Amazon_question_answer.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Install Necessary Library\n"
      ],
      "metadata": {
        "id": "IcGXwGo4o8gV"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w4QFrmePn5uJ",
        "outputId": "039a53b2-a9e3-4aa1-f655-13d260837a93"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting tensorflow==2.10\n",
            "  Downloading tensorflow-2.10.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (578.0 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m578.0/578.0 MB\u001b[0m \u001b[31m2.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (1.4.0)\n",
            "Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (1.6.3)\n",
            "Requirement already satisfied: flatbuffers>=2.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (24.3.25)\n",
            "Collecting gast<=0.4.0,>=0.2.1 (from tensorflow==2.10)\n",
            "  Downloading gast-0.4.0-py3-none-any.whl (9.8 kB)\n",
            "Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (0.2.0)\n",
            "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (1.63.0)\n",
            "Requirement already satisfied: h5py>=2.9.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (3.9.0)\n",
            "Collecting keras<2.11,>=2.10.0 (from tensorflow==2.10)\n",
            "  Downloading keras-2.10.0-py2.py3-none-any.whl (1.7 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.7/1.7 MB\u001b[0m \u001b[31m44.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting keras-preprocessing>=1.1.1 (from tensorflow==2.10)\n",
            "  Downloading Keras_Preprocessing-1.1.2-py2.py3-none-any.whl (42 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m42.6/42.6 kB\u001b[0m \u001b[31m4.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (18.1.1)\n",
            "Requirement already satisfied: numpy>=1.20 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (1.25.2)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (3.3.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (24.0)\n",
            "Collecting protobuf<3.20,>=3.9.2 (from tensorflow==2.10)\n",
            "  Downloading protobuf-3.19.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.1 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.1/1.1 MB\u001b[0m \u001b[31m23.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (67.7.2)\n",
            "Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (1.16.0)\n",
            "Collecting tensorboard<2.11,>=2.10 (from tensorflow==2.10)\n",
            "  Downloading tensorboard-2.10.1-py3-none-any.whl (5.9 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.9/5.9 MB\u001b[0m \u001b[31m25.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (0.37.0)\n",
            "Collecting tensorflow-estimator<2.11,>=2.10.0 (from tensorflow==2.10)\n",
            "  Downloading tensorflow_estimator-2.10.0-py2.py3-none-any.whl (438 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m438.7/438.7 kB\u001b[0m \u001b[31m13.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (2.4.0)\n",
            "Requirement already satisfied: typing-extensions>=3.6.6 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (4.11.0)\n",
            "Requirement already satisfied: wrapt>=1.11.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow==2.10) (1.14.1)\n",
            "Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.10/dist-packages (from astunparse>=1.6.0->tensorflow==2.10) (0.43.0)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10) (2.27.0)\n",
            "Collecting google-auth-oauthlib<0.5,>=0.4.1 (from tensorboard<2.11,>=2.10->tensorflow==2.10)\n",
            "  Downloading google_auth_oauthlib-0.4.6-py2.py3-none-any.whl (18 kB)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10) (3.6)\n",
            "Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10) (2.31.0)\n",
            "Collecting tensorboard-data-server<0.7.0,>=0.6.0 (from tensorboard<2.11,>=2.10->tensorflow==2.10)\n",
            "  Downloading tensorboard_data_server-0.6.1-py3-none-manylinux2010_x86_64.whl (4.9 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m4.9/4.9 MB\u001b[0m \u001b[31m31.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting tensorboard-plugin-wit>=1.6.0 (from tensorboard<2.11,>=2.10->tensorflow==2.10)\n",
            "  Downloading tensorboard_plugin_wit-1.8.1-py3-none-any.whl (781 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m781.3/781.3 kB\u001b[0m \u001b[31m13.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow==2.10) (3.0.3)\n",
            "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow==2.10) (5.3.3)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow==2.10) (0.4.0)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow==2.10) (4.9)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.11,>=2.10->tensorflow==2.10) (1.3.1)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.11,>=2.10->tensorflow==2.10) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.11,>=2.10->tensorflow==2.10) (3.7)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.11,>=2.10->tensorflow==2.10) (2.0.7)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2.21.0->tensorboard<2.11,>=2.10->tensorflow==2.10) (2024.2.2)\n",
            "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.10/dist-packages (from werkzeug>=1.0.1->tensorboard<2.11,>=2.10->tensorflow==2.10) (2.1.5)\n",
            "Requirement already satisfied: pyasn1<0.7.0,>=0.4.6 in /usr/local/lib/python3.10/dist-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow==2.10) (0.6.0)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.10/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.11,>=2.10->tensorflow==2.10) (3.2.2)\n",
            "Installing collected packages: tensorboard-plugin-wit, keras, tensorflow-estimator, tensorboard-data-server, protobuf, keras-preprocessing, gast, google-auth-oauthlib, tensorboard, tensorflow\n",
            "  Attempting uninstall: keras\n",
            "    Found existing installation: keras 2.15.0\n",
            "    Uninstalling keras-2.15.0:\n",
            "      Successfully uninstalled keras-2.15.0\n",
            "  Attempting uninstall: tensorflow-estimator\n",
            "    Found existing installation: tensorflow-estimator 2.15.0\n",
            "    Uninstalling tensorflow-estimator-2.15.0:\n",
            "      Successfully uninstalled tensorflow-estimator-2.15.0\n",
            "  Attempting uninstall: tensorboard-data-server\n",
            "    Found existing installation: tensorboard-data-server 0.7.2\n",
            "    Uninstalling tensorboard-data-server-0.7.2:\n",
            "      Successfully uninstalled tensorboard-data-server-0.7.2\n",
            "  Attempting uninstall: protobuf\n",
            "    Found existing installation: protobuf 3.20.3\n",
            "    Uninstalling protobuf-3.20.3:\n",
            "      Successfully uninstalled protobuf-3.20.3\n",
            "  Attempting uninstall: gast\n",
            "    Found existing installation: gast 0.5.4\n",
            "    Uninstalling gast-0.5.4:\n",
            "      Successfully uninstalled gast-0.5.4\n",
            "  Attempting uninstall: google-auth-oauthlib\n",
            "    Found existing installation: google-auth-oauthlib 1.2.0\n",
            "    Uninstalling google-auth-oauthlib-1.2.0:\n",
            "      Successfully uninstalled google-auth-oauthlib-1.2.0\n",
            "  Attempting uninstall: tensorboard\n",
            "    Found existing installation: tensorboard 2.15.2\n",
            "    Uninstalling tensorboard-2.15.2:\n",
            "      Successfully uninstalled tensorboard-2.15.2\n",
            "  Attempting uninstall: tensorflow\n",
            "    Found existing installation: tensorflow 2.15.0\n",
            "    Uninstalling tensorflow-2.15.0:\n",
            "      Successfully uninstalled tensorflow-2.15.0\n",
            "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
            "cudf-cu12 24.4.1 requires protobuf<5,>=3.20, but you have protobuf 3.19.6 which is incompatible.\n",
            "pandas-gbq 0.19.2 requires google-auth-oauthlib>=0.7.0, but you have google-auth-oauthlib 0.4.6 which is incompatible.\n",
            "tensorflow-datasets 4.9.4 requires protobuf>=3.20, but you have protobuf 3.19.6 which is incompatible.\n",
            "tensorflow-metadata 1.15.0 requires protobuf<4.21,>=3.20.3; python_version < \"3.11\", but you have protobuf 3.19.6 which is incompatible.\n",
            "tf-keras 2.15.1 requires tensorflow<2.16,>=2.15, but you have tensorflow 2.10.0 which is incompatible.\u001b[0m\u001b[31m\n",
            "\u001b[0mSuccessfully installed gast-0.4.0 google-auth-oauthlib-0.4.6 keras-2.10.0 keras-preprocessing-1.1.2 protobuf-3.19.6 tensorboard-2.10.1 tensorboard-data-server-0.6.1 tensorboard-plugin-wit-1.8.1 tensorflow-2.10.0 tensorflow-estimator-2.10.0\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "google"
                ]
              },
              "id": "b36fafae46a54f0389b35133aad61747"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting deepctr[cpu]\n",
            "  Downloading deepctr-0.9.3-py3-none-any.whl (141 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m141.2/141.2 kB\u001b[0m \u001b[31m1.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from deepctr[cpu]) (2.31.0)\n",
            "Collecting h5py==3.7.0 (from deepctr[cpu])\n",
            "  Downloading h5py-3.7.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (4.5 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m4.5/4.5 MB\u001b[0m \u001b[31m21.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: tensorflow!=1.7.*,!=1.8.*,>=1.4.0 in /usr/local/lib/python3.10/dist-packages (from deepctr[cpu]) (2.10.0)\n",
            "Requirement already satisfied: numpy>=1.14.5 in /usr/local/lib/python3.10/dist-packages (from h5py==3.7.0->deepctr[cpu]) (1.25.2)\n",
            "Requirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (1.4.0)\n",
            "Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (1.6.3)\n",
            "Requirement already satisfied: flatbuffers>=2.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (24.3.25)\n",
            "Requirement already satisfied: gast<=0.4.0,>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (0.4.0)\n",
            "Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (0.2.0)\n",
            "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (1.63.0)\n",
            "Requirement already satisfied: keras<2.11,>=2.10.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (2.10.0)\n",
            "Requirement already satisfied: keras-preprocessing>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (1.1.2)\n",
            "Requirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (18.1.1)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (3.3.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (24.0)\n",
            "Requirement already satisfied: protobuf<3.20,>=3.9.2 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (3.19.6)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (67.7.2)\n",
            "Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (1.16.0)\n",
            "Requirement already satisfied: tensorboard<2.11,>=2.10 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (2.10.1)\n",
            "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (0.37.0)\n",
            "Requirement already satisfied: tensorflow-estimator<2.11,>=2.10.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (2.10.0)\n",
            "Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (2.4.0)\n",
            "Requirement already satisfied: typing-extensions>=3.6.6 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (4.11.0)\n",
            "Requirement already satisfied: wrapt>=1.11.0 in /usr/local/lib/python3.10/dist-packages (from tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (1.14.1)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->deepctr[cpu]) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->deepctr[cpu]) (3.7)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->deepctr[cpu]) (2.0.7)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->deepctr[cpu]) (2024.2.2)\n",
            "Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.10/dist-packages (from astunparse>=1.6.0->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (0.43.0)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (2.27.0)\n",
            "Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (0.4.6)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (3.6)\n",
            "Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (0.6.1)\n",
            "Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (1.8.1)\n",
            "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (3.0.3)\n",
            "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (5.3.3)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (0.4.0)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (4.9)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (1.3.1)\n",
            "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.10/dist-packages (from werkzeug>=1.0.1->tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (2.1.5)\n",
            "Requirement already satisfied: pyasn1<0.7.0,>=0.4.6 in /usr/local/lib/python3.10/dist-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (0.6.0)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.10/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.11,>=2.10->tensorflow!=1.7.*,!=1.8.*,>=1.4.0->deepctr[cpu]) (3.2.2)\n",
            "Installing collected packages: h5py, deepctr\n",
            "  Attempting uninstall: h5py\n",
            "    Found existing installation: h5py 3.9.0\n",
            "    Uninstalling h5py-3.9.0:\n",
            "      Successfully uninstalled h5py-3.9.0\n",
            "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
            "tf-keras 2.15.1 requires tensorflow<2.16,>=2.15, but you have tensorflow 2.10.0 which is incompatible.\u001b[0m\u001b[31m\n",
            "\u001b[0mSuccessfully installed deepctr-0.9.3 h5py-3.7.0\n",
            "Dataset URL: https://www.kaggle.com/datasets/praneshmukhopadhyay/amazon-questionanswer-dataset\n",
            "License(s): CC0-1.0\n",
            "Downloading amazon-questionanswer-dataset.zip to /content\n",
            " 99% 423M/426M [00:03<00:00, 125MB/s]\n",
            "100% 426M/426M [00:03<00:00, 121MB/s]\n",
            "Archive:  /content/amazon-questionanswer-dataset.zip\n",
            "  inflating: multi_answers.csv       \n",
            "  inflating: multi_questions.csv     \n",
            "  inflating: single_qna.csv          \n"
          ]
        }
      ],
      "source": [
        "!pip install tensorflow==2.10\n",
        "!pip install deepctr[cpu]\n",
        "! mkdir ~/.kaggle\n",
        "! cp kaggle.json ~/.kaggle/\n",
        "! chmod 600 ~/.kaggle/kaggle.json\n",
        "! kaggle datasets download praneshmukhopadhyay/amazon-questionanswer-dataset\n",
        "!unzip /content/amazon-questionanswer-dataset.zip"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import LabelEncoder, MinMaxScaler\n",
        "from sklearn.metrics import log_loss, roc_auc_score , mean_squared_error\n",
        "\n",
        "from itertools import chain\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow.python.keras.models import Model\n",
        "from tensorflow.python.keras.layers import Dense, Concatenate ,Flatten ,Lambda\n",
        "\n",
        "from deepctr.feature_column import SparseFeat, DenseFeat, get_feature_names\n",
        "from deepctr.feature_column import build_input_features, get_linear_logit, DEFAULT_GROUP_NAME, input_from_feature_columns\n",
        "from deepctr.layers.core import PredictionLayer, DNN\n",
        "from deepctr.layers.interaction import FM ,InteractingLayer\n",
        "from deepctr.layers.utils import concat_func, add_func, combined_dnn_input\n",
        "\n",
        "from deepctr.feature_column import build_input_features, get_linear_logit, input_from_feature_columns ,VarLenSparseFeat\n",
        "\n",
        "from deepctr.layers.core import PredictionLayer, DNN ,RegulationModule\n",
        "from deepctr.layers.interaction import CrossNet ,BridgeModule\n",
        "from deepctr.layers.utils import add_func, combined_dnn_input\n"
      ],
      "metadata": {
        "id": "3aIlDcpNpcWB"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data = pd.read_csv(\"multi_answers.csv\")\n",
        "data=data.dropna()\n",
        "data"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "nBimWA5fpAcU",
        "outputId": "d7db4d39-cdbd-4788-af9e-3873c9493e7c"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        QuestionID                                         AnswerText  \\\n",
              "2             C1Q1  It's been a while since I installed them, but ...   \n",
              "3             C1Q1           1 pair rear flaps and mounting hardware.   \n",
              "4             C1Q1  I didn't buy these for myself I bought them fo...   \n",
              "5             C1Q1                                                 no   \n",
              "6             C1Q1  Not sure, maybe measuring width of wheels and ...   \n",
              "...            ...                                                ...   \n",
              "4019732   C17Q1183                              To be honest idk srry   \n",
              "4019733   C17Q1183  Yes it should work and to completely honest do...   \n",
              "4019734   C17Q1183  http://www.systemrequirementslab.com/cyri/requ...   \n",
              "4019742   C17Q1183  Yes you can. I personally download from spinti...   \n",
              "4019743   C17Q1183                                          Yes I can   \n",
              "\n",
              "             AnswererID        AnswerTime AnswerType  AnswerScore  \n",
              "2        A3BWPG98KF0TAV     July 19, 2013          ?       0.5428  \n",
              "3                     1     July 19, 2013          ?       0.5565  \n",
              "4        A1MGZTOLD2C0VS     July 19, 2013          ?       0.4623  \n",
              "5        A1GU54G5GCLOGV     July 20, 2013          N       0.0366  \n",
              "6        A2FQI9PZSTWQTO     July 19, 2013          ?       0.1260  \n",
              "...                 ...               ...        ...          ...  \n",
              "4019732  A24GF8XJ81UJCM    March 24, 2015          ?       0.5820  \n",
              "4019733  A3HVTKFMKZD9U8    March 24, 2015          Y       0.9774  \n",
              "4019734  A215AZEU7D23KI    March 25, 2015          ?       0.6033  \n",
              "4019742  A215AZEU7D23KI  January 10, 2015          Y       0.9897  \n",
              "4019743   ARR5TN6Y7R824  January 10, 2015          Y       0.9807  \n",
              "\n",
              "[600736 rows x 6 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-73cf5aca-66c3-460e-b550-fb5f22e35baf\" class=\"colab-df-container\">\n",
              "    <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>QuestionID</th>\n",
              "      <th>AnswerText</th>\n",
              "      <th>AnswererID</th>\n",
              "      <th>AnswerTime</th>\n",
              "      <th>AnswerType</th>\n",
              "      <th>AnswerScore</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>C1Q1</td>\n",
              "      <td>It's been a while since I installed them, but ...</td>\n",
              "      <td>A3BWPG98KF0TAV</td>\n",
              "      <td>July 19, 2013</td>\n",
              "      <td>?</td>\n",
              "      <td>0.5428</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>C1Q1</td>\n",
              "      <td>1 pair rear flaps and mounting hardware.</td>\n",
              "      <td>1</td>\n",
              "      <td>July 19, 2013</td>\n",
              "      <td>?</td>\n",
              "      <td>0.5565</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>C1Q1</td>\n",
              "      <td>I didn't buy these for myself I bought them fo...</td>\n",
              "      <td>A1MGZTOLD2C0VS</td>\n",
              "      <td>July 19, 2013</td>\n",
              "      <td>?</td>\n",
              "      <td>0.4623</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>C1Q1</td>\n",
              "      <td>no</td>\n",
              "      <td>A1GU54G5GCLOGV</td>\n",
              "      <td>July 20, 2013</td>\n",
              "      <td>N</td>\n",
              "      <td>0.0366</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>C1Q1</td>\n",
              "      <td>Not sure, maybe measuring width of wheels and ...</td>\n",
              "      <td>A2FQI9PZSTWQTO</td>\n",
              "      <td>July 19, 2013</td>\n",
              "      <td>?</td>\n",
              "      <td>0.1260</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4019732</th>\n",
              "      <td>C17Q1183</td>\n",
              "      <td>To be honest idk srry</td>\n",
              "      <td>A24GF8XJ81UJCM</td>\n",
              "      <td>March 24, 2015</td>\n",
              "      <td>?</td>\n",
              "      <td>0.5820</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4019733</th>\n",
              "      <td>C17Q1183</td>\n",
              "      <td>Yes it should work and to completely honest do...</td>\n",
              "      <td>A3HVTKFMKZD9U8</td>\n",
              "      <td>March 24, 2015</td>\n",
              "      <td>Y</td>\n",
              "      <td>0.9774</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4019734</th>\n",
              "      <td>C17Q1183</td>\n",
              "      <td>http://www.systemrequirementslab.com/cyri/requ...</td>\n",
              "      <td>A215AZEU7D23KI</td>\n",
              "      <td>March 25, 2015</td>\n",
              "      <td>?</td>\n",
              "      <td>0.6033</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4019742</th>\n",
              "      <td>C17Q1183</td>\n",
              "      <td>Yes you can. I personally download from spinti...</td>\n",
              "      <td>A215AZEU7D23KI</td>\n",
              "      <td>January 10, 2015</td>\n",
              "      <td>Y</td>\n",
              "      <td>0.9897</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4019743</th>\n",
              "      <td>C17Q1183</td>\n",
              "      <td>Yes I can</td>\n",
              "      <td>ARR5TN6Y7R824</td>\n",
              "      <td>January 10, 2015</td>\n",
              "      <td>Y</td>\n",
              "      <td>0.9807</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>600736 rows × 6 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-73cf5aca-66c3-460e-b550-fb5f22e35baf')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-73cf5aca-66c3-460e-b550-fb5f22e35baf button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-73cf5aca-66c3-460e-b550-fb5f22e35baf');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-d96fa442-f114-4505-8ffc-3bfe237dc65c\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-d96fa442-f114-4505-8ffc-3bfe237dc65c')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-d96fa442-f114-4505-8ffc-3bfe237dc65c button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data"
            }
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sparse_features = [\"QuestionID\", \"AnswererID\",\n",
        "                    \"AnswerTime\", \"AnswerType\"]\n",
        "target = ['AnswerScore']\n",
        "\n",
        "for feat in sparse_features:\n",
        "    lbe = LabelEncoder()\n",
        "    data[feat] = lbe.fit_transform(data[feat])\n",
        "fixlen_feature_columns = [SparseFeat(feat, data[feat].max() + 1, embedding_dim=4)\n",
        "                          for feat in sparse_features]\n",
        "linear_feature_columns = fixlen_feature_columns\n",
        "dnn_feature_columns = fixlen_feature_columns\n",
        "feature_names = get_feature_names(linear_feature_columns + dnn_feature_columns)\n",
        "\n",
        "train, test = train_test_split(data, test_size=0.2, random_state=2020)\n",
        "train_model_input = {name: train[name].values for name in feature_names}\n",
        "test_model_input = {name: test[name].values for name in feature_names}\n"
      ],
      "metadata": {
        "id": "5Dp_P3jrqaL7"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Deep and Cross Network\n"
      ],
      "metadata": {
        "id": "25uFSajz1vZ5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def DCN(linear_feature_columns, dnn_feature_columns, cross_num=2, cross_parameterization='vector',\n",
        "        dnn_hidden_units=(256, 128, 64), l2_reg_linear=1e-5, l2_reg_embedding=1e-5,\n",
        "        l2_reg_cross=1e-5, l2_reg_dnn=0, seed=1024, dnn_dropout=0, dnn_use_bn=False,\n",
        "        dnn_activation='relu', task='regression'):\n",
        "\n",
        "    if len(dnn_hidden_units) == 0 and cross_num == 0:\n",
        "        raise ValueError(\"Either hidden_layer or cross layer must > 0\")\n",
        "\n",
        "    features = build_input_features(dnn_feature_columns)\n",
        "    inputs_list = list(features.values())\n",
        "\n",
        "    linear_logit = get_linear_logit(features, linear_feature_columns, seed=seed, prefix='linear',\n",
        "                                    l2_reg=l2_reg_linear)\n",
        "\n",
        "    sparse_embedding_list, dense_value_list = input_from_feature_columns(features, dnn_feature_columns,\n",
        "                                                                         l2_reg_embedding, seed)\n",
        "\n",
        "    dnn_input = combined_dnn_input(sparse_embedding_list, dense_value_list)\n",
        "\n",
        "    if len(dnn_hidden_units) > 0 and cross_num > 0:  # Deep & Cross\n",
        "        deep_out = DNN(dnn_hidden_units, dnn_activation, l2_reg_dnn, dnn_dropout, dnn_use_bn, seed=seed)(dnn_input)\n",
        "        cross_out = CrossNet(cross_num, parameterization=cross_parameterization, l2_reg=l2_reg_cross)(dnn_input)\n",
        "        stack_out = Concatenate()([cross_out, deep_out])\n",
        "        final_logit = Dense(1, use_bias=False)(stack_out)\n",
        "    elif len(dnn_hidden_units) > 0:  # Only Deep\n",
        "        deep_out = DNN(dnn_hidden_units, dnn_activation, l2_reg_dnn, dnn_dropout, dnn_use_bn, seed=seed)(dnn_input)\n",
        "        final_logit = Dense(1, use_bias=False)(deep_out)\n",
        "    elif cross_num > 0:  # Only Cross\n",
        "        cross_out = CrossNet(cross_num, parameterization=cross_parameterization, l2_reg=l2_reg_cross)(dnn_input)\n",
        "        final_logit = Dense(1, use_bias=False)(cross_out)\n",
        "    else:  # Error\n",
        "        raise NotImplementedError\n",
        "\n",
        "    final_logit = add_func([final_logit, linear_logit])\n",
        "    output = PredictionLayer(task)(final_logit)\n",
        "\n",
        "    model = Model(inputs=inputs_list, outputs=output)\n",
        "\n",
        "    return model"
      ],
      "metadata": {
        "id": "SKLsyTS0qua9"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model = DCN(linear_feature_columns, dnn_feature_columns, task='regression')\n",
        "model.compile(\"adam\", \"mse\", metrics=['mse'], )\n",
        "\n",
        "history = model.fit(train_model_input, train[target].values,\n",
        "                    batch_size=256, epochs=10, verbose=2, validation_split=0.2, )\n",
        "pred_ans = model.predict(test_model_input, batch_size=256)\n",
        "print(\"test MSE\", round(mean_squared_error(\n",
        "    test[target].values, pred_ans), 4))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q3TgtKWCqyrV",
        "outputId": "74fb2ad5-c68f-4ac0-cabb-47526988128e"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CrossNet parameterization: vector\n",
            "Epoch 1/10\n",
            "1502/1502 - 72s - loss: 0.0399 - mse: 0.0378 - val_loss: 0.0373 - val_mse: 0.0329\n",
            "Epoch 2/10\n",
            "1502/1502 - 55s - loss: 0.0140 - mse: 0.0098 - val_loss: 0.0401 - val_mse: 0.0353\n",
            "Epoch 3/10\n",
            "1502/1502 - 64s - loss: 0.0227 - mse: 0.0157 - val_loss: 0.0454 - val_mse: 0.0364\n",
            "Epoch 4/10\n",
            "1502/1502 - 53s - loss: 0.0178 - mse: 0.0095 - val_loss: 0.0519 - val_mse: 0.0440\n",
            "Epoch 5/10\n",
            "1502/1502 - 54s - loss: 0.0148 - mse: 0.0076 - val_loss: 0.0438 - val_mse: 0.0369\n",
            "Epoch 6/10\n",
            "1502/1502 - 63s - loss: 0.0137 - mse: 0.0072 - val_loss: 0.0417 - val_mse: 0.0353\n",
            "Epoch 7/10\n",
            "1502/1502 - 56s - loss: 0.0138 - mse: 0.0076 - val_loss: 0.0424 - val_mse: 0.0361\n",
            "Epoch 8/10\n",
            "1502/1502 - 55s - loss: 0.0147 - mse: 0.0086 - val_loss: 0.0449 - val_mse: 0.0384\n",
            "Epoch 9/10\n",
            "1502/1502 - 55s - loss: 0.0150 - mse: 0.0088 - val_loss: 0.0475 - val_mse: 0.0410\n",
            "Epoch 10/10\n",
            "1502/1502 - 63s - loss: 0.0150 - mse: 0.0088 - val_loss: 0.0503 - val_mse: 0.0437\n",
            "test MSE 0.0437\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "plt.style.use('ggplot')\n",
        "\n",
        "def plot_history(history):\n",
        "    mse = history.history['mse']\n",
        "    loss = history.history['loss']\n",
        "    x = range(1, len(mse) + 1)\n",
        "\n",
        "    plt.figure(figsize=(12, 5))\n",
        "    plt.subplot(1, 2, 1)\n",
        "    plt.plot(x, mse, 'b', label='mse')\n",
        "    plt.title('MSE')\n",
        "    plt.legend()\n",
        "    plt.subplot(1, 2, 2)\n",
        "    plt.plot(x, loss, 'b', label='loss')\n",
        "    plt.title('loss')\n",
        "    plt.legend()\n",
        "\n",
        "plot_history(history)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 469
        },
        "id": "dp3GXINX1z4o",
        "outputId": "ea26723e-05d0-457e-ee66-addc375f069e"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def DIFM(linear_feature_columns, dnn_feature_columns,\n",
        "         att_embedding_size=8, att_head_num=8, att_res=True, dnn_hidden_units=(256, 128, 64),\n",
        "         l2_reg_linear=0.00001, l2_reg_embedding=0.00001, l2_reg_dnn=0, seed=1024, dnn_dropout=0,\n",
        "         dnn_activation='relu', dnn_use_bn=False, task='regression'):\n",
        "\n",
        "\n",
        "    if not len(dnn_hidden_units) > 0:\n",
        "        raise ValueError(\"dnn_hidden_units is null!\")\n",
        "\n",
        "    features = build_input_features(\n",
        "        linear_feature_columns + dnn_feature_columns)\n",
        "\n",
        "    sparse_feat_num = len(list(filter(lambda x: isinstance(x, SparseFeat) or isinstance(x, VarLenSparseFeat),\n",
        "                                      dnn_feature_columns)))\n",
        "    inputs_list = list(features.values())\n",
        "\n",
        "    sparse_embedding_list, _ = input_from_feature_columns(features, dnn_feature_columns,\n",
        "                                                          l2_reg_embedding, seed)\n",
        "\n",
        "    if not len(sparse_embedding_list) > 0:\n",
        "        raise ValueError(\"there are no sparse features\")\n",
        "\n",
        "    att_input = concat_func(sparse_embedding_list, axis=1)\n",
        "    att_out = InteractingLayer(att_embedding_size, att_head_num, att_res, scaling=True)(att_input)\n",
        "    att_out = Flatten()(att_out)\n",
        "    m_vec = Dense(sparse_feat_num, use_bias=False)(att_out)\n",
        "\n",
        "    dnn_input = combined_dnn_input(sparse_embedding_list, [])\n",
        "    dnn_output = DNN(dnn_hidden_units, dnn_activation, l2_reg_dnn, dnn_dropout, dnn_use_bn, seed=seed)(dnn_input)\n",
        "    m_bit = Dense(sparse_feat_num, use_bias=False)(dnn_output)\n",
        "\n",
        "    input_aware_factor = add_func([m_vec, m_bit])  # the complete input-aware factor m_x\n",
        "\n",
        "    linear_logit = get_linear_logit(features, linear_feature_columns, seed=seed, prefix='linear',\n",
        "                                    l2_reg=l2_reg_linear, sparse_feat_refine_weight=input_aware_factor)\n",
        "\n",
        "    fm_input = concat_func(sparse_embedding_list, axis=1)\n",
        "    refined_fm_input = Lambda(lambda x: x[0] * tf.expand_dims(x[1], axis=-1))(\n",
        "        [fm_input, input_aware_factor])\n",
        "    fm_logit = FM()(refined_fm_input)\n",
        "\n",
        "    final_logit = add_func([linear_logit, fm_logit])\n",
        "\n",
        "    output = PredictionLayer(task)(final_logit)\n",
        "    model = Model(inputs=inputs_list, outputs=output)\n",
        "    return model"
      ],
      "metadata": {
        "id": "aqtOrtU_txkb"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model = DIFM(linear_feature_columns, dnn_feature_columns, task='regression')\n",
        "model.compile(\"adam\", \"mse\",\n",
        "              metrics=['accuracy'], )\n",
        "\n",
        "history = model.fit(train_model_input, train[target].values,\n",
        "                    batch_size=256, epochs=10, verbose=2, validation_split=0.2, )\n",
        "pred_ans = model.predict(test_model_input, batch_size=256)\n",
        "print(\"test MSE\", round(mean_squared_error(\n",
        "    test[target].values, pred_ans), 4))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fr5GRifht59I",
        "outputId": "d652fe4d-ab51-4104-af45-9e8e6be09cf8"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "1502/1502 - 82s - loss: 0.0469 - accuracy: 2.6010e-06 - val_loss: 0.0371 - val_accuracy: 1.0404e-05\n",
            "Epoch 2/10\n",
            "1502/1502 - 71s - loss: 0.0135 - accuracy: 2.6010e-06 - val_loss: 0.0392 - val_accuracy: 1.0404e-05\n",
            "Epoch 3/10\n",
            "1502/1502 - 76s - loss: 0.0158 - accuracy: 2.6010e-06 - val_loss: 0.0425 - val_accuracy: 1.0404e-05\n",
            "Epoch 4/10\n",
            "1502/1502 - 66s - loss: 0.0147 - accuracy: 2.6010e-06 - val_loss: 0.0473 - val_accuracy: 1.0404e-05\n",
            "Epoch 5/10\n",
            "1502/1502 - 77s - loss: 0.0147 - accuracy: 2.6010e-06 - val_loss: 0.0417 - val_accuracy: 1.0404e-05\n",
            "Epoch 6/10\n",
            "1502/1502 - 82s - loss: 0.0130 - accuracy: 2.6010e-06 - val_loss: 0.0415 - val_accuracy: 1.0404e-05\n",
            "Epoch 7/10\n",
            "1502/1502 - 68s - loss: 0.0133 - accuracy: 2.6010e-06 - val_loss: 0.0422 - val_accuracy: 1.0404e-05\n",
            "Epoch 8/10\n",
            "1502/1502 - 72s - loss: 0.0138 - accuracy: 2.6010e-06 - val_loss: 0.0416 - val_accuracy: 1.0404e-05\n",
            "Epoch 9/10\n",
            "1502/1502 - 64s - loss: 0.0131 - accuracy: 2.6010e-06 - val_loss: 0.0415 - val_accuracy: 1.0404e-05\n",
            "Epoch 10/10\n",
            "1502/1502 - 67s - loss: 0.0125 - accuracy: 2.6010e-06 - val_loss: 0.0409 - val_accuracy: 1.0404e-05\n",
            "test MSE 0.0349\n"
          ]
        }
      ]
    }
  ]
}