\n    plt.legend()\\n\\n    plt.show()\";\n",
       "                var nbb_formatted_code = \"def plot_history(history):\\n    loss_list = [s for s in history.history.keys() if \\\"loss\\\" in s and \\\"val\\\" not in s]\\n    val_loss_list = [s for s in history.history.keys() if \\\"loss\\\" in s and \\\"val\\\" in s]\\n    acc_list = [s for s in history.history.keys() if \\\"AUC\\\" in s and \\\"val\\\" not in s]\\n    val_acc_list = [s for s in history.history.keys() if \\\"AUC\\\" in s and \\\"val\\\" in s]\\n\\n    if len(loss_list) == 0:\\n        print(\\\"Loss is missing in history\\\")\\n        return\\n\\n    ## As loss always exists\\n    epochs = range(1, len(history.history[loss_list[0]]) + 1)\\n\\n    ## Loss\\n    plt.figure(1)\\n    for l in loss_list:\\n        plt.plot(\\n            epochs,\\n            history.history[l],\\n            \\\"b\\\",\\n            label=\\\"Training loss (\\\"\\n            + str(str(format(history.history[l][-1], \\\".5f\\\")) + \\\")\\\"),\\n        )\\n    for l in val_loss_list:\\n        plt.plot(\\n            epochs,\\n            history.history[l],\\n            \\\"g\\\",\\n            label=\\\"Validation loss (\\\"\\n            + str(str(format(history.history[l][-1], \\\".5f\\\")) + \\\")\\\"),\\n        )\\n\\n    plt.title(\\\"Loss\\\")\\n    plt.xlabel(\\\"Epochs\\\")\\n    plt.ylabel(\\\"Loss\\\")\\n    plt.legend()\\n\\n    plt.show()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_history(history):\n",
    "    loss_list = [s for s in history.history.keys() if \"loss\" in s and \"val\" not in s]\n",
    "    val_loss_list = [s for s in history.history.keys() if \"loss\" in s and \"val\" in s]\n",
    "    acc_list = [s for s in history.history.keys() if \"AUC\" in s and \"val\" not in s]\n",
    "    val_acc_list = [s for s in history.history.keys() if \"AUC\" in s and \"val\" in s]\n",
    "\n",
    "    if len(loss_list) == 0:\n",
    "        print(\"Loss is missing in history\")\n",
    "        return\n",
    "\n",
    "    ## As loss always exists\n",
    "    epochs = range(1, len(history.history[loss_list[0]]) + 1)\n",
    "\n",
    "    ## Loss\n",
    "    plt.figure(1)\n",
    "    for l in loss_list:\n",
    "        plt.plot(\n",
    "            epochs,\n",
    "            history.history[l],\n",
    "            \"b\",\n",
    "            label=\"Training loss (\"\n",
    "            + str(str(format(history.history[l][-1], \".5f\")) + \")\"),\n",
    "        )\n",
    "    for l in val_loss_list:\n",
    "        plt.plot(\n",
    "            epochs,\n",
    "            history.history[l],\n",
    "            \"g\",\n",
    "            label=\"Validation loss (\"\n",
    "            + str(str(format(history.history[l][-1], \".5f\")) + \")\"),\n",
    "        )\n",
    "\n",
    "    plt.title(\"Loss\")\n",
    "    plt.xlabel(\"Epochs\")\n",
    "    plt.ylabel(\"Loss\")\n",
    "    plt.legend()\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bank marketing : loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:20.881689Z",
     "start_time": "2020-09-19T17:11:19.202265Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading https://archive.ics.uci.edu/ml/machine-learning-databases/00222/bank-additional.zip at /data/bank-additional.zip ...\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 7;\n",
       "                var nbb_unformatted_code = \"url = \\\"https://archive.ics.uci.edu/ml/machine-learning-databases/00222/bank-additional.zip\\\"\\ndataset_name = \\\"bank-additional-full\\\"\\nout_zip = Path(os.getcwd().rsplit(\\\"/\\\", 1)[0] + \\\"/data/bank-additional.zip\\\")\\nout = Path(\\n    os.getcwd().rsplit(\\\"/\\\", 1)[0] + \\\"/data/bank-additional/\\\" + dataset_name + \\\".csv\\\"\\n)\\n\\ndownload(url, out_zip)\\nwith ZipFile(out_zip, \\\"r\\\") as zipObj:\\n    zipObj.extractall(\\\"/data\\\")\\n\\ntarget = \\\"y\\\"\\nto_remove = []\\ntrain = pd.read_csv(out, sep=\\\";\\\", low_memory=False)\";\n",
       "                var nbb_formatted_code = \"url = \\\"https://archive.ics.uci.edu/ml/machine-learning-databases/00222/bank-additional.zip\\\"\\ndataset_name = \\\"bank-additional-full\\\"\\nout_zip = Path(os.getcwd().rsplit(\\\"/\\\", 1)[0] + \\\"/data/bank-additional.zip\\\")\\nout = Path(\\n    os.getcwd().rsplit(\\\"/\\\", 1)[0] + \\\"/data/bank-additional/\\\" + dataset_name + \\\".csv\\\"\\n)\\n\\ndownload(url, out_zip)\\nwith ZipFile(out_zip, \\\"r\\\") as zipObj:\\n    zipObj.extractall(\\\"/data\\\")\\n\\ntarget = \\\"y\\\"\\nto_remove = []\\ntrain = pd.read_csv(out, sep=\\\";\\\", low_memory=False)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "url = \"https://archive.ics.uci.edu/ml/machine-learning-databases/00222/bank-additional.zip\"\n",
    "dataset_name = \"bank-additional-full\"\n",
    "out_zip = Path(os.getcwd().rsplit(\"/\", 1)[0] + \"/data/bank-additional.zip\")\n",
    "out = Path(\n",
    "    os.getcwd().rsplit(\"/\", 1)[0] + \"/data/bank-additional/\" + dataset_name + \".csv\"\n",
    ")\n",
    "\n",
    "download(url, out_zip)\n",
    "with ZipFile(out_zip, \"r\") as zipObj:\n",
    "    zipObj.extractall(\"/data\")\n",
    "\n",
    "target = \"y\"\n",
    "to_remove = []\n",
    "train = pd.read_csv(out, sep=\";\", low_memory=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:20.906178Z",
     "start_time": "2020-09-19T17:11:20.883318Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(41188, 21)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 8;\n",
       "                var nbb_unformatted_code = \"train.shape\";\n",
       "                var nbb_formatted_code = \"train.shape\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:20.931140Z",
     "start_time": "2020-09-19T17:11:20.907434Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['age', 'job', 'marital', 'education', 'default', 'housing', 'loan',\n",
       "       'contact', 'month', 'day_of_week', 'duration', 'campaign', 'pdays',\n",
       "       'previous', 'poutcome', 'emp.var.rate', 'cons.price.idx',\n",
       "       'cons.conf.idx', 'euribor3m', 'nr.employed', 'y'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 9;\n",
       "                var nbb_unformatted_code = \"train.columns\";\n",
       "                var nbb_formatted_code = \"train.columns\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.022086Z",
     "start_time": "2020-09-19T17:11:20.932087Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Building tailored column\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/work/.cache/poetry/xplainet-KS9Ipbc1-py3.7/lib/python3.7/site-packages/ipykernel_launcher.py:14: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \n",
      "/work/.cache/poetry/xplainet-KS9Ipbc1-py3.7/lib/python3.7/site-packages/pandas/core/indexing.py:670: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  iloc._setitem_with_indexer(indexer, value)\n",
      "/work/.cache/poetry/xplainet-KS9Ipbc1-py3.7/lib/python3.7/site-packages/ipykernel_launcher.py:15: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  from ipykernel import kernelapp as app\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 10;\n",
       "                var nbb_unformatted_code = \"if \\\"Set\\\" not in train.columns:\\n    print(\\\"Building tailored column\\\")\\n    train_valid_index, test_index = next(\\n        StratifiedShuffleSplit(n_splits=1, test_size=0.1, random_state=SEED).split(\\n            range(train[target].shape[0]), train[target].values\\n        )\\n    )\\n    train_index, valid_index = next(\\n        StratifiedShuffleSplit(n_splits=1, test_size=0.1, random_state=SEED).split(\\n            train_valid_index, train[target].values[train_valid_index]\\n        )\\n    )\\n    train[\\\"Set\\\"] = \\\"train\\\"\\n    train[\\\"Set\\\"][valid_index] = \\\"valid\\\"\\n    train[\\\"Set\\\"][test_index] = \\\"test\\\"\\n    # train.to_csv((out.parent / \\\"train_bench.csv\\\").as_posix(), index=False)\";\n",
       "                var nbb_formatted_code = \"if \\\"Set\\\" not in train.columns:\\n    print(\\\"Building tailored column\\\")\\n    train_valid_index, test_index = next(\\n        StratifiedShuffleSplit(n_splits=1, test_size=0.1, random_state=SEED).split(\\n            range(train[target].shape[0]), train[target].values\\n        )\\n    )\\n    train_index, valid_index = next(\\n        StratifiedShuffleSplit(n_splits=1, test_size=0.1, random_state=SEED).split(\\n            train_valid_index, train[target].values[train_valid_index]\\n        )\\n    )\\n    train[\\\"Set\\\"] = \\\"train\\\"\\n    train[\\\"Set\\\"][valid_index] = \\\"valid\\\"\\n    train[\\\"Set\\\"][test_index] = \\\"test\\\"\\n    # train.to_csv((out.parent / \\\"train_bench.csv\\\").as_posix(), index=False)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if \"Set\" not in train.columns:\n",
    "    print(\"Building tailored column\")\n",
    "    train_valid_index, test_index = next(\n",
    "        StratifiedShuffleSplit(n_splits=1, test_size=0.1, random_state=SEED).split(\n",
    "            range(train[target].shape[0]), train[target].values\n",
    "        )\n",
    "    )\n",
    "    train_index, valid_index = next(\n",
    "        StratifiedShuffleSplit(n_splits=1, test_size=0.1, random_state=SEED).split(\n",
    "            train_valid_index, train[target].values[train_valid_index]\n",
    "        )\n",
    "    )\n",
    "    train[\"Set\"] = \"train\"\n",
    "    train[\"Set\"][valid_index] = \"valid\"\n",
    "    train[\"Set\"][test_index] = \"test\"\n",
    "    # train.to_csv((out.parent / \"train_bench.csv\").as_posix(), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.062820Z",
     "start_time": "2020-09-19T17:11:21.023273Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 11;\n",
       "                var nbb_unformatted_code = \"train_indices = train[train.Set == \\\"train\\\"].index\\nvalid_indices = train[train.Set == \\\"valid\\\"].index\\ntest_indices = train[train.Set == \\\"test\\\"].index\";\n",
       "                var nbb_formatted_code = \"train_indices = train[train.Set == \\\"train\\\"].index\\nvalid_indices = train[train.Set == \\\"valid\\\"].index\\ntest_indices = train[train.Set == \\\"test\\\"].index\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_indices = train[train.Set == \"train\"].index\n",
    "valid_indices = train[train.Set == \"valid\"].index\n",
    "test_indices = train[train.Set == \"test\"].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.596221Z",
     "start_time": "2020-09-19T17:11:21.063777Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'constant_cols': [],\n",
       " 'bool_cols': ['contact'],\n",
       " 'num_cols': [],\n",
       " 'cat_cols': ['month',\n",
       "  'pdays',\n",
       "  'duration',\n",
       "  'marital',\n",
       "  'cons.conf.idx',\n",
       "  'age',\n",
       "  'nr.employed',\n",
       "  'education',\n",
       "  'job',\n",
       "  'euribor3m',\n",
       "  'loan',\n",
       "  'emp.var.rate',\n",
       "  'poutcome',\n",
       "  'campaign',\n",
       "  'day_of_week',\n",
       "  'default',\n",
       "  'cons.price.idx',\n",
       "  'housing',\n",
       "  'previous'],\n",
       " 'num_encoder': [],\n",
       " 'bool_encoder': [SafeLabelEncoder()],\n",
       " 'max_nb': 1482,\n",
       " 'cat_encoder': [SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder()]}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 12;\n",
       "                var nbb_unformatted_code = \"input_train, params = preproc_dataset(train.loc[train_indices], target, [\\\"Set\\\"])\\nparams\";\n",
       "                var nbb_formatted_code = \"input_train, params = preproc_dataset(train.loc[train_indices], target, [\\\"Set\\\"])\\nparams\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "input_train, params = preproc_dataset(train.loc[train_indices], target, [\"Set\"])\n",
    "params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.621499Z",
     "start_time": "2020-09-19T17:11:21.597049Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "33764"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 13;\n",
       "                var nbb_unformatted_code = \"len(train_indices)\";\n",
       "                var nbb_formatted_code = \"len(train_indices)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "len(train_indices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.724799Z",
     "start_time": "2020-09-19T17:11:21.622551Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 14;\n",
       "                var nbb_unformatted_code = \"input_valid, _ = preproc_dataset(train.loc[valid_indices], target, [\\\"Set\\\"], params)\\ninput_test, _ = preproc_dataset(train.loc[test_indices], target, [\\\"Set\\\"], params)\";\n",
       "                var nbb_formatted_code = \"input_valid, _ = preproc_dataset(train.loc[valid_indices], target, [\\\"Set\\\"], params)\\ninput_test, _ = preproc_dataset(train.loc[test_indices], target, [\\\"Set\\\"], params)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "input_valid, _ = preproc_dataset(train.loc[valid_indices], target, [\"Set\"], params)\n",
    "input_test, _ = preproc_dataset(train.loc[test_indices], target, [\"Set\"], params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.748489Z",
     "start_time": "2020-09-19T17:11:21.725846Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 15;\n",
       "                var nbb_unformatted_code = \"target_encoder = LabelEncoder()\";\n",
       "                var nbb_formatted_code = \"target_encoder = LabelEncoder()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "target_encoder = LabelEncoder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.782395Z",
     "start_time": "2020-09-19T17:11:21.749455Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 16;\n",
       "                var nbb_unformatted_code = \"train[target] = target_encoder.fit_transform(train[target].values.reshape(-1))\\ny_train = train[target].values[train_indices]\\ny_valid = train[target].values[valid_indices]\\ny_test = train[target].values[test_indices]\";\n",
       "                var nbb_formatted_code = \"train[target] = target_encoder.fit_transform(train[target].values.reshape(-1))\\ny_train = train[target].values[train_indices]\\ny_valid = train[target].values[valid_indices]\\ny_test = train[target].values[test_indices]\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train[target] = target_encoder.fit_transform(train[target].values.reshape(-1))\n",
    "y_train = train[target].values[train_indices]\n",
    "y_valid = train[target].values[valid_indices]\n",
    "y_test = train[target].values[test_indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.812260Z",
     "start_time": "2020-09-19T17:11:21.783287Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'constant_cols': [],\n",
       " 'bool_cols': ['contact'],\n",
       " 'num_cols': [],\n",
       " 'cat_cols': ['month',\n",
       "  'pdays',\n",
       "  'duration',\n",
       "  'marital',\n",
       "  'cons.conf.idx',\n",
       "  'age',\n",
       "  'nr.employed',\n",
       "  'education',\n",
       "  'job',\n",
       "  'euribor3m',\n",
       "  'loan',\n",
       "  'emp.var.rate',\n",
       "  'poutcome',\n",
       "  'campaign',\n",
       "  'day_of_week',\n",
       "  'default',\n",
       "  'cons.price.idx',\n",
       "  'housing',\n",
       "  'previous'],\n",
       " 'num_encoder': [],\n",
       " 'bool_encoder': [SafeLabelEncoder()],\n",
       " 'max_nb': 1482,\n",
       " 'cat_encoder': [SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder(),\n",
       "  SafeLabelEncoder()]}"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 17;\n",
       "                var nbb_unformatted_code = \"params\";\n",
       "                var nbb_formatted_code = \"params\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.926369Z",
     "start_time": "2020-09-19T17:11:21.813368Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/work/.cache/poetry/xplainet-KS9Ipbc1-py3.7/lib/python3.7/site-packages/tensorflow_addons/utils/resource_loader.py:103: UserWarning: You are currently using TensorFlow 2.3.0 and trying to load a custom op (custom_ops/activations/_activation_ops.so).\n",
      "TensorFlow Addons has compiled its custom ops against TensorFlow 2.2.0, and there are no compatibility guarantees between the two versions. \n",
      "This means that you might get segfaults when loading the custom op, or other kind of low-level errors.\n",
      " If you do, do not file an issue on Github. This is a known limitation.\n",
      "\n",
      "It might help you to fallback to pure Python ops with TF_ADDONS_PY_OPS . To do that, see https://github.com/tensorflow/addons#gpucpu-custom-ops \n",
      "\n",
      "You can also change the TensorFlow version installed on your system. You would need a TensorFlow version equal to or above 2.2.0 and strictly below 2.3.0.\n",
      " Note that nightly versions of TensorFlow, as well as non-pip TensorFlow like `conda install tensorflow` or compiled from source are not supported.\n",
      "\n",
      "The last solution is to find the TensorFlow Addons version that has custom ops compatible with the TensorFlow installed on your system. To do that, refer to the readme: https://github.com/tensorflow/addons\n",
      "  UserWarning,\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 18;\n",
       "                var nbb_unformatted_code = \"model = build_model(\\n    params,\\n    lconv_dim=[4],\\n    lconv_num_dim=[8],\\n    emb_size=16,\\n    activation_num_first_layer=None,  # \\\"tanh\\\",\\n)\";\n",
       "                var nbb_formatted_code = \"model = build_model(\\n    params,\\n    lconv_dim=[4],\\n    lconv_num_dim=[8],\\n    emb_size=16,\\n    activation_num_first_layer=None,  # \\\"tanh\\\",\\n)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = build_model(\n",
    "    params,\n",
    "    lconv_dim=[4],\n",
    "    lconv_num_dim=[8],\n",
    "    emb_size=16,\n",
    "    activation_num_first_layer=None,  # \"tanh\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.951028Z",
     "start_time": "2020-09-19T17:11:21.927381Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TensorShape([None, 77])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 19;\n",
       "                var nbb_unformatted_code = \"model.get_layer(\\\"output\\\")._build_input_shape\";\n",
       "                var nbb_formatted_code = \"model.get_layer(\\\"output\\\")._build_input_shape\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.get_layer(\"output\")._build_input_shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:21.978303Z",
     "start_time": "2020-09-19T17:11:21.951948Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"explainable_model\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_cat (InputLayer)          [(None, 19)]         0                                            \n",
      "__________________________________________________________________________________________________\n",
      "large_emb (Embedding)           (None, 19, 16)       23728       input_cat[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "block_cat_0_conv (LocallyConnec (None, 19, 4)        1216        large_emb[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "block_cat_0_activation (Activat (None, 19, 4)        0           block_cat_0_conv[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "input_bool (InputLayer)         [(None, 1)]          0                                            \n",
      "__________________________________________________________________________________________________\n",
      "reshape_cat_output (Reshape)    (None, 76)           0           block_cat_0_activation[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "concatenate (Concatenate)       (None, 77)           0           input_bool[0][0]                 \n",
      "                                                                 reshape_cat_output[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "output (Dense)                  (None, 1)            78          concatenate[0][0]                \n",
      "==================================================================================================\n",
      "Total params: 25,022\n",
      "Trainable params: 25,022\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 20;\n",
       "                var nbb_unformatted_code = \"model.summary()\";\n",
       "                var nbb_formatted_code = \"model.summary()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:22.001003Z",
     "start_time": "2020-09-19T17:11:21.979199Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 21;\n",
       "                var nbb_unformatted_code = \"#!pip install pydot graphviz\";\n",
       "                var nbb_formatted_code = \"#!pip install pydot graphviz\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#!pip install pydot graphviz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:22.119220Z",
     "start_time": "2020-09-19T17:11:22.001997Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 22;\n",
       "                var nbb_unformatted_code = \"plot_model(\\n    model,\\n    # to_file=\\\"model.png\\\",\\n    show_shapes=True,\\n    show_layer_names=True,\\n    rankdir=\\\"TB\\\",\\n    expand_nested=False,\\n    dpi=96,\\n)\";\n",
       "                var nbb_formatted_code = \"plot_model(\\n    model,\\n    # to_file=\\\"model.png\\\",\\n    show_shapes=True,\\n    show_layer_names=True,\\n    rankdir=\\\"TB\\\",\\n    expand_nested=False,\\n    dpi=96,\\n)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_model(\n",
    "    model,\n",
    "    # to_file=\"model.png\",\n",
    "    show_shapes=True,\n",
    "    show_layer_names=True,\n",
    "    rankdir=\"TB\",\n",
    "    expand_nested=False,\n",
    "    dpi=96,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:22.148184Z",
     "start_time": "2020-09-19T17:11:22.122036Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(33764,)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 23;\n",
       "                var nbb_unformatted_code = \"y_train.shape\";\n",
       "                var nbb_formatted_code = \"y_train.shape\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:22.174802Z",
     "start_time": "2020-09-19T17:11:22.149379Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 24;\n",
       "                var nbb_unformatted_code = \"counts = np.unique(y_train, return_counts=True)[1]\\ncounts = counts.sum() / counts\";\n",
       "                var nbb_formatted_code = \"counts = np.unique(y_train, return_counts=True)[1]\\ncounts = counts.sum() / counts\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "counts = np.unique(y_train, return_counts=True)[1]\n",
    "counts = counts.sum() / counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:22.202992Z",
     "start_time": "2020-09-19T17:11:22.176381Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 1.131539260699085, 1: 8.602292993630574}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 25;\n",
       "                var nbb_unformatted_code = \"class_weight = {\\n    0: counts[0],\\n    1: counts[1],\\n}\\nclass_weight\";\n",
       "                var nbb_formatted_code = \"class_weight = {\\n    0: counts[0],\\n    1: counts[1],\\n}\\nclass_weight\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "class_weight = {\n",
    "    0: counts[0],\n",
    "    1: counts[1],\n",
    "}\n",
    "class_weight"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-07-14T10:13:28.824624Z",
     "start_time": "2020-07-14T10:13:28.800384Z"
    }
   },
   "source": [
    "class_weight = {\n",
    "    0: 1,\n",
    "    1: 1,\n",
    "}\n",
    "class_weight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:46.327278Z",
     "start_time": "2020-09-19T17:11:22.203842Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/2000\n",
      "33/33 - 0s - loss: 1.4087 - val_loss: 0.7771\n",
      "Epoch 2/2000\n",
      "33/33 - 0s - loss: 1.4069 - val_loss: 0.7720\n",
      "Epoch 3/2000\n",
      "33/33 - 0s - loss: 1.4043 - val_loss: 0.7644\n",
      "Epoch 4/2000\n",
      "33/33 - 0s - loss: 1.4008 - val_loss: 0.7565\n",
      "Epoch 5/2000\n",
      "33/33 - 0s - loss: 1.3963 - val_loss: 0.7447\n",
      "Epoch 6/2000\n",
      "33/33 - 0s - loss: 1.3902 - val_loss: 0.7325\n",
      "Epoch 7/2000\n",
      "33/33 - 0s - loss: 1.3818 - val_loss: 0.7142\n",
      "Epoch 8/2000\n",
      "33/33 - 0s - loss: 1.3702 - val_loss: 0.6954\n",
      "Epoch 9/2000\n",
      "33/33 - 0s - loss: 1.3547 - val_loss: 0.6678\n",
      "Epoch 10/2000\n",
      "33/33 - 0s - loss: 1.3340 - val_loss: 0.6423\n",
      "Epoch 11/2000\n",
      "33/33 - 0s - loss: 1.3084 - val_loss: 0.6103\n",
      "Epoch 12/2000\n",
      "33/33 - 0s - loss: 1.2770 - val_loss: 0.5837\n",
      "Epoch 13/2000\n",
      "33/33 - 0s - loss: 1.2415 - val_loss: 0.5552\n",
      "Epoch 14/2000\n",
      "33/33 - 0s - loss: 1.2023 - val_loss: 0.5328\n",
      "Epoch 15/2000\n",
      "33/33 - 0s - loss: 1.1632 - val_loss: 0.5132\n",
      "Epoch 16/2000\n",
      "33/33 - 0s - loss: 1.1254 - val_loss: 0.4959\n",
      "Epoch 17/2000\n",
      "33/33 - 0s - loss: 1.0918 - val_loss: 0.4778\n",
      "Epoch 18/2000\n",
      "33/33 - 0s - loss: 1.0613 - val_loss: 0.4641\n",
      "Epoch 19/2000\n",
      "33/33 - 0s - loss: 1.0342 - val_loss: 0.4462\n",
      "Epoch 20/2000\n",
      "33/33 - 0s - loss: 1.0087 - val_loss: 0.4334\n",
      "Epoch 21/2000\n",
      "33/33 - 0s - loss: 0.9847 - val_loss: 0.4193\n",
      "Epoch 22/2000\n",
      "33/33 - 0s - loss: 0.9608 - val_loss: 0.4084\n",
      "Epoch 23/2000\n",
      "33/33 - 0s - loss: 0.9375 - val_loss: 0.3987\n",
      "Epoch 24/2000\n",
      "33/33 - 0s - loss: 0.9138 - val_loss: 0.3903\n",
      "Epoch 25/2000\n",
      "33/33 - 0s - loss: 0.8907 - val_loss: 0.3806\n",
      "Epoch 26/2000\n",
      "33/33 - 0s - loss: 0.8677 - val_loss: 0.3703\n",
      "Epoch 27/2000\n",
      "33/33 - 0s - loss: 0.8455 - val_loss: 0.3630\n",
      "Epoch 28/2000\n",
      "33/33 - 0s - loss: 0.8239 - val_loss: 0.3551\n",
      "Epoch 29/2000\n",
      "33/33 - 0s - loss: 0.8037 - val_loss: 0.3479\n",
      "Epoch 30/2000\n",
      "33/33 - 0s - loss: 0.7843 - val_loss: 0.3415\n",
      "Epoch 31/2000\n",
      "33/33 - 0s - loss: 0.7661 - val_loss: 0.3362\n",
      "Epoch 32/2000\n",
      "33/33 - 0s - loss: 0.7488 - val_loss: 0.3249\n",
      "Epoch 33/2000\n",
      "33/33 - 0s - loss: 0.7327 - val_loss: 0.3293\n",
      "Epoch 34/2000\n",
      "33/33 - 0s - loss: 0.7175 - val_loss: 0.3164\n",
      "Epoch 35/2000\n",
      "33/33 - 0s - loss: 0.7032 - val_loss: 0.3101\n",
      "Epoch 36/2000\n",
      "33/33 - 0s - loss: 0.6900 - val_loss: 0.3072\n",
      "Epoch 37/2000\n",
      "33/33 - 0s - loss: 0.6775 - val_loss: 0.3012\n",
      "Epoch 38/2000\n",
      "33/33 - 0s - loss: 0.6657 - val_loss: 0.2987\n",
      "Epoch 39/2000\n",
      "33/33 - 0s - loss: 0.6548 - val_loss: 0.2953\n",
      "Epoch 40/2000\n",
      "33/33 - 0s - loss: 0.6445 - val_loss: 0.2940\n",
      "Epoch 41/2000\n",
      "33/33 - 0s - loss: 0.6351 - val_loss: 0.2896\n",
      "Epoch 42/2000\n",
      "33/33 - 0s - loss: 0.6260 - val_loss: 0.2820\n",
      "Epoch 43/2000\n",
      "33/33 - 0s - loss: 0.6180 - val_loss: 0.2776\n",
      "Epoch 44/2000\n",
      "33/33 - 0s - loss: 0.6099 - val_loss: 0.2777\n",
      "Epoch 45/2000\n",
      "33/33 - 0s - loss: 0.6026 - val_loss: 0.2771\n",
      "Epoch 46/2000\n",
      "33/33 - 0s - loss: 0.5958 - val_loss: 0.2749\n",
      "Epoch 47/2000\n",
      "33/33 - 0s - loss: 0.5894 - val_loss: 0.2725\n",
      "Epoch 48/2000\n",
      "33/33 - 0s - loss: 0.5837 - val_loss: 0.2716\n",
      "Epoch 49/2000\n",
      "33/33 - 0s - loss: 0.5780 - val_loss: 0.2729\n",
      "Epoch 50/2000\n",
      "33/33 - 0s - loss: 0.5728 - val_loss: 0.2659\n",
      "Epoch 51/2000\n",
      "33/33 - 0s - loss: 0.5679 - val_loss: 0.2656\n",
      "Epoch 52/2000\n",
      "33/33 - 0s - loss: 0.5633 - val_loss: 0.2667\n",
      "Epoch 53/2000\n",
      "33/33 - 0s - loss: 0.5592 - val_loss: 0.2661\n",
      "Epoch 54/2000\n",
      "33/33 - 0s - loss: 0.5550 - val_loss: 0.2638\n",
      "Epoch 55/2000\n",
      "33/33 - 0s - loss: 0.5512 - val_loss: 0.2583\n",
      "Epoch 56/2000\n",
      "33/33 - 0s - loss: 0.5481 - val_loss: 0.2555\n",
      "Epoch 57/2000\n",
      "33/33 - 0s - loss: 0.5444 - val_loss: 0.2596\n",
      "Epoch 58/2000\n",
      "33/33 - 0s - loss: 0.5411 - val_loss: 0.2623\n",
      "Epoch 59/2000\n",
      "33/33 - 0s - loss: 0.5380 - val_loss: 0.2546\n",
      "Epoch 60/2000\n",
      "33/33 - 0s - loss: 0.5353 - val_loss: 0.2581\n",
      "Epoch 61/2000\n",
      "33/33 - 0s - loss: 0.5326 - val_loss: 0.2666\n",
      "Epoch 62/2000\n",
      "33/33 - 0s - loss: 0.5302 - val_loss: 0.2560\n",
      "Epoch 63/2000\n",
      "33/33 - 0s - loss: 0.5277 - val_loss: 0.2599\n",
      "Epoch 64/2000\n",
      "33/33 - 0s - loss: 0.5253 - val_loss: 0.2554\n",
      "Epoch 65/2000\n",
      "33/33 - 0s - loss: 0.5232 - val_loss: 0.2575\n",
      "Epoch 66/2000\n",
      "33/33 - 0s - loss: 0.5212 - val_loss: 0.2559\n",
      "Epoch 67/2000\n",
      "33/33 - 0s - loss: 0.5193 - val_loss: 0.2642\n",
      "Epoch 68/2000\n",
      "33/33 - 0s - loss: 0.5175 - val_loss: 0.2558\n",
      "Epoch 69/2000\n",
      "33/33 - 0s - loss: 0.5154 - val_loss: 0.2566\n",
      "Epoch 70/2000\n",
      "33/33 - 0s - loss: 0.5137 - val_loss: 0.2537\n",
      "Epoch 71/2000\n",
      "33/33 - 0s - loss: 0.5120 - val_loss: 0.2607\n",
      "Epoch 72/2000\n",
      "33/33 - 0s - loss: 0.5104 - val_loss: 0.2535\n",
      "Epoch 73/2000\n",
      "33/33 - 0s - loss: 0.5088 - val_loss: 0.2537\n",
      "Epoch 74/2000\n",
      "33/33 - 0s - loss: 0.5074 - val_loss: 0.2580\n",
      "Epoch 75/2000\n",
      "33/33 - 0s - loss: 0.5060 - val_loss: 0.2566\n",
      "Epoch 76/2000\n",
      "33/33 - 0s - loss: 0.5044 - val_loss: 0.2529\n",
      "Epoch 77/2000\n",
      "33/33 - 0s - loss: 0.5031 - val_loss: 0.2541\n",
      "Epoch 78/2000\n",
      "33/33 - 0s - loss: 0.5018 - val_loss: 0.2565\n",
      "Epoch 79/2000\n",
      "33/33 - 0s - loss: 0.5007 - val_loss: 0.2550\n",
      "Epoch 80/2000\n",
      "33/33 - 0s - loss: 0.4995 - val_loss: 0.2533\n",
      "Epoch 81/2000\n",
      "33/33 - 0s - loss: 0.4983 - val_loss: 0.2521\n",
      "Epoch 82/2000\n",
      "33/33 - 0s - loss: 0.4970 - val_loss: 0.2589\n",
      "Epoch 83/2000\n",
      "33/33 - 0s - loss: 0.4962 - val_loss: 0.2577\n",
      "Epoch 84/2000\n",
      "33/33 - 0s - loss: 0.4951 - val_loss: 0.2549\n",
      "Epoch 85/2000\n",
      "33/33 - 0s - loss: 0.4940 - val_loss: 0.2513\n",
      "Epoch 86/2000\n",
      "33/33 - 0s - loss: 0.4934 - val_loss: 0.2555\n",
      "Epoch 87/2000\n",
      "33/33 - 0s - loss: 0.4922 - val_loss: 0.2587\n",
      "Epoch 88/2000\n",
      "33/33 - 0s - loss: 0.4912 - val_loss: 0.2494\n",
      "Epoch 89/2000\n",
      "33/33 - 0s - loss: 0.4904 - val_loss: 0.2528\n",
      "Epoch 90/2000\n",
      "33/33 - 0s - loss: 0.4894 - val_loss: 0.2551\n",
      "Epoch 91/2000\n",
      "33/33 - 0s - loss: 0.4886 - val_loss: 0.2552\n",
      "Epoch 92/2000\n",
      "33/33 - 0s - loss: 0.4878 - val_loss: 0.2526\n",
      "Epoch 93/2000\n",
      "33/33 - 0s - loss: 0.4869 - val_loss: 0.2514\n",
      "Epoch 94/2000\n",
      "33/33 - 0s - loss: 0.4862 - val_loss: 0.2560\n",
      "Epoch 95/2000\n",
      "33/33 - 0s - loss: 0.4854 - val_loss: 0.2543\n",
      "Epoch 96/2000\n",
      "33/33 - 0s - loss: 0.4847 - val_loss: 0.2540\n",
      "Epoch 97/2000\n",
      "33/33 - 0s - loss: 0.4841 - val_loss: 0.2562\n",
      "Epoch 98/2000\n",
      "33/33 - 0s - loss: 0.4832 - val_loss: 0.2587\n",
      "Epoch 99/2000\n",
      "33/33 - 0s - loss: 0.4828 - val_loss: 0.2571\n",
      "Epoch 100/2000\n",
      "33/33 - 0s - loss: 0.4821 - val_loss: 0.2585\n",
      "Epoch 101/2000\n",
      "33/33 - 0s - loss: 0.4814 - val_loss: 0.2621\n",
      "Epoch 102/2000\n",
      "33/33 - 0s - loss: 0.4807 - val_loss: 0.2605\n",
      "Epoch 103/2000\n",
      "33/33 - 0s - loss: 0.4801 - val_loss: 0.2589\n",
      "Epoch 104/2000\n",
      "33/33 - 0s - loss: 0.4795 - val_loss: 0.2538\n",
      "Epoch 105/2000\n",
      "33/33 - 0s - loss: 0.4788 - val_loss: 0.2591\n",
      "Epoch 106/2000\n",
      "33/33 - 0s - loss: 0.4783 - val_loss: 0.2594\n",
      "Epoch 107/2000\n",
      "33/33 - 0s - loss: 0.4776 - val_loss: 0.2555\n",
      "Epoch 108/2000\n",
      "33/33 - 0s - loss: 0.4770 - val_loss: 0.2609\n",
      "Epoch 00108: early stopping\n",
      "CPU times: user 2min 50s, sys: 4.8 s, total: 2min 55s\n",
      "Wall time: 24.1 s\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 26;\n",
       "                var nbb_unformatted_code = \"%%time\\nhistory = model.fit(\\n    input_train,\\n    y_train.reshape(-1, 1),\\n    epochs=2000,\\n    batch_size=1024,\\n    validation_data=(input_valid, y_valid.reshape(-1, 1),),\\n    verbose=2,\\n    callbacks=[EarlyStopping(monitor=\\\"val_loss\\\", patience=20, verbose=1)],\\n    class_weight=class_weight\\n)\";\n",
       "                var nbb_formatted_code = \"%%time\\nhistory = model.fit(\\n    input_train,\\n    y_train.reshape(-1, 1),\\n    epochs=2000,\\n    batch_size=1024,\\n    validation_data=(input_valid, y_valid.reshape(-1, 1),),\\n    verbose=2,\\n    callbacks=[EarlyStopping(monitor=\\\"val_loss\\\", patience=20, verbose=1)],\\n    class_weight=class_weight\\n)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%time\n",
    "history = model.fit(\n",
    "    input_train,\n",
    "    y_train.reshape(-1, 1),\n",
    "    epochs=2000,\n",
    "    batch_size=1024,\n",
    "    validation_data=(input_valid, y_valid.reshape(-1, 1),),\n",
    "    verbose=2,\n",
    "    callbacks=[EarlyStopping(monitor=\"val_loss\", patience=20, verbose=1)],\n",
    "    class_weight=class_weight\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:46.466045Z",
     "start_time": "2020-09-19T17:11:46.328206Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 27;\n",
       "                var nbb_unformatted_code = \"plot_history(history)\";\n",
       "                var nbb_formatted_code = \"plot_history(history)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_history(history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:46.644197Z",
     "start_time": "2020-09-19T17:11:46.466991Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9228612726565905"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 28;\n",
       "                var nbb_unformatted_code = \"model_auc = roc_auc_score(\\n    y_true=y_valid, y_score=model.predict(input_valid).reshape(-1),\\n)\\nmodel_auc\";\n",
       "                var nbb_formatted_code = \"model_auc = roc_auc_score(\\n    y_true=y_valid,\\n    y_score=model.predict(input_valid).reshape(-1),\\n)\\nmodel_auc\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_auc = roc_auc_score(\n",
    "    y_true=y_valid,\n",
    "    y_score=model.predict(input_valid).reshape(-1),\n",
    ")\n",
    "model_auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:46.744448Z",
     "start_time": "2020-09-19T17:11:46.645166Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9285325958771641"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 29;\n",
       "                var nbb_unformatted_code = \"model_auc = roc_auc_score(y_true=y_test, y_score=model.predict(input_test).reshape(-1),)\\nmodel_auc\";\n",
       "                var nbb_formatted_code = \"model_auc = roc_auc_score(\\n    y_true=y_test,\\n    y_score=model.predict(input_test).reshape(-1),\\n)\\nmodel_auc\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_auc = roc_auc_score(\n",
    "    y_true=y_test,\n",
    "    y_score=model.predict(input_test).reshape(-1),\n",
    ")\n",
    "model_auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:46.768610Z",
     "start_time": "2020-09-19T17:11:46.745343Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 30;\n",
       "                var nbb_unformatted_code = \"from xplainet.model import predict, encode\";\n",
       "                var nbb_formatted_code = \"from xplainet.model import predict, encode\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from xplainet.model import predict, encode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:46.950091Z",
     "start_time": "2020-09-19T17:11:46.769492Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 31;\n",
       "                var nbb_unformatted_code = \"probs, explanations = predict(model, input_test)\";\n",
       "                var nbb_formatted_code = \"probs, explanations = predict(model, input_test)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "probs, explanations = predict(model, input_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.125105Z",
     "start_time": "2020-09-19T17:11:46.951082Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 32;\n",
       "                var nbb_unformatted_code = \"probs, encoded_output = encode(model, input_test)\";\n",
       "                var nbb_formatted_code = \"probs, encoded_output = encode(model, input_test)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "probs, encoded_output = encode(model, input_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.148633Z",
     "start_time": "2020-09-19T17:11:47.126151Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4119,)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 33;\n",
       "                var nbb_unformatted_code = \"y_test.shape\";\n",
       "                var nbb_formatted_code = \"y_test.shape\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.172422Z",
     "start_time": "2020-09-19T17:11:47.149566Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4119, 77)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 34;\n",
       "                var nbb_unformatted_code = \"encoded_output.shape\";\n",
       "                var nbb_formatted_code = \"encoded_output.shape\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "encoded_output.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.197498Z",
     "start_time": "2020-09-19T17:11:47.173373Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4119, 20)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 35;\n",
       "                var nbb_unformatted_code = \"explanations.shape\";\n",
       "                var nbb_formatted_code = \"explanations.shape\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "explanations.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.228708Z",
     "start_time": "2020-09-19T17:11:47.198513Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 36;\n",
       "                var nbb_unformatted_code = \"import matplotlib.pyplot as plt\\n\\n# plt.rcdefaults()\\nimport numpy as np\\nimport matplotlib.pyplot as plt\\n\\n\\ndef explain_plot(importances, columns):\\n    selection = np.argsort(-np.absolute(importances))[:10]\\n    # indexes = np.argsort(importances)\\n    performance = importances[selection]\\n    #     print(performance.shape)512, 256, 128, 64, 32, 1\\n    y_pos = np.arange(performance.shape[0])\\n\\n    plt.barh(y_pos, performance, align=\\\"center\\\", alpha=0.5)\\n    plt.yticks(y_pos, columns[selection])\\n    # plt.xlabel('Usage')\\n    plt.title(\\\"Feature importance\\\")\\n\\n    plt.show()\";\n",
       "                var nbb_formatted_code = \"import matplotlib.pyplot as plt\\n\\n# plt.rcdefaults()\\nimport numpy as np\\nimport matplotlib.pyplot as plt\\n\\n\\ndef explain_plot(importances, columns):\\n    selection = np.argsort(-np.absolute(importances))[:10]\\n    # indexes = np.argsort(importances)\\n    performance = importances[selection]\\n    #     print(performance.shape)512, 256, 128, 64, 32, 1\\n    y_pos = np.arange(performance.shape[0])\\n\\n    plt.barh(y_pos, performance, align=\\\"center\\\", alpha=0.5)\\n    plt.yticks(y_pos, columns[selection])\\n    # plt.xlabel('Usage')\\n    plt.title(\\\"Feature importance\\\")\\n\\n    plt.show()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# plt.rcdefaults()\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "def explain_plot(importances, columns):\n",
    "    selection = np.argsort(-np.absolute(importances))[:10]\n",
    "    # indexes = np.argsort(importances)\n",
    "    performance = importances[selection]\n",
    "    #     print(performance.shape)512, 256, 128, 64, 32, 1\n",
    "    y_pos = np.arange(performance.shape[0])\n",
    "\n",
    "    plt.barh(y_pos, performance, align=\"center\", alpha=0.5)\n",
    "    plt.yticks(y_pos, columns[selection])\n",
    "    # plt.xlabel('Usage')\n",
    "    plt.title(\"Feature importance\")\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.254358Z",
     "start_time": "2020-09-19T17:11:47.229714Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['contact', 'month', 'pdays', 'duration', 'marital',\n",
       "       'cons.conf.idx', 'age', 'nr.employed', 'education', 'job',\n",
       "       'euribor3m', 'loan', 'emp.var.rate', 'poutcome', 'campaign',\n",
       "       'day_of_week', 'default', 'cons.price.idx', 'housing', 'previous'],\n",
       "      dtype='<U14')"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 37;\n",
       "                var nbb_unformatted_code = \"all_cols = np.array(params[\\\"bool_cols\\\"] + params[\\\"num_cols\\\"] + params[\\\"cat_cols\\\"])\\nall_cols\";\n",
       "                var nbb_formatted_code = \"all_cols = np.array(params[\\\"bool_cols\\\"] + params[\\\"num_cols\\\"] + params[\\\"cat_cols\\\"])\\nall_cols\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "all_cols = np.array(params[\"bool_cols\"] + params[\"num_cols\"] + params[\"cat_cols\"])\n",
    "all_cols"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T18:48:54.151671Z",
     "start_time": "2020-09-18T18:48:54.127488Z"
    }
   },
   "source": [
    "## Explain global"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.839638Z",
     "start_time": "2020-09-19T17:11:47.255303Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 38;\n",
       "                var nbb_unformatted_code = \"probs_train, explanations_train = predict(model, input_train)\";\n",
       "                var nbb_formatted_code = \"probs_train, explanations_train = predict(model, input_train)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "probs_train, explanations_train = predict(model, input_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.864894Z",
     "start_time": "2020-09-19T17:11:47.840699Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 39;\n",
       "                var nbb_unformatted_code = \"global_explain = np.sum(explanations_train, axis=0)\\nglobal_explain = global_explain / np.abs(global_explain).sum()\";\n",
       "                var nbb_formatted_code = \"global_explain = np.sum(explanations_train, axis=0)\\nglobal_explain = global_explain / np.abs(global_explain).sum()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "global_explain = np.sum(explanations_train, axis=0)\n",
    "global_explain = global_explain / np.abs(global_explain).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:47.888696Z",
     "start_time": "2020-09-19T17:11:47.865831Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 5.51899192e-03, -1.21002763e-02, -2.84195636e-02, -5.93051374e-01,\n",
       "        2.45436422e-02, -1.50825474e-03, -2.77059989e-04, -1.12226852e-02,\n",
       "       -6.63148540e-04,  4.89590013e-03, -7.27010199e-02,  2.35344764e-02,\n",
       "        1.40137751e-02,  6.68540125e-02,  2.23898893e-02,  3.18636484e-02,\n",
       "        4.10649480e-02, -6.90272925e-03,  3.30870987e-02,  5.38750549e-03])"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 40;\n",
       "                var nbb_unformatted_code = \"global_explain\";\n",
       "                var nbb_formatted_code = \"global_explain\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "global_explain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:48.018462Z",
     "start_time": "2020-09-19T17:11:47.889608Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 41;\n",
       "                var nbb_unformatted_code = \"explain_plot(global_explain, all_cols)\";\n",
       "                var nbb_formatted_code = \"explain_plot(global_explain, all_cols)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "explain_plot(global_explain, all_cols)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Explain local"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:50.185529Z",
     "start_time": "2020-09-19T17:11:48.019519Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.4338725527049974e-05\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0025712251663208008\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.12140670418739319\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.008202970027923584\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7.297842239495367e-05\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.04883086681365967\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.027219325304031372\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.007348418235778809\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0015024244785308838\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0028031766414642334\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0036596059799194336\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.1416583220125176e-05\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.005013465881347656\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.40163514018058777\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.030781149864196777\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.02529105544090271\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.464015930891037\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.32752472162246704\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0065393149852752686\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.006181776523590088\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 42;\n",
       "                var nbb_unformatted_code = \"for i in range(20):\\n    explain_plot(explanations[i], all_cols)\\n    print(probs[i].item())\";\n",
       "                var nbb_formatted_code = \"for i in range(20):\\n    explain_plot(explanations[i], all_cols)\\n    print(probs[i].item())\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(20):\n",
    "    explain_plot(explanations[i], all_cols)\n",
    "    print(probs[i].item())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:50.208716Z",
     "start_time": "2020-09-19T17:11:50.186507Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 43;\n",
       "                var nbb_unformatted_code = \"from sklearn.metrics import confusion_matrix\";\n",
       "                var nbb_formatted_code = \"from sklearn.metrics import confusion_matrix\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:50.233603Z",
     "start_time": "2020-09-19T17:11:50.209952Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1]), array([3655,  464]))"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 44;\n",
       "                var nbb_unformatted_code = \"np.unique(y_test, return_counts=True)\";\n",
       "                var nbb_formatted_code = \"np.unique(y_test, return_counts=True)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "np.unique(y_test, return_counts=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:50.257537Z",
     "start_time": "2020-09-19T17:11:50.234481Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 1, 0, 1])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 45;\n",
       "                var nbb_unformatted_code = \"y_test\";\n",
       "                var nbb_formatted_code = \"y_test\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:50.282652Z",
     "start_time": "2020-09-19T17:11:50.258630Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['no', 'yes'], dtype=object)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 46;\n",
       "                var nbb_unformatted_code = \"target_encoder.classes_\";\n",
       "                var nbb_formatted_code = \"target_encoder.classes_\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "target_encoder.classes_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-19T17:11:50.384699Z",
     "start_time": "2020-09-19T17:11:50.283544Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[3193,  462],\n",
       "       [  72,  392]])"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 47;\n",
       "                var nbb_unformatted_code = \"confusion_matrix(\\n    y_true=y_test,\\n    y_pred=model.predict(input_test).reshape(-1) >= 0.5,\\n    # labels=target_encoder.classes_,\\n)\";\n",
       "                var nbb_formatted_code = \"confusion_matrix(\\n    y_true=y_test,\\n    y_pred=model.predict(input_test).reshape(-1) >= 0.5,\\n    # labels=target_encoder.classes_,\\n)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "confusion_matrix(\n",
    "    y_true=y_test,\n",
    "    y_pred=model.predict(input_test).reshape(-1) >= 0.5,\n",
    "    # labels=target_encoder.classes_,\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}