ventilator_gb_rescaling_eda_V8_GPU.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "language": "python",
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.7.10",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "colab": {
      "name": "ventilator-gb-rescaling-eda-V8-GPU.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/bobbercheng/ventilator-pressure-prediction/blob/master/ventilator_gb_rescaling_eda_V8_GPU.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Yh6Zg50KJQBQ"
      },
      "source": [
        "V1: Keep R/C, improve\n",
        "\n",
        "V2: Add R**2, no improve\n",
        "\n",
        "V3: Update model from https://www.kaggle.com/dlaststark/gb-vpp-pulp-fiction. Not much improve compare V1\n",
        "\n",
        "V4: Use whole db for test Fold-1\n",
        "Fold-1 | OOF Score: 0.15912771348961205\n",
        "\n",
        "V5: Add loading ScaleLayer from saved model\n",
        "\n",
        "V6: Add TransformerEncoder. TransformerEncoder(128, 512, 8, name=\"transformer_layer\")(x). Testing in GPU.\n",
        "\n",
        "V7: Try to scale targe to 0-1, but it slow the train as MASE cannot be scaled.\n",
        "\n",
        "V8: Removed target scale.\n",
        "Epoch 00022: val_loss improved from 0.32912 to 0.30967, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
        "Epoch 23/150\n",
        "127/127 [==============================] - 161s 1s/step - loss: 0.3176 - val_loss: 0.2917\n",
        "\n",
        "Not Good as without TransformerEncoder. Change paramemter back to \n",
        "\n",
        "V8: Change attention result to multiple instead of concate.\n",
        "\n",
        "Loss is smaller than without LayerNormalization, but CV is bigger. It means it overfit the train data.\n",
        "\n",
        "TransformerEncoder(128, 512, 8, 0.3, name=\"transformer_layer\")\n",
        "\n",
        "Epoch 00181: val_loss did not improve from 0.16212\n",
        "Epoch 182/200\n",
        "127/127 [==============================] - 163s 1s/step - loss: 0.0448 - val_loss: 0.1655\n",
        "\n",
        "Epoch 00182: ReduceLROnPlateau reducing learning rate to 5.6313510867767036e-05.\n",
        "\n",
        "Epoch 00182: val_loss did not improve from 0.16212\n",
        "Restoring model weights from the end of the best epoch.\n",
        "Epoch 00182: early stopping\n",
        "Fold-1 | OOF Score: 0.16211654680482343\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uslq2jL49nGk",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5a62b77b-e612-4c68-dd5a-3e52cb3bde1c"
      },
      "source": [
        "# Update pandas version for Colab TPU\n",
        "!pip install pandas==1.3.2"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: pandas==1.3.2 in /usr/local/lib/python3.7/dist-packages (1.3.2)\n",
            "Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas==1.3.2) (2018.9)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas==1.3.2) (2.8.2)\n",
            "Requirement already satisfied: numpy>=1.17.3 in /usr/local/lib/python3.7/dist-packages (from pandas==1.3.2) (1.19.5)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas==1.3.2) (1.15.0)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RBJlpUNI9com"
      },
      "source": [
        "## Import libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iXzLriPA9coq"
      },
      "source": [
        "try:\n",
        "  import google.colab\n",
        "  IN_COLAB = True\n",
        "except:\n",
        "  IN_COLAB = False"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9esF__aG9cor",
        "outputId": "8f314dcb-62de-489e-e2e7-546fcadba777"
      },
      "source": [
        "if IN_COLAB:\n",
        "  from google.colab import drive\n",
        "  drive.mount(\"/gdrive\", force_remount=True)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /gdrive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MZ_f3Z2M9cos",
        "outputId": "845d4778-0731-429a-a719-d9dac65fb59f"
      },
      "source": [
        "if IN_COLAB:\n",
        "  !pip install kaggle\n",
        "  !mkdir /root/.kaggle\n",
        "  !cp /gdrive/MyDrive/ventilator-pressure-prediction/kaggle.json /root/.kaggle\n",
        "  !kaggle competitions download -c ventilator-pressure-prediction\n",
        "  !mkdir -p ../input/ventilator-pressure-prediction\n",
        "  !unzip '*.zip' -d ../input/ventilator-pressure-prediction\n",
        "  !ls ../input/ventilator-pressure-prediction\n",
        "  !mkdir /gdrive/MyDrive/ventilator-pressure-prediction/gb-rescaling-eda-v3-gpu\n",
        "  #Map Google driver folder to local folder model to save model\n",
        "  !ln -s /gdrive/MyDrive/ventilator-pressure-prediction/gb-rescaling-eda-v3-gpu model-gpu\n",
        "else:\n",
        "  !mkdir ./model-gpu"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: kaggle in /usr/local/lib/python3.7/dist-packages (1.5.12)\n",
            "Requirement already satisfied: python-dateutil in /usr/local/lib/python3.7/dist-packages (from kaggle) (2.8.2)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.7/dist-packages (from kaggle) (4.62.3)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.7/dist-packages (from kaggle) (2021.5.30)\n",
            "Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.7/dist-packages (from kaggle) (1.15.0)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.7/dist-packages (from kaggle) (5.0.2)\n",
            "Requirement already satisfied: urllib3 in /usr/local/lib/python3.7/dist-packages (from kaggle) (1.24.3)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from kaggle) (2.23.0)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.7/dist-packages (from python-slugify->kaggle) (1.3)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->kaggle) (3.0.4)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->kaggle) (2.10)\n",
            "mkdir: cannot create directory ‘/root/.kaggle’: File exists\n",
            "Warning: Looks like you're using an outdated API Version, please consider updating (server 1.5.12 / client 1.5.4)\n",
            "test.csv.zip: Skipping, found more recently modified local copy (use --force to force download)\n",
            "train.csv.zip: Skipping, found more recently modified local copy (use --force to force download)\n",
            "sample_submission.csv.zip: Skipping, found more recently modified local copy (use --force to force download)\n",
            "Archive:  train.csv.zip\n",
            "replace ../input/ventilator-pressure-prediction/train.csv? [y]es, [n]o, [A]ll, [N]one, [r]ename: N\n",
            "\n",
            "Archive:  features.zip\n",
            "  inflating: ../input/ventilator-pressure-prediction/train.npy  N\n",
            "N\n",
            "N\n",
            "\n",
            "  inflating: ../input/ventilator-pressure-prediction/test.npy  \n",
            "\n",
            "Archive:  sample_submission.csv.zip\n",
            "\n",
            "Archive:  test.csv.zip\n",
            "\n",
            "4 archives were successfully processed.\n",
            "sample_submission.csv  test.csv  test.npy  train.csv  train.npy\n",
            "mkdir: cannot create directory ‘/gdrive/MyDrive/ventilator-pressure-prediction/gb-rescaling-eda-v3-gpu’: File exists\n",
            "ln: failed to create symbolic link 'model-gpu/gb-rescaling-eda-v3-gpu': File exists\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.status.busy": "2021-10-21T17:30:55.782617Z",
          "iopub.execute_input": "2021-10-21T17:30:55.783998Z",
          "iopub.status.idle": "2021-10-21T17:30:55.793669Z",
          "shell.execute_reply.started": "2021-10-21T17:30:55.783937Z",
          "shell.execute_reply": "2021-10-21T17:30:55.792667Z"
        },
        "trusted": true,
        "id": "1TfjZidC9cos"
      },
      "source": [
        "import gc\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import random\n",
        "import os\n",
        "from sklearn.model_selection import KFold\n",
        "from sklearn.preprocessing import RobustScaler, MinMaxScaler\n",
        "from sklearn.metrics import mean_absolute_error\n",
        "\n",
        "import tensorflow as tf\n",
        "gpus = tf.config.experimental.list_physical_devices('GPU')\n",
        "for gpu in gpus:\n",
        "  tf.config.experimental.set_memory_growth(gpu, True)\n",
        "\n",
        "from tensorflow.keras.utils import plot_model\n",
        "from tensorflow.keras.models import Model, load_model\n",
        "from tensorflow.keras.callbacks import EarlyStopping\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint\n",
        "from tensorflow.keras.callbacks import ReduceLROnPlateau\n",
        "from tensorflow.keras.layers import BatchNormalization\n",
        "from tensorflow.keras.layers import Dense, Dropout, Input\n",
        "from tensorflow.keras.layers import Concatenate, LSTM, GRU\n",
        "from tensorflow.keras.layers import Bidirectional, Multiply\n",
        "\n",
        "\n",
        "\n",
        "# Random Seed Initialize\n",
        "RANDOM_SEED = 42\n",
        "\n",
        "def seed_everything(seed=RANDOM_SEED):\n",
        "    os.environ['PYTHONHASHSEED'] = str(seed)\n",
        "    np.random.seed(seed)\n",
        "    random.seed(seed)\n",
        "    tf.random.set_seed(seed)\n",
        "    \n",
        "seed_everything()"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XycyrO7g9cot"
      },
      "source": [
        "## Load datasets"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.status.busy": "2021-10-21T17:24:55.352925Z",
          "iopub.execute_input": "2021-10-21T17:24:55.353455Z",
          "iopub.status.idle": "2021-10-21T17:25:06.101910Z",
          "shell.execute_reply.started": "2021-10-21T17:24:55.353418Z",
          "shell.execute_reply": "2021-10-21T17:25:06.100969Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 220
        },
        "id": "p_vUMe_39cot",
        "outputId": "11147c0c-8c38-423a-e9af-1b73d8ed317b"
      },
      "source": [
        "train_df = pd.read_csv('../input/ventilator-pressure-prediction/train.csv')\n",
        "print(f\"train_df: {train_df.shape}\")\n",
        "train_df.head()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "train_df: (6036000, 8)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>breath_id</th>\n",
              "      <th>R</th>\n",
              "      <th>C</th>\n",
              "      <th>time_step</th>\n",
              "      <th>u_in</th>\n",
              "      <th>u_out</th>\n",
              "      <th>pressure</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0</td>\n",
              "      <td>5.837492</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.033652</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0</td>\n",
              "      <td>5.907794</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.067514</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0</td>\n",
              "      <td>7.876254</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.101542</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>0</td>\n",
              "      <td>11.742872</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.135756</td>\n",
              "      <td>25.355850</td>\n",
              "      <td>0</td>\n",
              "      <td>12.234987</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   id  breath_id   R   C  time_step       u_in  u_out   pressure\n",
              "0   1          1  20  50   0.000000   0.083334      0   5.837492\n",
              "1   2          1  20  50   0.033652  18.383041      0   5.907794\n",
              "2   3          1  20  50   0.067514  22.509278      0   7.876254\n",
              "3   4          1  20  50   0.101542  22.808822      0  11.742872\n",
              "4   5          1  20  50   0.135756  25.355850      0  12.234987"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.status.busy": "2021-10-21T17:25:11.213742Z",
          "iopub.execute_input": "2021-10-21T17:25:11.214675Z",
          "iopub.status.idle": "2021-10-21T17:25:11.297152Z",
          "shell.execute_reply.started": "2021-10-21T17:25:11.214617Z",
          "shell.execute_reply": "2021-10-21T17:25:11.295985Z"
        },
        "trusted": true,
        "id": "R5oBMjYJ9cov"
      },
      "source": [
        "all_pressure = np.sort(train_df['pressure'].unique())\n",
        "pressure_min =  all_pressure[0].item()\n",
        "pressure_max = all_pressure[-1].item()\n",
        "pressure_step = (all_pressure[1] - all_pressure[0]).item()\n",
        "\n",
        "pressure_min,pressure_max,pressure_step"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "80hIcGg09cou"
      },
      "source": [
        "DEBUG = False\n",
        "\n",
        "if DEBUG:\n",
        "    train_df = train_df[:80*10000]"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.status.busy": "2021-10-21T17:25:06.103441Z",
          "iopub.execute_input": "2021-10-21T17:25:06.104426Z",
          "iopub.status.idle": "2021-10-21T17:25:11.211412Z",
          "shell.execute_reply.started": "2021-10-21T17:25:06.104376Z",
          "shell.execute_reply": "2021-10-21T17:25:11.209506Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 220
        },
        "id": "y-0XhQ3a9cov",
        "outputId": "824aeabc-2428-46d5-fd1d-4a75638c66c5"
      },
      "source": [
        "test_df = pd.read_csv('../input/ventilator-pressure-prediction/test.csv')\n",
        "print(f\"test_df: {test_df.shape}\")\n",
        "test_df.head()"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "test_df: (4024000, 7)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>breath_id</th>\n",
              "      <th>R</th>\n",
              "      <th>C</th>\n",
              "      <th>time_step</th>\n",
              "      <th>u_in</th>\n",
              "      <th>u_out</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>20</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>20</td>\n",
              "      <td>0.031904</td>\n",
              "      <td>7.515046</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>20</td>\n",
              "      <td>0.063827</td>\n",
              "      <td>14.651675</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>20</td>\n",
              "      <td>0.095751</td>\n",
              "      <td>21.230610</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>20</td>\n",
              "      <td>0.127644</td>\n",
              "      <td>26.320956</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   id  breath_id  R   C  time_step       u_in  u_out\n",
              "0   1          0  5  20   0.000000   0.000000      0\n",
              "1   2          0  5  20   0.031904   7.515046      0\n",
              "2   3          0  5  20   0.063827  14.651675      0\n",
              "3   4          0  5  20   0.095751  21.230610      0\n",
              "4   5          0  5  20   0.127644  26.320956      0"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IVPWgQDL9cow"
      },
      "source": [
        "## Feature Engineering"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.status.busy": "2021-10-21T17:25:11.298767Z",
          "iopub.execute_input": "2021-10-21T17:25:11.299164Z",
          "iopub.status.idle": "2021-10-21T17:27:48.006762Z",
          "shell.execute_reply.started": "2021-10-21T17:25:11.299125Z",
          "shell.execute_reply": "2021-10-21T17:27:48.005658Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1NHNQSZp9cow",
        "outputId": "ef4ea428-902a-4060-bea8-e1fd7dee094b"
      },
      "source": [
        "def add_features(df):\n",
        "    df['cross']= df['u_in'] * df['u_out']\n",
        "    df['cross2']= df['time_step'] * df['u_out']\n",
        "    df['area'] = df['time_step'] * df['u_in']\n",
        "    df['area'] = df.groupby('breath_id')['area'].cumsum()\n",
        "    df['time_step_cumsum'] = df.groupby(['breath_id'])['time_step'].cumsum()\n",
        "    df['u_in_cumsum'] = (df['u_in']).groupby(df['breath_id']).cumsum()\n",
        "    print(\"Step-1...Completed\")\n",
        "    \n",
        "    df['u_in_lag1'] = df.groupby('breath_id')['u_in'].shift(1)\n",
        "    df['u_out_lag1'] = df.groupby('breath_id')['u_out'].shift(1)\n",
        "    df['u_in_lag_back1'] = df.groupby('breath_id')['u_in'].shift(-1)\n",
        "    df['u_out_lag_back1'] = df.groupby('breath_id')['u_out'].shift(-1)\n",
        "    df['u_in_lag2'] = df.groupby('breath_id')['u_in'].shift(2)\n",
        "    df['u_out_lag2'] = df.groupby('breath_id')['u_out'].shift(2)\n",
        "    df['u_in_lag_back2'] = df.groupby('breath_id')['u_in'].shift(-2)\n",
        "    df['u_out_lag_back2'] = df.groupby('breath_id')['u_out'].shift(-2)\n",
        "    df['u_in_lag3'] = df.groupby('breath_id')['u_in'].shift(3)\n",
        "    df['u_out_lag3'] = df.groupby('breath_id')['u_out'].shift(3)\n",
        "    df['u_in_lag_back3'] = df.groupby('breath_id')['u_in'].shift(-3)\n",
        "    df['u_out_lag_back3'] = df.groupby('breath_id')['u_out'].shift(-3)\n",
        "    df['u_in_lag4'] = df.groupby('breath_id')['u_in'].shift(4)\n",
        "    df['u_out_lag4'] = df.groupby('breath_id')['u_out'].shift(4)\n",
        "    df['u_in_lag_back4'] = df.groupby('breath_id')['u_in'].shift(-4)\n",
        "    df['u_out_lag_back4'] = df.groupby('breath_id')['u_out'].shift(-4)\n",
        "    df = df.fillna(0)\n",
        "    print(\"Step-2...Completed\")\n",
        "    \n",
        "    df['breath_id__u_in__max'] = df.groupby(['breath_id'])['u_in'].transform('max')\n",
        "    df['breath_id__u_out__max'] = df.groupby(['breath_id'])['u_out'].transform('max')\n",
        "    df['breath_id__u_in__diffmax'] = df.groupby(['breath_id'])['u_in'].transform('max') - df['u_in']\n",
        "    df['breath_id__u_in__diffmean'] = df.groupby(['breath_id'])['u_in'].transform('mean') - df['u_in']\n",
        "    df['breath_id__u_in__diffmax'] = df.groupby(['breath_id'])['u_in'].transform('max') - df['u_in']\n",
        "    df['breath_id__u_in__diffmean'] = df.groupby(['breath_id'])['u_in'].transform('mean') - df['u_in']\n",
        "    print(\"Step-3...Completed\")\n",
        "    \n",
        "    df['u_in_diff1'] = df['u_in'] - df['u_in_lag1']\n",
        "    df['u_out_diff1'] = df['u_out'] - df['u_out_lag1']\n",
        "    df['u_in_diff2'] = df['u_in'] - df['u_in_lag2']\n",
        "    df['u_out_diff2'] = df['u_out'] - df['u_out_lag2']\n",
        "    df['u_in_diff3'] = df['u_in'] - df['u_in_lag3']\n",
        "    df['u_out_diff3'] = df['u_out'] - df['u_out_lag3']\n",
        "    df['u_in_diff4'] = df['u_in'] - df['u_in_lag4']\n",
        "    df['u_out_diff4'] = df['u_out'] - df['u_out_lag4']\n",
        "    print(\"Step-4...Completed\")\n",
        "    \n",
        "    df['one'] = 1\n",
        "    df['count'] = (df['one']).groupby(df['breath_id']).cumsum()\n",
        "    df['u_in_cummean'] =df['u_in_cumsum'] /df['count']\n",
        "    \n",
        "    df['breath_id_lag']=df['breath_id'].shift(1).fillna(0)\n",
        "    df['breath_id_lag2']=df['breath_id'].shift(2).fillna(0)\n",
        "    df['breath_id_lagsame']=np.select([df['breath_id_lag']==df['breath_id']],[1],0)\n",
        "    df['breath_id_lag2same']=np.select([df['breath_id_lag2']==df['breath_id']],[1],0)\n",
        "    df['breath_id__u_in_lag'] = df['u_in'].shift(1).fillna(0)\n",
        "    df['breath_id__u_in_lag'] = df['breath_id__u_in_lag'] * df['breath_id_lagsame']\n",
        "    df['breath_id__u_in_lag2'] = df['u_in'].shift(2).fillna(0)\n",
        "    df['breath_id__u_in_lag2'] = df['breath_id__u_in_lag2'] * df['breath_id_lag2same']\n",
        "    print(\"Step-5...Completed\")\n",
        "    \n",
        "    df['time_step_diff'] = df.groupby('breath_id')['time_step'].diff().fillna(0)\n",
        "    df['ewm_u_in_mean'] = (df\\\n",
        "                           .groupby('breath_id')['u_in']\\\n",
        "                           .ewm(halflife=9)\\\n",
        "                           .mean()\\\n",
        "                           .reset_index(level=0,drop=True))\n",
        "    df[[\"15_in_sum\",\"15_in_min\",\"15_in_max\",\"15_in_mean\"]] = (df\\\n",
        "                                                              .groupby('breath_id')['u_in']\\\n",
        "                                                              .rolling(window=15,min_periods=1)\\\n",
        "                                                              .agg({\"15_in_sum\":\"sum\",\n",
        "                                                                    \"15_in_min\":\"min\",\n",
        "                                                                    \"15_in_max\":\"max\",\n",
        "                                                                    \"15_in_mean\":\"mean\"\n",
        "                                                                    #\"15_in_std\":\"std\"\n",
        "                                                               })\\\n",
        "                                                               .reset_index(level=0,drop=True))\n",
        "    print(\"Step-6...Completed\")\n",
        "    \n",
        "    #df['u_in_diff_1_2'] = df['u_in_lag1'] - df['u_in_lag2']\n",
        "    #df['u_out_diff_1_2'] = df['u_out_lag1'] - df['u_out_lag2']\n",
        "    #df['u_in_lagback_diff_1_2'] = df['u_in_lag_back1'] - df['u_in_lag_back2']\n",
        "    #df['u_out_lagback_diff_1_2'] = df['u_out_lag_back1'] - df['u_out_lag_back2']\n",
        "    \n",
        "    df['u_in_lagback_diff1'] = df['u_in'] - df['u_in_lag_back1']\n",
        "    df['u_out_lagback_diff1'] = df['u_out'] - df['u_out_lag_back1']\n",
        "    df['u_in_lagback_diff2'] = df['u_in'] - df['u_in_lag_back2']\n",
        "    df['u_out_lagback_diff2'] = df['u_out'] - df['u_out_lag_back2']\n",
        "    print(\"Step-7...Completed\")\n",
        "    \n",
        "    df['R_cat'] = df['R'].astype(str)\n",
        "    df['C_cat'] = df['C'].astype(str)\n",
        "    df['R__C'] = df[\"R\"].astype(str) + '__' + df[\"C\"].astype(str)\n",
        "    df = pd.get_dummies(df)\n",
        "    print(\"Step-8...Completed\")\n",
        "    \n",
        "    return df\n",
        "\n",
        "\n",
        "print(\"Train data...\\n\")\n",
        "train = add_features(train_df)\n",
        "\n",
        "print(\"\\nTest data...\\n\")\n",
        "test = add_features(test_df)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train data...\n",
            "\n",
            "Step-1...Completed\n",
            "Step-2...Completed\n",
            "Step-3...Completed\n",
            "Step-4...Completed\n",
            "Step-5...Completed\n",
            "Step-6...Completed\n",
            "Step-7...Completed\n",
            "Step-8...Completed\n",
            "\n",
            "Test data...\n",
            "\n",
            "Step-1...Completed\n",
            "Step-2...Completed\n",
            "Step-3...Completed\n",
            "Step-4...Completed\n",
            "Step-5...Completed\n",
            "Step-6...Completed\n",
            "Step-7...Completed\n",
            "Step-8...Completed\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.status.busy": "2021-10-21T17:28:07.762733Z",
          "iopub.execute_input": "2021-10-21T17:28:07.763099Z",
          "iopub.status.idle": "2021-10-21T17:28:07.795291Z",
          "shell.execute_reply.started": "2021-10-21T17:28:07.763065Z",
          "shell.execute_reply": "2021-10-21T17:28:07.794126Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 223
        },
        "id": "PgRvxlmB9co0",
        "outputId": "6db9caf1-04a4-44fa-cab6-0cc437fef346"
      },
      "source": [
        "train.head()"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>breath_id</th>\n",
              "      <th>R</th>\n",
              "      <th>C</th>\n",
              "      <th>time_step</th>\n",
              "      <th>u_in</th>\n",
              "      <th>u_out</th>\n",
              "      <th>pressure</th>\n",
              "      <th>cross</th>\n",
              "      <th>cross2</th>\n",
              "      <th>area</th>\n",
              "      <th>time_step_cumsum</th>\n",
              "      <th>u_in_cumsum</th>\n",
              "      <th>u_in_lag1</th>\n",
              "      <th>u_out_lag1</th>\n",
              "      <th>u_in_lag_back1</th>\n",
              "      <th>u_out_lag_back1</th>\n",
              "      <th>u_in_lag2</th>\n",
              "      <th>u_out_lag2</th>\n",
              "      <th>u_in_lag_back2</th>\n",
              "      <th>u_out_lag_back2</th>\n",
              "      <th>u_in_lag3</th>\n",
              "      <th>u_out_lag3</th>\n",
              "      <th>u_in_lag_back3</th>\n",
              "      <th>u_out_lag_back3</th>\n",
              "      <th>u_in_lag4</th>\n",
              "      <th>u_out_lag4</th>\n",
              "      <th>u_in_lag_back4</th>\n",
              "      <th>u_out_lag_back4</th>\n",
              "      <th>breath_id__u_in__max</th>\n",
              "      <th>breath_id__u_out__max</th>\n",
              "      <th>breath_id__u_in__diffmax</th>\n",
              "      <th>breath_id__u_in__diffmean</th>\n",
              "      <th>u_in_diff1</th>\n",
              "      <th>u_out_diff1</th>\n",
              "      <th>u_in_diff2</th>\n",
              "      <th>u_out_diff2</th>\n",
              "      <th>u_in_diff3</th>\n",
              "      <th>u_out_diff3</th>\n",
              "      <th>u_in_diff4</th>\n",
              "      <th>u_out_diff4</th>\n",
              "      <th>one</th>\n",
              "      <th>count</th>\n",
              "      <th>u_in_cummean</th>\n",
              "      <th>breath_id_lag</th>\n",
              "      <th>breath_id_lag2</th>\n",
              "      <th>breath_id_lagsame</th>\n",
              "      <th>breath_id_lag2same</th>\n",
              "      <th>breath_id__u_in_lag</th>\n",
              "      <th>breath_id__u_in_lag2</th>\n",
              "      <th>time_step_diff</th>\n",
              "      <th>ewm_u_in_mean</th>\n",
              "      <th>15_in_sum</th>\n",
              "      <th>15_in_min</th>\n",
              "      <th>15_in_max</th>\n",
              "      <th>15_in_mean</th>\n",
              "      <th>u_in_lagback_diff1</th>\n",
              "      <th>u_out_lagback_diff1</th>\n",
              "      <th>u_in_lagback_diff2</th>\n",
              "      <th>u_out_lagback_diff2</th>\n",
              "      <th>R_cat_20</th>\n",
              "      <th>R_cat_5</th>\n",
              "      <th>R_cat_50</th>\n",
              "      <th>C_cat_10</th>\n",
              "      <th>C_cat_20</th>\n",
              "      <th>C_cat_50</th>\n",
              "      <th>R__C_20__10</th>\n",
              "      <th>R__C_20__20</th>\n",
              "      <th>R__C_20__50</th>\n",
              "      <th>R__C_50__10</th>\n",
              "      <th>R__C_50__20</th>\n",
              "      <th>R__C_50__50</th>\n",
              "      <th>R__C_5__10</th>\n",
              "      <th>R__C_5__20</th>\n",
              "      <th>R__C_5__50</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0</td>\n",
              "      <td>5.837492</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.355850</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.313036</td>\n",
              "      <td>1</td>\n",
              "      <td>28.229702</td>\n",
              "      <td>10.062673</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>-18.299707</td>\n",
              "      <td>0.0</td>\n",
              "      <td>-22.425944</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.033652</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0</td>\n",
              "      <td>5.907794</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.618632</td>\n",
              "      <td>0.033652</td>\n",
              "      <td>18.466375</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.355850</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.259866</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.313036</td>\n",
              "      <td>1</td>\n",
              "      <td>9.929994</td>\n",
              "      <td>-8.237035</td>\n",
              "      <td>18.299707</td>\n",
              "      <td>0.0</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.0</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.0</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>9.233188</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.033652</td>\n",
              "      <td>9.585358</td>\n",
              "      <td>18.466375</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>9.233188</td>\n",
              "      <td>-4.126236</td>\n",
              "      <td>0.0</td>\n",
              "      <td>-4.425781</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.067514</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0</td>\n",
              "      <td>7.876254</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.138333</td>\n",
              "      <td>0.101167</td>\n",
              "      <td>40.975653</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.355850</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.259866</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.127486</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.313036</td>\n",
              "      <td>1</td>\n",
              "      <td>5.803758</td>\n",
              "      <td>-12.363271</td>\n",
              "      <td>4.126236</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.425944</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>13.658551</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.033862</td>\n",
              "      <td>14.229040</td>\n",
              "      <td>40.975653</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>13.658551</td>\n",
              "      <td>-0.299544</td>\n",
              "      <td>0.0</td>\n",
              "      <td>-2.846573</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.101542</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>0</td>\n",
              "      <td>11.742872</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>4.454391</td>\n",
              "      <td>0.202709</td>\n",
              "      <td>63.784476</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.355850</td>\n",
              "      <td>0.0</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.259866</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.127486</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>26.807732</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.313036</td>\n",
              "      <td>1</td>\n",
              "      <td>5.504214</td>\n",
              "      <td>-12.662816</td>\n",
              "      <td>0.299544</td>\n",
              "      <td>0.0</td>\n",
              "      <td>4.425781</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.725488</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>15.946119</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.034028</td>\n",
              "      <td>16.627759</td>\n",
              "      <td>63.784476</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>15.946119</td>\n",
              "      <td>-2.547028</td>\n",
              "      <td>0.0</td>\n",
              "      <td>-4.451044</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>50</td>\n",
              "      <td>0.135756</td>\n",
              "      <td>25.355850</td>\n",
              "      <td>0</td>\n",
              "      <td>12.234987</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.896588</td>\n",
              "      <td>0.338464</td>\n",
              "      <td>89.140326</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.259866</td>\n",
              "      <td>0.0</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.127486</td>\n",
              "      <td>0.0</td>\n",
              "      <td>18.383041</td>\n",
              "      <td>0.0</td>\n",
              "      <td>26.807732</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>0.0</td>\n",
              "      <td>27.864715</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.313036</td>\n",
              "      <td>1</td>\n",
              "      <td>2.957185</td>\n",
              "      <td>-15.209844</td>\n",
              "      <td>2.547028</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.846573</td>\n",
              "      <td>0.0</td>\n",
              "      <td>6.972809</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.272516</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>17.828065</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>22.808822</td>\n",
              "      <td>22.509278</td>\n",
              "      <td>0.034213</td>\n",
              "      <td>18.652046</td>\n",
              "      <td>89.140326</td>\n",
              "      <td>0.083334</td>\n",
              "      <td>25.355850</td>\n",
              "      <td>17.828065</td>\n",
              "      <td>-1.904016</td>\n",
              "      <td>0.0</td>\n",
              "      <td>-1.771635</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   id  breath_id   R   C  ...  R__C_50__50  R__C_5__10  R__C_5__20  R__C_5__50\n",
              "0   1          1  20  50  ...            0           0           0           0\n",
              "1   2          1  20  50  ...            0           0           0           0\n",
              "2   3          1  20  50  ...            0           0           0           0\n",
              "3   4          1  20  50  ...            0           0           0           0\n",
              "4   5          1  20  50  ...            0           0           0           0\n",
              "\n",
              "[5 rows x 75 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zcyv-iig9co0",
        "outputId": "658af9aa-5216-484f-c277-72a364a7511a"
      },
      "source": [
        "del train_df\n",
        "del test_df\n",
        "gc.collect()"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "50"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qip05b0r9co0",
        "outputId": "7b4f12c1-5dcb-4ce2-d230-440f94ad0323"
      },
      "source": [
        "targets = train[['pressure']].to_numpy().reshape(-1, 80)\n",
        "\n",
        "train.drop(['pressure','id', 'breath_id','one','count',\n",
        "            'breath_id_lag','breath_id_lag2','breath_id_lagsame',\n",
        "            'breath_id_lag2same'], axis=1, inplace=True)\n",
        "\n",
        "test = test.drop(['id', 'breath_id','one','count','breath_id_lag',\n",
        "                  'breath_id_lag2','breath_id_lagsame',\n",
        "                  'breath_id_lag2same'], axis=1)\n",
        "\n",
        "print(f\"train: {train.shape} \\ntest: {test.shape}\")"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "train: (6036000, 66) \n",
            "test: (4024000, 66)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OYWzQtL49co1",
        "outputId": "94c211d5-155d-4b03-df5b-521ec4ed9d91"
      },
      "source": [
        "scaler = RobustScaler()\n",
        "train = scaler.fit_transform(train)\n",
        "test = scaler.transform(test)\n",
        "\n",
        "train = train.reshape(-1, 80, train.shape[-1])\n",
        "test = test.reshape(-1, 80, train.shape[-1])\n",
        "\n",
        "print(f\"train: {train.shape} \\ntest: {test.shape} \\ntargets: {targets.shape}\")"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "train: (75450, 80, 66) \n",
            "test: (50300, 80, 66) \n",
            "targets: (75450, 80)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2AQScob_Bg43"
      },
      "source": [
        "# np.save('train.npy', train)\n",
        "# np.save('test.npy', test)\n",
        "# !zip features.zip train.npy test.npy"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Osu14DeV9co1"
      },
      "source": [
        "## Hardware config"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CisxazED9co1",
        "outputId": "5873d72d-6193-43c6-86ee-fff69f911240"
      },
      "source": [
        "try:\n",
        "    tpu = tf.distribute.cluster_resolver.TPUClusterResolver()\n",
        "    tf.config.experimental_connect_to_cluster(tpu)\n",
        "    tf.tpu.experimental.initialize_tpu_system(tpu)\n",
        "    tpu_strategy = tf.distribute.experimental.TPUStrategy(tpu)\n",
        "    BATCH_SIZE = tpu_strategy.num_replicas_in_sync * 64\n",
        "    print(\"Running on TPU:\", tpu.master())\n",
        "    print(f\"Batch Size: {BATCH_SIZE}\")\n",
        "    \n",
        "except ValueError:\n",
        "    tpu_strategy = tf.distribute.get_strategy()\n",
        "    BATCH_SIZE = 512\n",
        "    print(f\"Running on {tpu_strategy.num_replicas_in_sync} replicas\")\n",
        "    print(f\"Batch Size: {BATCH_SIZE}\")"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Running on 1 replicas\n",
            "Batch Size: 512\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GAZMw10A9co1"
      },
      "source": [
        "## Rescaling layer"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qLou5BDt9co2"
      },
      "source": [
        "@tf.custom_gradient\n",
        "def round_with_gradients(x):\n",
        "    def grad(dy):\n",
        "        return dy\n",
        "    return tf.round(x), grad\n",
        "\n",
        "class ScaleLayer(tf.keras.layers.Layer):\n",
        "    def __init__(self, **kwargs):\n",
        "        super(ScaleLayer, self).__init__(**kwargs)\n",
        "        self.min = tf.constant(pressure_min, dtype=np.float32)\n",
        "        self.max = tf.constant(pressure_max, dtype=np.float32)\n",
        "        self.step = tf.constant(pressure_step, dtype=np.float32)\n",
        "\n",
        "    def call(self, inputs):\n",
        "        steps = tf.math.divide(tf.math.add(inputs, -self.min), self.step)\n",
        "        int_steps = round_with_gradients(steps)\n",
        "        rescaled_steps = tf.math.add(tf.math.multiply(int_steps, self.step), self.min)\n",
        "        clipped = tf.clip_by_value(rescaled_steps, self.min, self.max)\n",
        "        return clipped\n",
        "    # def get_config(self):\n",
        "    #     return super(ScaleLayer, self).get_config()\n"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PfmPPh-U2EUY"
      },
      "source": [
        "class TransformerEncoder(tf.keras.layers.Layer):\n",
        "    def __init__(self, embed_dim, dense_dim, num_heads, dropout=0.3, **kwargs):\n",
        "        super().__init__(**kwargs)\n",
        "        self.embed_dim = embed_dim\n",
        "        self.dense_dim = dense_dim\n",
        "        self.num_heads = num_heads\n",
        "        self.attention_dropout = dropout\n",
        "        self.attention = tf.keras.layers.MultiHeadAttention(\n",
        "            num_heads=num_heads, key_dim=embed_dim, dropout=0.3\n",
        "        )\n",
        "        self.dense_proj = tf.keras.Sequential(\n",
        "            [tf.keras.layers.Dense(dense_dim, activation=tf.nn.gelu), tf.keras.layers.Dense(embed_dim),]\n",
        "        )\n",
        "        self.layernorm_1 = tf.keras.layers.LayerNormalization()\n",
        "        self.layernorm_2 = tf.keras.layers.LayerNormalization()\n",
        "\n",
        "    def call(self, inputs, mask=None):\n",
        "        if mask is not None:\n",
        "            mask = mask[:, tf.newaxis, :]\n",
        "\n",
        "        attention_output = self.attention(inputs, inputs, attention_mask=mask)\n",
        "        proj_input = self.layernorm_1(inputs + attention_output)\n",
        "        proj_output = self.dense_proj(proj_input)\n",
        "        return self.layernorm_2(proj_input + proj_output)\n",
        "    def get_config(self):\n",
        "        config = super(TransformerEncoder, self).get_config()\n",
        "        config.update({\"embed_dim\": self.embed_dim,\n",
        "            \"dense_dim\": self.dense_dim,\n",
        "            \"num_heads\": self.num_heads})\n",
        "        return config"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5Ft0JBPC9co2"
      },
      "source": [
        "## Keras DNN Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_HuMFGY39co2"
      },
      "source": [
        "def dnn_model():\n",
        "    \n",
        "    x_input = Input(shape=(train.shape[-2:]))\n",
        "    \n",
        "    x1 = Bidirectional(LSTM(units=768, return_sequences=True))(x_input)\n",
        "    x2 = Bidirectional(LSTM(units=512, return_sequences=True))(x1)\n",
        "    x3 = Bidirectional(LSTM(units=384, return_sequences=True))(x2)\n",
        "    x4 = Bidirectional(LSTM(units=256, return_sequences=True))(x3)\n",
        "    x5 = Bidirectional(LSTM(units=128, return_sequences=True))(x4)\n",
        "    \n",
        "    z2 = Bidirectional(GRU(units=384, return_sequences=True))(x2)\n",
        "    \n",
        "    z31 = Multiply()([x3, z2])\n",
        "    z31 = BatchNormalization()(z31)\n",
        "    z3 = Bidirectional(GRU(units=256, return_sequences=True))(z31)\n",
        "    \n",
        "    z41 = Multiply()([x4, z3])\n",
        "    z41 = BatchNormalization()(z41)\n",
        "    z4 = Bidirectional(GRU(units=128, return_sequences=True))(z41)\n",
        "    \n",
        "    z51 = Multiply()([x5, z4])\n",
        "    z51 = BatchNormalization()(z51)\n",
        "    z5 = Bidirectional(GRU(units=64, return_sequences=True))(z51)\n",
        "    \n",
        "    x = Concatenate(axis=2)([x5, z2, z3, z4, z5])\n",
        "\n",
        "    x = Dense(units=128)(x)\n",
        "\n",
        "    x_transformer = BatchNormalization()(x)\n",
        "    x_transformer = TransformerEncoder(128, 512, 8, 0.3, name=\"transformer_layer\")(x_transformer)\n",
        "    x = Multiply()([x, x_transformer])\n",
        "    \n",
        "    x = Dense(units=128, activation='selu')(x)\n",
        "    \n",
        "    x_output = Dense(units=1)(x)\n",
        "    \n",
        "    x_output = ScaleLayer()(x_output)\n",
        "\n",
        "    model = Model(inputs=x_input, outputs=x_output, \n",
        "                  name='DNN_Model')\n",
        "    return model"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_d8ybRJc9co2",
        "outputId": "34b24203-7271-46b1-cea4-f6574bf716e8"
      },
      "source": [
        "model = dnn_model()\n",
        "model.summary()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"DNN_Model\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_1 (InputLayer)            [(None, 80, 66)]     0                                            \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional (Bidirectional)   (None, 80, 1536)     5130240     input_1[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional_1 (Bidirectional) (None, 80, 1024)     8392704     bidirectional[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional_2 (Bidirectional) (None, 80, 768)      4328448     bidirectional_1[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional_5 (Bidirectional) (None, 80, 768)      3248640     bidirectional_1[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "multiply (Multiply)             (None, 80, 768)      0           bidirectional_2[0][0]            \n",
            "                                                                 bidirectional_5[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization (BatchNorma (None, 80, 768)      3072        multiply[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional_3 (Bidirectional) (None, 80, 512)      2099200     bidirectional_2[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional_6 (Bidirectional) (None, 80, 512)      1575936     batch_normalization[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "multiply_1 (Multiply)           (None, 80, 512)      0           bidirectional_3[0][0]            \n",
            "                                                                 bidirectional_6[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_1 (BatchNor (None, 80, 512)      2048        multiply_1[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional_4 (Bidirectional) (None, 80, 256)      656384      bidirectional_3[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional_7 (Bidirectional) (None, 80, 256)      493056      batch_normalization_1[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "multiply_2 (Multiply)           (None, 80, 256)      0           bidirectional_4[0][0]            \n",
            "                                                                 bidirectional_7[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_2 (BatchNor (None, 80, 256)      1024        multiply_2[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "bidirectional_8 (Bidirectional) (None, 80, 128)      123648      batch_normalization_2[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "concatenate (Concatenate)       (None, 80, 1920)     0           bidirectional_4[0][0]            \n",
            "                                                                 bidirectional_5[0][0]            \n",
            "                                                                 bidirectional_6[0][0]            \n",
            "                                                                 bidirectional_7[0][0]            \n",
            "                                                                 bidirectional_8[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "dense (Dense)                   (None, 80, 128)      245888      concatenate[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_3 (BatchNor (None, 80, 128)      512         dense[0][0]                      \n",
            "__________________________________________________________________________________________________\n",
            "transformer_layer (TransformerE (None, 80, 128)      659712      batch_normalization_3[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "multiply_3 (Multiply)           (None, 80, 128)      0           dense[0][0]                      \n",
            "                                                                 transformer_layer[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "dense_3 (Dense)                 (None, 80, 128)      16512       multiply_3[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "dense_4 (Dense)                 (None, 80, 1)        129         dense_3[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "scale_layer (ScaleLayer)        (None, 80, 1)        0           dense_4[0][0]                    \n",
            "==================================================================================================\n",
            "Total params: 26,977,153\n",
            "Trainable params: 26,973,825\n",
            "Non-trainable params: 3,328\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "e-M0ZMvC9co3",
        "outputId": "c9ebe4bc-b060-449c-9666-431b1642d338"
      },
      "source": [
        "plot_model(\n",
        "    model, \n",
        "    to_file='Google_Brain_Keras_Model.png', \n",
        "    show_shapes=True,\n",
        "    show_layer_names=True\n",
        ")\n"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rxWQVBYm-0VK",
        "outputId": "981a73cb-446c-4ae8-e759-b53cea4e2147"
      },
      "source": [
        "del model\n",
        "gc.collect()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "22831"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u1v9SMji9co3"
      },
      "source": [
        "#reset Keras Session\n",
        "def reset_keras():\n",
        "    sess = tf.compat.v1.keras.backend.get_session()\n",
        "    tf.compat.v1.keras.backend.clear_session()\n",
        "    sess.close()\n",
        "    sess = tf.compat.v1.keras.backend.get_session()\n",
        "\n",
        "    # use the same config as you used to create the session\n",
        "    config = tf.compat.v1.ConfigProto()\n",
        "    config.gpu_options.per_process_gpu_memory_fraction = 1\n",
        "    config.gpu_options.visible_device_list = \"0\"\n",
        "    tf.compat.v1.keras.backend.set_session(tf.compat.v1.Session(config=config))\n",
        "    gc.collect()\n",
        "\n",
        "reset_keras()"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qDsDMBnE9co3"
      },
      "source": [
        "train_pred = np.zeros(targets.shape)"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "INtJvC7xfKdV"
      },
      "source": [
        "#skip_folds = np.arange(1)\n",
        "skip_folds = []"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SYJkCi1-9co3",
        "outputId": "1cf7654a-32d3-4ada-a2a5-3206a365e195"
      },
      "source": [
        "with tpu_strategy.scope():\n",
        "    \n",
        "    VERBOSE = 1\n",
        "    EPOCHS = 200\n",
        "    FOLDS = 7\n",
        "    if DEBUG:\n",
        "      EPOCHS = 100\n",
        "    test_preds = []\n",
        "    kf = KFold(n_splits=FOLDS, shuffle=True, random_state=2021)\n",
        "    \n",
        "    for fold, (train_idx, test_idx) in enumerate(kf.split(train, targets)):\n",
        "        if len(test_preds) > fold:\n",
        "          print(f'Ignore corrent fold {fold+1} as we already predicted test data')\n",
        "\n",
        "        if DEBUG and fold > 0:\n",
        "          continue\n",
        "        X_train, X_valid = train[train_idx], train[test_idx]\n",
        "        y_train, y_valid = targets[train_idx], targets[test_idx]\n",
        "        \n",
        "        model = dnn_model()\n",
        "        model.compile(optimizer=\"adam\", loss=\"mae\")\n",
        "\n",
        "        lr = ReduceLROnPlateau(monitor=\"val_loss\", factor=0.75, \n",
        "                               patience=10, verbose=VERBOSE)\n",
        "        checkpoint_name = f'./model-gpu/Bidirect_LSTM_model_{fold+1}C.h5'\n",
        "        chk_point = ModelCheckpoint(checkpoint_name,\n",
        "                                    monitor='val_loss', verbose=VERBOSE, \n",
        "                                    save_best_only=True, mode='min')\n",
        "\n",
        "        es = EarlyStopping(monitor=\"val_loss\", patience=50, \n",
        "                           verbose=VERBOSE, mode=\"min\", \n",
        "                           restore_best_weights=True)\n",
        "        \n",
        "        if fold not in skip_folds:\n",
        "          model.fit(X_train, y_train, \n",
        "                    validation_data=(X_valid, y_valid), \n",
        "                    epochs=EPOCHS,\n",
        "                    verbose=VERBOSE,\n",
        "                    batch_size=BATCH_SIZE, \n",
        "                    callbacks=[lr, chk_point, es])\n",
        "        else:\n",
        "          print('Load pretrain weight from ', checkpoint_name)\n",
        "          load_locally = tf.saved_model.LoadOptions(experimental_io_device='/job:localhost')\n",
        "          try:\n",
        "            # At loading time, register the custom objects with a `custom_object_scope`:\n",
        "            custom_objects = {\"ScaleLayer\": ScaleLayer}\n",
        "            with tf.keras.utils.custom_object_scope(custom_objects):\n",
        "                model = load_model(checkpoint_name, options=load_locally)\n",
        "          except (OSError):\n",
        "            print('Cannot load pretrain weight from ', checkpoint_name)\n",
        "            break\n",
        "                \n",
        "        y_true = y_valid.squeeze().reshape(-1, 1)\n",
        "        y_pred = model.predict(X_valid, batch_size=BATCH_SIZE).squeeze().reshape(-1, 1)\n",
        "        train_pred[test_idx] = y_pred.reshape(-1, 80)\n",
        "        score = mean_absolute_error(y_true, y_pred)\n",
        "        test_preds.append(model.predict(test, batch_size=BATCH_SIZE).squeeze().reshape(-1, 1).squeeze())\n",
        "\n",
        "        print(f\"Fold-{fold+1} | OOF Score: {score}\")\n",
        "        \n",
        "        del X_train, X_valid, y_train, y_valid, model\n",
        "        reset_keras()\n",
        "        gc.collect()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/200\n",
            "127/127 [==============================] - 199s 1s/step - loss: 1.4014 - val_loss: 1.5946\n",
            "\n",
            "Epoch 00001: val_loss improved from inf to 1.59460, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/utils/generic_utils.py:497: CustomMaskWarning: Custom mask layers require a config and must override get_config. When loading, the custom mask layer must be passed to the custom_objects argument.\n",
            "  category=CustomMaskWarning)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 2/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.6075 - val_loss: 0.8406\n",
            "\n",
            "Epoch 00002: val_loss improved from 1.59460 to 0.84064, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 3/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.5435 - val_loss: 0.9137\n",
            "\n",
            "Epoch 00003: val_loss did not improve from 0.84064\n",
            "Epoch 4/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.4927 - val_loss: 0.5022\n",
            "\n",
            "Epoch 00004: val_loss improved from 0.84064 to 0.50222, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 5/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.4673 - val_loss: 0.6438\n",
            "\n",
            "Epoch 00005: val_loss did not improve from 0.50222\n",
            "Epoch 6/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.4613 - val_loss: 0.5085\n",
            "\n",
            "Epoch 00006: val_loss did not improve from 0.50222\n",
            "Epoch 7/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.4261 - val_loss: 0.4284\n",
            "\n",
            "Epoch 00007: val_loss improved from 0.50222 to 0.42835, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 8/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3988 - val_loss: 0.3771\n",
            "\n",
            "Epoch 00008: val_loss improved from 0.42835 to 0.37713, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 9/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.4065 - val_loss: 0.3854\n",
            "\n",
            "Epoch 00009: val_loss did not improve from 0.37713\n",
            "Epoch 10/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.4075 - val_loss: 0.3849\n",
            "\n",
            "Epoch 00010: val_loss did not improve from 0.37713\n",
            "Epoch 11/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3531 - val_loss: 0.3331\n",
            "\n",
            "Epoch 00011: val_loss improved from 0.37713 to 0.33308, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 12/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3436 - val_loss: 0.3283\n",
            "\n",
            "Epoch 00012: val_loss improved from 0.33308 to 0.32833, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 13/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3515 - val_loss: 0.5256\n",
            "\n",
            "Epoch 00013: val_loss did not improve from 0.32833\n",
            "Epoch 14/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3322 - val_loss: 0.3758\n",
            "\n",
            "Epoch 00014: val_loss did not improve from 0.32833\n",
            "Epoch 15/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3180 - val_loss: 0.3818\n",
            "\n",
            "Epoch 00015: val_loss did not improve from 0.32833\n",
            "Epoch 16/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3216 - val_loss: 0.3306\n",
            "\n",
            "Epoch 00016: val_loss did not improve from 0.32833\n",
            "Epoch 17/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3024 - val_loss: 0.3977\n",
            "\n",
            "Epoch 00017: val_loss did not improve from 0.32833\n",
            "Epoch 18/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.3083 - val_loss: 0.2980\n",
            "\n",
            "Epoch 00018: val_loss improved from 0.32833 to 0.29797, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 19/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2886 - val_loss: 0.3006\n",
            "\n",
            "Epoch 00019: val_loss did not improve from 0.29797\n",
            "Epoch 20/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2779 - val_loss: 0.3142\n",
            "\n",
            "Epoch 00020: val_loss did not improve from 0.29797\n",
            "Epoch 21/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2873 - val_loss: 0.3195\n",
            "\n",
            "Epoch 00021: val_loss did not improve from 0.29797\n",
            "Epoch 22/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2778 - val_loss: 0.3022\n",
            "\n",
            "Epoch 00022: val_loss did not improve from 0.29797\n",
            "Epoch 23/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2647 - val_loss: 0.2591\n",
            "\n",
            "Epoch 00023: val_loss improved from 0.29797 to 0.25913, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 24/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2570 - val_loss: 0.2762\n",
            "\n",
            "Epoch 00024: val_loss did not improve from 0.25913\n",
            "Epoch 25/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2587 - val_loss: 0.2662\n",
            "\n",
            "Epoch 00025: val_loss did not improve from 0.25913\n",
            "Epoch 26/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2635 - val_loss: 0.2732\n",
            "\n",
            "Epoch 00026: val_loss did not improve from 0.25913\n",
            "Epoch 27/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2702 - val_loss: 0.2815\n",
            "\n",
            "Epoch 00027: val_loss did not improve from 0.25913\n",
            "Epoch 28/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2505 - val_loss: 0.2708\n",
            "\n",
            "Epoch 00028: val_loss did not improve from 0.25913\n",
            "Epoch 29/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2559 - val_loss: 0.2681\n",
            "\n",
            "Epoch 00029: val_loss did not improve from 0.25913\n",
            "Epoch 30/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2577 - val_loss: 0.2633\n",
            "\n",
            "Epoch 00030: val_loss did not improve from 0.25913\n",
            "Epoch 31/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2563 - val_loss: 0.2539\n",
            "\n",
            "Epoch 00031: val_loss improved from 0.25913 to 0.25389, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 32/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2497 - val_loss: 0.2458\n",
            "\n",
            "Epoch 00032: val_loss improved from 0.25389 to 0.24584, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 33/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2467 - val_loss: 0.2580\n",
            "\n",
            "Epoch 00033: val_loss did not improve from 0.24584\n",
            "Epoch 34/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2388 - val_loss: 0.2380\n",
            "\n",
            "Epoch 00034: val_loss improved from 0.24584 to 0.23799, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 35/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2317 - val_loss: 0.2398\n",
            "\n",
            "Epoch 00035: val_loss did not improve from 0.23799\n",
            "Epoch 36/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2281 - val_loss: 0.2392\n",
            "\n",
            "Epoch 00036: val_loss did not improve from 0.23799\n",
            "Epoch 37/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2323 - val_loss: 0.2668\n",
            "\n",
            "Epoch 00037: val_loss did not improve from 0.23799\n",
            "Epoch 38/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2282 - val_loss: 0.2736\n",
            "\n",
            "Epoch 00038: val_loss did not improve from 0.23799\n",
            "Epoch 39/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.2359 - val_loss: 0.2325\n",
            "\n",
            "Epoch 00039: val_loss improved from 0.23799 to 0.23247, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 40/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2382 - val_loss: 0.2294\n",
            "\n",
            "Epoch 00040: val_loss improved from 0.23247 to 0.22945, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 41/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2327 - val_loss: 0.2400\n",
            "\n",
            "Epoch 00041: val_loss did not improve from 0.22945\n",
            "Epoch 42/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2248 - val_loss: 0.2437\n",
            "\n",
            "Epoch 00042: val_loss did not improve from 0.22945\n",
            "Epoch 43/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2263 - val_loss: 0.2445\n",
            "\n",
            "Epoch 00043: val_loss did not improve from 0.22945\n",
            "Epoch 44/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.2274 - val_loss: 0.2333\n",
            "\n",
            "Epoch 00044: val_loss did not improve from 0.22945\n",
            "Epoch 45/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2192 - val_loss: 0.2208\n",
            "\n",
            "Epoch 00045: val_loss improved from 0.22945 to 0.22082, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 46/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2114 - val_loss: 0.2296\n",
            "\n",
            "Epoch 00046: val_loss did not improve from 0.22082\n",
            "Epoch 47/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2029 - val_loss: 0.2141\n",
            "\n",
            "Epoch 00047: val_loss improved from 0.22082 to 0.21408, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 48/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2018 - val_loss: 0.2136\n",
            "\n",
            "Epoch 00048: val_loss improved from 0.21408 to 0.21356, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 49/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1982 - val_loss: 0.2283\n",
            "\n",
            "Epoch 00049: val_loss did not improve from 0.21356\n",
            "Epoch 50/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1973 - val_loss: 0.2396\n",
            "\n",
            "Epoch 00050: val_loss did not improve from 0.21356\n",
            "Epoch 51/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.2018 - val_loss: 0.2335\n",
            "\n",
            "Epoch 00051: val_loss did not improve from 0.21356\n",
            "Epoch 52/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1975 - val_loss: 0.2136\n",
            "\n",
            "Epoch 00052: val_loss did not improve from 0.21356\n",
            "Epoch 53/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1956 - val_loss: 0.2154\n",
            "\n",
            "Epoch 00053: val_loss did not improve from 0.21356\n",
            "Epoch 54/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1956 - val_loss: 0.2199\n",
            "\n",
            "Epoch 00054: val_loss did not improve from 0.21356\n",
            "Epoch 55/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1964 - val_loss: 0.2144\n",
            "\n",
            "Epoch 00055: val_loss did not improve from 0.21356\n",
            "Epoch 56/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1963 - val_loss: 0.2087\n",
            "\n",
            "Epoch 00056: val_loss improved from 0.21356 to 0.20875, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 57/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.1887 - val_loss: 0.2085\n",
            "\n",
            "Epoch 00057: val_loss improved from 0.20875 to 0.20847, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 58/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1992 - val_loss: 0.2204\n",
            "\n",
            "Epoch 00058: val_loss did not improve from 0.20847\n",
            "Epoch 59/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1895 - val_loss: 0.1946\n",
            "\n",
            "Epoch 00059: val_loss improved from 0.20847 to 0.19459, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 60/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1950 - val_loss: 0.2208\n",
            "\n",
            "Epoch 00060: val_loss did not improve from 0.19459\n",
            "Epoch 61/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1933 - val_loss: 0.2173\n",
            "\n",
            "Epoch 00061: val_loss did not improve from 0.19459\n",
            "Epoch 62/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1910 - val_loss: 0.2152\n",
            "\n",
            "Epoch 00062: val_loss did not improve from 0.19459\n",
            "Epoch 63/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1796 - val_loss: 0.2095\n",
            "\n",
            "Epoch 00063: val_loss did not improve from 0.19459\n",
            "Epoch 64/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1891 - val_loss: 0.2355\n",
            "\n",
            "Epoch 00064: val_loss did not improve from 0.19459\n",
            "Epoch 65/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.1768 - val_loss: 0.2008\n",
            "\n",
            "Epoch 00065: val_loss did not improve from 0.19459\n",
            "Epoch 66/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1781 - val_loss: 0.2129\n",
            "\n",
            "Epoch 00066: val_loss did not improve from 0.19459\n",
            "Epoch 67/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1805 - val_loss: 0.2006\n",
            "\n",
            "Epoch 00067: val_loss did not improve from 0.19459\n",
            "Epoch 68/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1786 - val_loss: 0.2058\n",
            "\n",
            "Epoch 00068: val_loss did not improve from 0.19459\n",
            "Epoch 69/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1773 - val_loss: 0.2035\n",
            "\n",
            "Epoch 00069: ReduceLROnPlateau reducing learning rate to 0.0007500000356230885.\n",
            "\n",
            "Epoch 00069: val_loss did not improve from 0.19459\n",
            "Epoch 70/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1590 - val_loss: 0.1994\n",
            "\n",
            "Epoch 00070: val_loss did not improve from 0.19459\n",
            "Epoch 71/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1537 - val_loss: 0.1756\n",
            "\n",
            "Epoch 00071: val_loss improved from 0.19459 to 0.17564, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 72/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1488 - val_loss: 0.1833\n",
            "\n",
            "Epoch 00072: val_loss did not improve from 0.17564\n",
            "Epoch 73/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1547 - val_loss: 0.1908\n",
            "\n",
            "Epoch 00073: val_loss did not improve from 0.17564\n",
            "Epoch 74/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1503 - val_loss: 0.1853\n",
            "\n",
            "Epoch 00074: val_loss did not improve from 0.17564\n",
            "Epoch 75/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1514 - val_loss: 0.1762\n",
            "\n",
            "Epoch 00075: val_loss did not improve from 0.17564\n",
            "Epoch 76/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1429 - val_loss: 0.1940\n",
            "\n",
            "Epoch 00076: val_loss did not improve from 0.17564\n",
            "Epoch 77/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.1486 - val_loss: 0.1782\n",
            "\n",
            "Epoch 00077: val_loss did not improve from 0.17564\n",
            "Epoch 78/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.1506 - val_loss: 0.1804\n",
            "\n",
            "Epoch 00078: val_loss did not improve from 0.17564\n",
            "Epoch 79/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1445 - val_loss: 0.1756\n",
            "\n",
            "Epoch 00079: val_loss improved from 0.17564 to 0.17557, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 80/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1451 - val_loss: 0.1999\n",
            "\n",
            "Epoch 00080: val_loss did not improve from 0.17557\n",
            "Epoch 81/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.1447 - val_loss: 0.1824\n",
            "\n",
            "Epoch 00081: ReduceLROnPlateau reducing learning rate to 0.0005625000048894435.\n",
            "\n",
            "Epoch 00081: val_loss did not improve from 0.17557\n",
            "Epoch 82/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1340 - val_loss: 0.1717\n",
            "\n",
            "Epoch 00082: val_loss improved from 0.17557 to 0.17168, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 83/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1294 - val_loss: 0.1714\n",
            "\n",
            "Epoch 00083: val_loss improved from 0.17168 to 0.17136, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 84/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1278 - val_loss: 0.1820\n",
            "\n",
            "Epoch 00084: val_loss did not improve from 0.17136\n",
            "Epoch 85/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.1296 - val_loss: 0.1692\n",
            "\n",
            "Epoch 00085: val_loss improved from 0.17136 to 0.16925, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 86/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1240 - val_loss: 0.1674\n",
            "\n",
            "Epoch 00086: val_loss improved from 0.16925 to 0.16735, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 87/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1219 - val_loss: 0.1678\n",
            "\n",
            "Epoch 00087: val_loss did not improve from 0.16735\n",
            "Epoch 88/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.1245 - val_loss: 0.1713\n",
            "\n",
            "Epoch 00088: val_loss did not improve from 0.16735\n",
            "Epoch 89/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1217 - val_loss: 0.1721\n",
            "\n",
            "Epoch 00089: val_loss did not improve from 0.16735\n",
            "Epoch 90/200\n",
            "127/127 [==============================] - 161s 1s/step - loss: 0.1263 - val_loss: 0.1729\n",
            "\n",
            "Epoch 00090: val_loss did not improve from 0.16735\n",
            "Epoch 91/200\n",
            "127/127 [==============================] - 162s 1s/step - loss: 0.1188 - val_loss: 0.1715\n",
            "\n",
            "Epoch 00091: val_loss did not improve from 0.16735\n",
            "Epoch 92/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1228 - val_loss: 0.1843\n",
            "\n",
            "Epoch 00092: val_loss did not improve from 0.16735\n",
            "Epoch 93/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1211 - val_loss: 0.1787\n",
            "\n",
            "Epoch 00093: val_loss did not improve from 0.16735\n",
            "Epoch 94/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1179 - val_loss: 0.1692\n",
            "\n",
            "Epoch 00094: val_loss did not improve from 0.16735\n",
            "Epoch 95/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1139 - val_loss: 0.1675\n",
            "\n",
            "Epoch 00095: val_loss did not improve from 0.16735\n",
            "Epoch 96/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1147 - val_loss: 0.1681\n",
            "\n",
            "Epoch 00096: ReduceLROnPlateau reducing learning rate to 0.0004218749818392098.\n",
            "\n",
            "Epoch 00096: val_loss did not improve from 0.16735\n",
            "Epoch 97/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1096 - val_loss: 0.1652\n",
            "\n",
            "Epoch 00097: val_loss improved from 0.16735 to 0.16523, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 98/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1054 - val_loss: 0.1664\n",
            "\n",
            "Epoch 00098: val_loss did not improve from 0.16523\n",
            "Epoch 99/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1030 - val_loss: 0.1721\n",
            "\n",
            "Epoch 00099: val_loss did not improve from 0.16523\n",
            "Epoch 100/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1043 - val_loss: 0.1707\n",
            "\n",
            "Epoch 00100: val_loss did not improve from 0.16523\n",
            "Epoch 101/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1022 - val_loss: 0.1665\n",
            "\n",
            "Epoch 00101: val_loss did not improve from 0.16523\n",
            "Epoch 102/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1013 - val_loss: 0.1650\n",
            "\n",
            "Epoch 00102: val_loss improved from 0.16523 to 0.16497, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 103/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1018 - val_loss: 0.1645\n",
            "\n",
            "Epoch 00103: val_loss improved from 0.16497 to 0.16449, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 104/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1003 - val_loss: 0.1648\n",
            "\n",
            "Epoch 00104: val_loss did not improve from 0.16449\n",
            "Epoch 105/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.1005 - val_loss: 0.1663\n",
            "\n",
            "Epoch 00105: val_loss did not improve from 0.16449\n",
            "Epoch 106/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0995 - val_loss: 0.1662\n",
            "\n",
            "Epoch 00106: val_loss did not improve from 0.16449\n",
            "Epoch 107/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0984 - val_loss: 0.1684\n",
            "\n",
            "Epoch 00107: val_loss did not improve from 0.16449\n",
            "Epoch 108/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0981 - val_loss: 0.1645\n",
            "\n",
            "Epoch 00108: val_loss did not improve from 0.16449\n",
            "Epoch 109/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0966 - val_loss: 0.1642\n",
            "\n",
            "Epoch 00109: val_loss improved from 0.16449 to 0.16418, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 110/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0980 - val_loss: 0.1649\n",
            "\n",
            "Epoch 00110: val_loss did not improve from 0.16418\n",
            "Epoch 111/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0964 - val_loss: 0.1672\n",
            "\n",
            "Epoch 00111: val_loss did not improve from 0.16418\n",
            "Epoch 112/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0950 - val_loss: 0.1675\n",
            "\n",
            "Epoch 00112: val_loss did not improve from 0.16418\n",
            "Epoch 113/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0920 - val_loss: 0.1733\n",
            "\n",
            "Epoch 00113: val_loss did not improve from 0.16418\n",
            "Epoch 114/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0913 - val_loss: 0.1667\n",
            "\n",
            "Epoch 00114: val_loss did not improve from 0.16418\n",
            "Epoch 115/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0908 - val_loss: 0.1657\n",
            "\n",
            "Epoch 00115: val_loss did not improve from 0.16418\n",
            "Epoch 116/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0934 - val_loss: 0.1695\n",
            "\n",
            "Epoch 00116: val_loss did not improve from 0.16418\n",
            "Epoch 117/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0895 - val_loss: 0.1656\n",
            "\n",
            "Epoch 00117: val_loss did not improve from 0.16418\n",
            "Epoch 118/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0911 - val_loss: 0.1702\n",
            "\n",
            "Epoch 00118: val_loss did not improve from 0.16418\n",
            "Epoch 119/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0882 - val_loss: 0.1653\n",
            "\n",
            "Epoch 00119: ReduceLROnPlateau reducing learning rate to 0.00031640623637940735.\n",
            "\n",
            "Epoch 00119: val_loss did not improve from 0.16418\n",
            "Epoch 120/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0821 - val_loss: 0.1641\n",
            "\n",
            "Epoch 00120: val_loss improved from 0.16418 to 0.16413, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 121/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0809 - val_loss: 0.1630\n",
            "\n",
            "Epoch 00121: val_loss improved from 0.16413 to 0.16299, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 122/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0783 - val_loss: 0.1676\n",
            "\n",
            "Epoch 00122: val_loss did not improve from 0.16299\n",
            "Epoch 123/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0808 - val_loss: 0.1659\n",
            "\n",
            "Epoch 00123: val_loss did not improve from 0.16299\n",
            "Epoch 124/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0793 - val_loss: 0.1630\n",
            "\n",
            "Epoch 00124: val_loss improved from 0.16299 to 0.16299, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 125/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0765 - val_loss: 0.1631\n",
            "\n",
            "Epoch 00125: val_loss did not improve from 0.16299\n",
            "Epoch 126/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0765 - val_loss: 0.1682\n",
            "\n",
            "Epoch 00126: val_loss did not improve from 0.16299\n",
            "Epoch 127/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0776 - val_loss: 0.1655\n",
            "\n",
            "Epoch 00127: val_loss did not improve from 0.16299\n",
            "Epoch 128/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0779 - val_loss: 0.1663\n",
            "\n",
            "Epoch 00128: val_loss did not improve from 0.16299\n",
            "Epoch 129/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0772 - val_loss: 0.1658\n",
            "\n",
            "Epoch 00129: val_loss did not improve from 0.16299\n",
            "Epoch 130/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0741 - val_loss: 0.1631\n",
            "\n",
            "Epoch 00130: val_loss did not improve from 0.16299\n",
            "Epoch 131/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0747 - val_loss: 0.1639\n",
            "\n",
            "Epoch 00131: ReduceLROnPlateau reducing learning rate to 0.00023730468819849193.\n",
            "\n",
            "Epoch 00131: val_loss did not improve from 0.16299\n",
            "Epoch 132/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0704 - val_loss: 0.1621\n",
            "\n",
            "Epoch 00132: val_loss improved from 0.16299 to 0.16212, saving model to ./model-gpu/Bidirect_LSTM_model_1C.h5\n",
            "Epoch 133/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0683 - val_loss: 0.1637\n",
            "\n",
            "Epoch 00133: val_loss did not improve from 0.16212\n",
            "Epoch 134/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0662 - val_loss: 0.1649\n",
            "\n",
            "Epoch 00134: val_loss did not improve from 0.16212\n",
            "Epoch 135/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0664 - val_loss: 0.1644\n",
            "\n",
            "Epoch 00135: val_loss did not improve from 0.16212\n",
            "Epoch 136/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0665 - val_loss: 0.1634\n",
            "\n",
            "Epoch 00136: val_loss did not improve from 0.16212\n",
            "Epoch 137/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0656 - val_loss: 0.1660\n",
            "\n",
            "Epoch 00137: val_loss did not improve from 0.16212\n",
            "Epoch 138/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0643 - val_loss: 0.1663\n",
            "\n",
            "Epoch 00138: val_loss did not improve from 0.16212\n",
            "Epoch 139/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0653 - val_loss: 0.1642\n",
            "\n",
            "Epoch 00139: val_loss did not improve from 0.16212\n",
            "Epoch 140/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0660 - val_loss: 0.1655\n",
            "\n",
            "Epoch 00140: val_loss did not improve from 0.16212\n",
            "Epoch 141/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0664 - val_loss: 0.1648\n",
            "\n",
            "Epoch 00141: val_loss did not improve from 0.16212\n",
            "Epoch 142/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0638 - val_loss: 0.1638\n",
            "\n",
            "Epoch 00142: ReduceLROnPlateau reducing learning rate to 0.00017797851614886895.\n",
            "\n",
            "Epoch 00142: val_loss did not improve from 0.16212\n",
            "Epoch 143/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0606 - val_loss: 0.1641\n",
            "\n",
            "Epoch 00143: val_loss did not improve from 0.16212\n",
            "Epoch 144/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0598 - val_loss: 0.1632\n",
            "\n",
            "Epoch 00144: val_loss did not improve from 0.16212\n",
            "Epoch 145/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0589 - val_loss: 0.1660\n",
            "\n",
            "Epoch 00145: val_loss did not improve from 0.16212\n",
            "Epoch 146/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0594 - val_loss: 0.1640\n",
            "\n",
            "Epoch 00146: val_loss did not improve from 0.16212\n",
            "Epoch 147/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0583 - val_loss: 0.1636\n",
            "\n",
            "Epoch 00147: val_loss did not improve from 0.16212\n",
            "Epoch 148/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0577 - val_loss: 0.1630\n",
            "\n",
            "Epoch 00148: val_loss did not improve from 0.16212\n",
            "Epoch 149/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0573 - val_loss: 0.1634\n",
            "\n",
            "Epoch 00149: val_loss did not improve from 0.16212\n",
            "Epoch 150/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0577 - val_loss: 0.1631\n",
            "\n",
            "Epoch 00150: val_loss did not improve from 0.16212\n",
            "Epoch 151/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0575 - val_loss: 0.1669\n",
            "\n",
            "Epoch 00151: val_loss did not improve from 0.16212\n",
            "Epoch 152/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0579 - val_loss: 0.1642\n",
            "\n",
            "Epoch 00152: ReduceLROnPlateau reducing learning rate to 0.0001334838816546835.\n",
            "\n",
            "Epoch 00152: val_loss did not improve from 0.16212\n",
            "Epoch 153/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0540 - val_loss: 0.1648\n",
            "\n",
            "Epoch 00153: val_loss did not improve from 0.16212\n",
            "Epoch 154/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0527 - val_loss: 0.1632\n",
            "\n",
            "Epoch 00154: val_loss did not improve from 0.16212\n",
            "Epoch 155/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0528 - val_loss: 0.1649\n",
            "\n",
            "Epoch 00155: val_loss did not improve from 0.16212\n",
            "Epoch 156/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0524 - val_loss: 0.1636\n",
            "\n",
            "Epoch 00156: val_loss did not improve from 0.16212\n",
            "Epoch 157/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0528 - val_loss: 0.1673\n",
            "\n",
            "Epoch 00157: val_loss did not improve from 0.16212\n",
            "Epoch 158/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0531 - val_loss: 0.1642\n",
            "\n",
            "Epoch 00158: val_loss did not improve from 0.16212\n",
            "Epoch 159/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0527 - val_loss: 0.1648\n",
            "\n",
            "Epoch 00159: val_loss did not improve from 0.16212\n",
            "Epoch 160/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0524 - val_loss: 0.1671\n",
            "\n",
            "Epoch 00160: val_loss did not improve from 0.16212\n",
            "Epoch 161/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0516 - val_loss: 0.1658\n",
            "\n",
            "Epoch 00161: val_loss did not improve from 0.16212\n",
            "Epoch 162/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0519 - val_loss: 0.1638\n",
            "\n",
            "Epoch 00162: ReduceLROnPlateau reducing learning rate to 0.00010011290578404441.\n",
            "\n",
            "Epoch 00162: val_loss did not improve from 0.16212\n",
            "Epoch 163/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0503 - val_loss: 0.1636\n",
            "\n",
            "Epoch 00163: val_loss did not improve from 0.16212\n",
            "Epoch 164/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0492 - val_loss: 0.1636\n",
            "\n",
            "Epoch 00164: val_loss did not improve from 0.16212\n",
            "Epoch 165/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0489 - val_loss: 0.1639\n",
            "\n",
            "Epoch 00165: val_loss did not improve from 0.16212\n",
            "Epoch 166/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0492 - val_loss: 0.1648\n",
            "\n",
            "Epoch 00166: val_loss did not improve from 0.16212\n",
            "Epoch 167/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0485 - val_loss: 0.1653\n",
            "\n",
            "Epoch 00167: val_loss did not improve from 0.16212\n",
            "Epoch 168/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0485 - val_loss: 0.1637\n",
            "\n",
            "Epoch 00168: val_loss did not improve from 0.16212\n",
            "Epoch 169/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0485 - val_loss: 0.1662\n",
            "\n",
            "Epoch 00169: val_loss did not improve from 0.16212\n",
            "Epoch 170/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0480 - val_loss: 0.1649\n",
            "\n",
            "Epoch 00170: val_loss did not improve from 0.16212\n",
            "Epoch 171/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0480 - val_loss: 0.1637\n",
            "\n",
            "Epoch 00171: val_loss did not improve from 0.16212\n",
            "Epoch 172/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0478 - val_loss: 0.1651\n",
            "\n",
            "Epoch 00172: ReduceLROnPlateau reducing learning rate to 7.508467933803331e-05.\n",
            "\n",
            "Epoch 00172: val_loss did not improve from 0.16212\n",
            "Epoch 173/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0467 - val_loss: 0.1644\n",
            "\n",
            "Epoch 00173: val_loss did not improve from 0.16212\n",
            "Epoch 174/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0462 - val_loss: 0.1638\n",
            "\n",
            "Epoch 00174: val_loss did not improve from 0.16212\n",
            "Epoch 175/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0457 - val_loss: 0.1641\n",
            "\n",
            "Epoch 00175: val_loss did not improve from 0.16212\n",
            "Epoch 176/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.0465 - val_loss: 0.1641\n",
            "\n",
            "Epoch 00176: val_loss did not improve from 0.16212\n",
            "Epoch 177/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0456 - val_loss: 0.1647\n",
            "\n",
            "Epoch 00177: val_loss did not improve from 0.16212\n",
            "Epoch 178/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0454 - val_loss: 0.1648\n",
            "\n",
            "Epoch 00178: val_loss did not improve from 0.16212\n",
            "Epoch 179/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0454 - val_loss: 0.1668\n",
            "\n",
            "Epoch 00179: val_loss did not improve from 0.16212\n",
            "Epoch 180/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0453 - val_loss: 0.1650\n",
            "\n",
            "Epoch 00180: val_loss did not improve from 0.16212\n",
            "Epoch 181/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0450 - val_loss: 0.1678\n",
            "\n",
            "Epoch 00181: val_loss did not improve from 0.16212\n",
            "Epoch 182/200\n",
            "127/127 [==============================] - 163s 1s/step - loss: 0.0448 - val_loss: 0.1655\n",
            "\n",
            "Epoch 00182: ReduceLROnPlateau reducing learning rate to 5.6313510867767036e-05.\n",
            "\n",
            "Epoch 00182: val_loss did not improve from 0.16212\n",
            "Restoring model weights from the end of the best epoch.\n",
            "Epoch 00182: early stopping\n",
            "Fold-1 | OOF Score: 0.16211654680482343\n",
            "Epoch 1/200\n",
            "127/127 [==============================] - 201s 1s/step - loss: 1.2274 - val_loss: 1.0773\n",
            "\n",
            "Epoch 00001: val_loss improved from inf to 1.07734, saving model to ./model-gpu/Bidirect_LSTM_model_2C.h5\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/utils/generic_utils.py:497: CustomMaskWarning: Custom mask layers require a config and must override get_config. When loading, the custom mask layer must be passed to the custom_objects argument.\n",
            "  category=CustomMaskWarning)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 2/200\n",
            "127/127 [==============================] - 164s 1s/step - loss: 0.5865 - val_loss: 0.7055\n",
            "\n",
            "Epoch 00002: val_loss improved from 1.07734 to 0.70552, saving model to ./model-gpu/Bidirect_LSTM_model_2C.h5\n",
            "Epoch 3/200\n",
            " 55/127 [===========>..................] - ETA: 1:27 - loss: 0.5373"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fUanGPDq9co3"
      },
      "source": [
        "if not DEBUG:\n",
        "  train_pred = train_pred.flatten()\n",
        "  targets = targets.flatten()\n",
        "  score = mean_absolute_error(targets, train_pred)\n",
        "  print(f\"Train Score: {score}\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QHx2oF849co4"
      },
      "source": [
        "## Create submission file"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tS9VjV7K9co4"
      },
      "source": [
        "if not DEBUG:\n",
        "  submission = pd.read_csv('../input/ventilator-pressure-prediction/sample_submission.csv')\n",
        "  submission[\"pressure\"] = sum(test_preds)/5\n",
        "  submission.to_csv('submission.csv', index=False)\n",
        "\n",
        "  # ENSEMBLE FOLDS WITH MEDIAN\n",
        "  #取中位数\n",
        "  submission[\"pressure\"] = np.median(np.vstack(test_preds),axis=0)\n",
        "  submission.to_csv('submission_median.csv', index=False)\n",
        "\n",
        "\n",
        "  # ENSEMBLE FOLDS WITH MEDIAN AND ROUND PREDICTIONS\n",
        "  submission[\"pressure\"] =\\\n",
        "      np.round( (submission.pressure - pressure_min)/pressure_step ) * pressure_step + pressure_min\n",
        "  submission.pressure = np.clip(submission.pressure, pressure_min, pressure_max)\n",
        "  submission.to_csv('submission_median_round.csv', index=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8lg88GUo4OCK"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}