cuisine_prediction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "DME project - cuisine prediction.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "source": [
        "# DME project - Classification models\n",
        "This notebook contains the classification models done for predicting the cuisines recipes using the `recipes.csv` data file from Bellosi (2011)"
      ],
      "cell_type": "markdown",
      "metadata": {}
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Install required packages\n",
        "!pip install -q -r requirements.txt"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LTXvV9CP6EUS"
      },
      "source": [
        "# Data libraries\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "# Tensorflow modules\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras import layers\n",
        "\n",
        "# sklearn modules\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.decomposition import PCA\n",
        "\n",
        "# Plotting libraries\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "# misc libraries\n",
        "import time\n",
        "import random"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yo4GmDd5zLv1",
        "outputId": "93aeb9eb-bea6-4c3f-dded-a94ce2b10dce"
      },
      "source": [
        "# umap\n",
        "!pip install umap-learn"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting umap-learn\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/75/69/85e7f950bb75792ad5d666d86c5f3e62eedbb942848e7e3126513af9999c/umap-learn-0.5.1.tar.gz (80kB)\n",
            "\u001b[K     |████████████████████████████████| 81kB 6.6MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.7/dist-packages (from umap-learn) (1.19.5)\n",
            "Requirement already satisfied: scikit-learn>=0.22 in /usr/local/lib/python3.7/dist-packages (from umap-learn) (0.22.2.post1)\n",
            "Requirement already satisfied: scipy>=1.0 in /usr/local/lib/python3.7/dist-packages (from umap-learn) (1.4.1)\n",
            "Requirement already satisfied: numba>=0.49 in /usr/local/lib/python3.7/dist-packages (from umap-learn) (0.51.2)\n",
            "Collecting pynndescent>=0.5\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/af/65/8189298dd3a05bbad716ee8e249764ff8800e365d8dc652ad2192ca01b4a/pynndescent-0.5.2.tar.gz (1.1MB)\n",
            "\u001b[K     |████████████████████████████████| 1.2MB 21.9MB/s \n",
            "\u001b[?25hRequirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.22->umap-learn) (1.0.1)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.7/dist-packages (from numba>=0.49->umap-learn) (54.2.0)\n",
            "Requirement already satisfied: llvmlite<0.35,>=0.34.0.dev0 in /usr/local/lib/python3.7/dist-packages (from numba>=0.49->umap-learn) (0.34.0)\n",
            "Building wheels for collected packages: umap-learn, pynndescent\n",
            "  Building wheel for umap-learn (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for umap-learn: filename=umap_learn-0.5.1-cp37-none-any.whl size=76569 sha256=36d8bf0ae72587c120ef9f85f4f869b3a389dac49897d6c4ad66dcd666d412c0\n",
            "  Stored in directory: /root/.cache/pip/wheels/ad/df/d5/a3691296ff779f25cd1cf415a3af954b987fb53111e3392cf4\n",
            "  Building wheel for pynndescent (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for pynndescent: filename=pynndescent-0.5.2-cp37-none-any.whl size=51351 sha256=e33be43aae60e28d7707d868d527a9fc04dff12632bedb4abc7725f641c33da8\n",
            "  Stored in directory: /root/.cache/pip/wheels/ba/52/4e/4c28d04d144a28f89e2575fb63628df6e6d49b56c5ddd0c74e\n",
            "Successfully built umap-learn pynndescent\n",
            "Installing collected packages: pynndescent, umap-learn\n",
            "Successfully installed pynndescent-0.5.2 umap-learn-0.5.1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YhB8djeP0sYA"
      },
      "source": [
        "import umap.umap_ as umap"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pgNHPLjKnO_4"
      },
      "source": [
        "---\n",
        "# Data import and preparation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fq4gKwabV7vR",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303
        },
        "outputId": "ab80ca86-08cb-451f-85d2-44a6fa52c908"
      },
      "source": [
        "recipes_df = pd.read_csv('recipes.csv')\n",
        "recipes_df.head()"
      ],
      "execution_count": 5,
      "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>'acorn squash'</th>\n",
              "      <th>adobo</th>\n",
              "      <th>'african birdseye chile pepper'</th>\n",
              "      <th>ale</th>\n",
              "      <th>'aleppo pepper'</th>\n",
              "      <th>'alfalfa sprouts'</th>\n",
              "      <th>'alfredo sauce'</th>\n",
              "      <th>allspice</th>\n",
              "      <th>almond</th>\n",
              "      <th>'almond butter'</th>\n",
              "      <th>amaretti</th>\n",
              "      <th>'amchoor powder'</th>\n",
              "      <th>'ancho chile pepper'</th>\n",
              "      <th>anchovy</th>\n",
              "      <th>'anchovy paste'</th>\n",
              "      <th>'andouille sausage'</th>\n",
              "      <th>anise</th>\n",
              "      <th>apple</th>\n",
              "      <th>'apple butter'</th>\n",
              "      <th>'apple juice'</th>\n",
              "      <th>applesauce</th>\n",
              "      <th>apricot</th>\n",
              "      <th>'arborio rice'</th>\n",
              "      <th>'argan oil'</th>\n",
              "      <th>arrowroot</th>\n",
              "      <th>artichoke</th>\n",
              "      <th>'artificial sweetener'</th>\n",
              "      <th>arugula</th>\n",
              "      <th>asadero</th>\n",
              "      <th>asafetida</th>\n",
              "      <th>'asiago cheese'</th>\n",
              "      <th>asparagus</th>\n",
              "      <th>'assorted vegetables'</th>\n",
              "      <th>avocado</th>\n",
              "      <th>'azuki bean'</th>\n",
              "      <th>'baby carrot'</th>\n",
              "      <th>'baby corn'</th>\n",
              "      <th>'baby onions'</th>\n",
              "      <th>'baby spinach'</th>\n",
              "      <th>bacon</th>\n",
              "      <th>...</th>\n",
              "      <th>vermouth</th>\n",
              "      <th>'Vidalia onion'</th>\n",
              "      <th>vinegar</th>\n",
              "      <th>'vital wheat gluten'</th>\n",
              "      <th>vodka</th>\n",
              "      <th>walnut</th>\n",
              "      <th>wasabi</th>\n",
              "      <th>'washed rind cheese'</th>\n",
              "      <th>water</th>\n",
              "      <th>'water chestnut'</th>\n",
              "      <th>watercress</th>\n",
              "      <th>watermelon</th>\n",
              "      <th>'wheat bread'</th>\n",
              "      <th>'whipping cream'</th>\n",
              "      <th>'white bean'</th>\n",
              "      <th>'white bread'</th>\n",
              "      <th>'white cheese'</th>\n",
              "      <th>'white fish'</th>\n",
              "      <th>'white onion'</th>\n",
              "      <th>'white peppercorns'</th>\n",
              "      <th>'white sesame seeds'</th>\n",
              "      <th>'whole grain mustard'</th>\n",
              "      <th>'whole milk'</th>\n",
              "      <th>'wild rice'</th>\n",
              "      <th>wine</th>\n",
              "      <th>'wine vinegar'</th>\n",
              "      <th>'winter squash'</th>\n",
              "      <th>'worcestershire sauce'</th>\n",
              "      <th>'xanthan gum'</th>\n",
              "      <th>yams</th>\n",
              "      <th>yeast</th>\n",
              "      <th>'yellow curry paste'</th>\n",
              "      <th>'yellow food coloring'</th>\n",
              "      <th>'yellow split pea'</th>\n",
              "      <th>'yellow squash'</th>\n",
              "      <th>yogurt</th>\n",
              "      <th>zaatar</th>\n",
              "      <th>zest</th>\n",
              "      <th>zucchini</th>\n",
              "      <th>cuisine</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <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>0</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",
              "      <td>0</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>0</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",
              "      <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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 710 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   'acorn squash'  adobo  ...  zucchini  cuisine\n",
              "0               0      0  ...         0        0\n",
              "1               0      0  ...         0        0\n",
              "2               0      0  ...         0        0\n",
              "3               0      0  ...         0        0\n",
              "4               0      0  ...         0        0\n",
              "\n",
              "[5 rows x 710 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0xnNfLznAXUl",
        "outputId": "6be4ee32-452d-487b-dbcb-0f485da293c6"
      },
      "source": [
        "recipes_df.cuisine.value_counts()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "11    353\n",
              "7     353\n",
              "3     353\n",
              "10    353\n",
              "6     353\n",
              "2     353\n",
              "9     353\n",
              "5     353\n",
              "1     353\n",
              "8     353\n",
              "4     353\n",
              "0     353\n",
              "Name: cuisine, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dpHXEOiXQpDx",
        "outputId": "6977369e-bdf6-412e-d96b-83b551fecdd9"
      },
      "source": [
        "recipes_df.info()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 4236 entries, 0 to 4235\n",
            "Columns: 710 entries, 'acorn squash' to cuisine\n",
            "dtypes: int64(710)\n",
            "memory usage: 22.9 MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V-JSHeInGbK9"
      },
      "source": [
        "### Clean column names"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 251
        },
        "id": "Kzj3G97IF-Yf",
        "outputId": "46027670-273e-4721-efc5-82b058e77182"
      },
      "source": [
        "for column in recipes_df.columns:\n",
        "  recipes_df.rename(columns={column:column.replace(' ', '_').replace(\"'\", \"\")}, inplace=True)\n",
        "  recipes_df.rename(columns={column:column.lower()}, inplace=True)\n",
        "\n",
        "recipes_df.head()"
      ],
      "execution_count": 8,
      "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>acorn_squash</th>\n",
              "      <th>adobo</th>\n",
              "      <th>african_birdseye_chile_pepper</th>\n",
              "      <th>ale</th>\n",
              "      <th>aleppo_pepper</th>\n",
              "      <th>alfalfa_sprouts</th>\n",
              "      <th>alfredo_sauce</th>\n",
              "      <th>allspice</th>\n",
              "      <th>almond</th>\n",
              "      <th>almond_butter</th>\n",
              "      <th>amaretti</th>\n",
              "      <th>amchoor_powder</th>\n",
              "      <th>ancho_chile_pepper</th>\n",
              "      <th>anchovy</th>\n",
              "      <th>anchovy_paste</th>\n",
              "      <th>andouille_sausage</th>\n",
              "      <th>anise</th>\n",
              "      <th>apple</th>\n",
              "      <th>apple_butter</th>\n",
              "      <th>apple_juice</th>\n",
              "      <th>applesauce</th>\n",
              "      <th>apricot</th>\n",
              "      <th>arborio_rice</th>\n",
              "      <th>argan_oil</th>\n",
              "      <th>arrowroot</th>\n",
              "      <th>artichoke</th>\n",
              "      <th>artificial_sweetener</th>\n",
              "      <th>arugula</th>\n",
              "      <th>asadero</th>\n",
              "      <th>asafetida</th>\n",
              "      <th>asiago_cheese</th>\n",
              "      <th>asparagus</th>\n",
              "      <th>assorted_vegetables</th>\n",
              "      <th>avocado</th>\n",
              "      <th>azuki_bean</th>\n",
              "      <th>baby_carrot</th>\n",
              "      <th>baby_corn</th>\n",
              "      <th>baby_onions</th>\n",
              "      <th>baby_spinach</th>\n",
              "      <th>bacon</th>\n",
              "      <th>...</th>\n",
              "      <th>vermouth</th>\n",
              "      <th>Vidalia_onion</th>\n",
              "      <th>vinegar</th>\n",
              "      <th>vital_wheat_gluten</th>\n",
              "      <th>vodka</th>\n",
              "      <th>walnut</th>\n",
              "      <th>wasabi</th>\n",
              "      <th>washed_rind_cheese</th>\n",
              "      <th>water</th>\n",
              "      <th>water_chestnut</th>\n",
              "      <th>watercress</th>\n",
              "      <th>watermelon</th>\n",
              "      <th>wheat_bread</th>\n",
              "      <th>whipping_cream</th>\n",
              "      <th>white_bean</th>\n",
              "      <th>white_bread</th>\n",
              "      <th>white_cheese</th>\n",
              "      <th>white_fish</th>\n",
              "      <th>white_onion</th>\n",
              "      <th>white_peppercorns</th>\n",
              "      <th>white_sesame_seeds</th>\n",
              "      <th>whole_grain_mustard</th>\n",
              "      <th>whole_milk</th>\n",
              "      <th>wild_rice</th>\n",
              "      <th>wine</th>\n",
              "      <th>wine_vinegar</th>\n",
              "      <th>winter_squash</th>\n",
              "      <th>worcestershire_sauce</th>\n",
              "      <th>xanthan_gum</th>\n",
              "      <th>yams</th>\n",
              "      <th>yeast</th>\n",
              "      <th>yellow_curry_paste</th>\n",
              "      <th>yellow_food_coloring</th>\n",
              "      <th>yellow_split_pea</th>\n",
              "      <th>yellow_squash</th>\n",
              "      <th>yogurt</th>\n",
              "      <th>zaatar</th>\n",
              "      <th>zest</th>\n",
              "      <th>zucchini</th>\n",
              "      <th>cuisine</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <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>0</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",
              "      <td>0</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>0</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",
              "      <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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</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",
              "      <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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</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",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 710 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   acorn_squash  adobo  african_birdseye_chile_pepper  ...  zest  zucchini  cuisine\n",
              "0             0      0                              0  ...     0         0        0\n",
              "1             0      0                              0  ...     0         0        0\n",
              "2             0      0                              0  ...     0         0        0\n",
              "3             0      0                              0  ...     0         0        0\n",
              "4             0      0                              0  ...     0         0        0\n",
              "\n",
              "[5 rows x 710 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vVB6tOYJGp5J"
      },
      "source": [
        "## Create train, validation and test split"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1eMEY2RJPo1i",
        "outputId": "53280b9d-1c31-49ec-f571-51937d275741"
      },
      "source": [
        "train, val = train_test_split(recipes_df, test_size=0.3, random_state=42, stratify=recipes_df['cuisine'])\n",
        "val, test = train_test_split(val, test_size=(1/3), random_state=42, stratify=val['cuisine'])\n",
        "print(len(train), 'train examples')\n",
        "print(len(val), 'validation examples')\n",
        "print(len(test), 'test examples')\n",
        "print(f'Percentage of train examples: {round(len(train) / len(recipes_df)*100)}%')\n",
        "print(f'Percentage of validation examples: {round(len(val) / len(recipes_df)*100)}%')\n",
        "print(f'Percentage of test examples: {round(len(test) / len(recipes_df)*100)}%')"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2965 train examples\n",
            "847 validation examples\n",
            "424 test examples\n",
            "Percentage of train examples: 70%\n",
            "Percentage of validation examples: 20%\n",
            "Percentage of test examples: 10%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gGn7UZ4yuy0K"
      },
      "source": [
        "## Dimensionally reduced versions of dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jZedzUAwOm19"
      },
      "source": [
        "y_train = train.pop('cuisine')\n",
        "X_train = train\n",
        "y_val = val.pop('cuisine')\n",
        "X_val = val\n",
        "y_test = test.pop('cuisine')\n",
        "X_test = test"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hfj8f2Rou2f2"
      },
      "source": [
        "### PCA"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mTsOgddgUTqW"
      },
      "source": [
        "# PCA reduced version of train, valid and test data\n",
        "pca = PCA(n_components=71)\n",
        "pca.fit(X_train)\n",
        "X_train_PCA = pca.transform(X_train)\n",
        "X_val_PCA = pca.transform(X_val)\n",
        "X_test_PCA = pca.transform(X_test)"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "icIdSDrM_4bj",
        "outputId": "7f74cd3f-51a2-4c50-8bc0-1bb3fd4464dc"
      },
      "source": [
        "plt.plot(np.cumsum(pca.explained_variance_ratio_))\n",
        "plt.xlabel('number of components')\n",
        "plt.ylabel('cumulative explained variance');"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEGCAYAAABo25JHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd3yV9fn/8ddFIGwIeyRsghg2hOXGiQu3gqu2bsXZ9lfbfqvW77dD22pta7WouOreqDhwgIuRhL0JgUBYCYQRCJnn+v1x39FjDMkdyJ37nJzr+XjkkXOPc5/3CYf7Ovf4fD6iqhhjjIldjYIOYIwxJlhWCIwxJsZZITDGmBhnhcAYY2KcFQJjjIlxjYMOUFsdO3bU3r17Bx3DGGOiSkZGxk5V7VTVsqgrBL179yY9PT3oGMYYE1VEJPtQy+zUkDHGxDgrBMYYE+OsEBhjTIyzQmCMMTHOCoExxsQ4KwTGGBPjrBAYY0yMi7p2BMYYE0v2FJawePMeFm/ew6lHd2FwYts6fw0rBMYYEyFKy0Os2V7Aok27WbTJ2fln7TwAgAh0aNXUCoExxjQkO/cXszB7NxmbdrMoew9Lt+yhqDQEQMdWTRnRM4GLRiUxokcCQ5La0rpZE19yWCEwxph6EAop63L3k5G9m/TsfBZm72bjrkIAmsQJg7q35fIxvRjRM4ERPRNITGiOiNRLNisExhjjg4Ml5SzJ2UNG9m7SNjo7/n1FZQB0aBnPqF7tmDKmJ6N6tWNwYluaNYkLLKsVAmOMqQN7CktI2+js9BdsyGfF1r2Uljtjwid3bsXZQ7sxqld7Unu1o1eHFvX2bd8LKwTGGHMYdu4vZn5WPgs27GL+hnzW7ChAFeIbN2JYUluuO74vo3u3Y2TPdiS0iA86brV8LQQiMhF4FIgDnlLVP1da/ggwwZ1sAXRW1QQ/MxljzOHILShiflY+87KcHX9m7n4AWsTHMapXO84e0o2xfTswNCnY0zyHw7dCICJxwGPAaUAOkCYiM1R1ZcU6qnpX2Pq3ASP8ymOMMbWRu6+IeRucHf+8rF1k5Tm3cbaMj2N0n/ZcNDKJsX3bMySxLU3iorttrp9HBGOATFXNAhCRV4DzgJWHWH8KcJ+PeYwx5pB27i9mXtYu5q53dvzr3R1/q6aNGdOnPZel9mBc3w4M6t6GxlG+46/Mz0KQCGwOm84Bxla1ooj0AvoAnx9i+Q3ADQA9e/as25TGmJi0r6iU+Vn5fLt+J3PX72L19gLA2fGP7t2OS1N7ML5fB1K6Nbwdf2WRcrF4MvCGqpZXtVBVpwHTAFJTU7U+gxljGoaSshCLNu3mm8ydfJ25kyU5eykPKc2aNGJ07/ZMGt6d8X07MCSxbYPf8VfmZyHYAvQIm05y51VlMnCrj1mMMTFGVVm7Yz9frcvjm8ydzN+QT2FJOY0EhvVI4JaT+nFMv46M7JVA08bRdXG3rvlZCNKAZBHpg1MAJgOXV15JRAYC7YC5PmYxxsSAnfuL+XrdTr5cl8dX63aSV1AMQN+OLbl4VBLH9OvI+H4daNvcn64aopVvhUBVy0RkKvAxzu2j01V1hYg8AKSr6gx31cnAK6pqp3yMMbVSWh4iI3s3c9bm8eXaPFZs3QdA+5bxHNu/I8cnd+TY/h1JTGgecNLIJtG2/01NTdX09PSgYxhjArJlz0Fmr8llzpo8vl2/i/3FZTRuJIzs1Y4TB3Ti+OSODO7elkaNIqflbiQQkQxVTa1qWaRcLDbGmCoVl5WTtmE3s9fkMntt3ncNuRITmjNpeHdOHNCJY/p18K1nzlhghcAYE3EqvvV/sTqPb9fvpLCknPi4Rozt257Jo3tw0lGd6NepVUT11xPNrBAYYwJXWh5iYfZuPl+Tyxerc1m74/tv/ReOTOSkAZ05pn8HWsTbLssP9lc1xgQi/0AJc9bm8tmqXOaszaOgyDnXP6ZPey4Z1YMJA+1bf32xQmCMqReqzsAsn67aweerclm4aTchdUbimjioKycP7MxxyR3tXH8ArBAYY3xTWh4ibUM+s1bt4NNVO9icfxCAwYltmHpyMqcM7MyQRLvDJ2hWCIwxdepAcRlfrs3j4xXb+Xx1LvuKyohv3Ijj+nfkphP7ccrALnRt2yzomCaMFQJjzBHbU1jCrJU7+Gj5dr7K3ElJWYh2LZpw+qCunJbSheOTO9qF3ghm/zLGmMOya38xn6zcwcxl25i7fhdlISUxoTlXjO3JGYO6ktqrXcx13hatPBUCt5voZFX9VESaA41VtcDfaMaYSJNbUMTHK3bw4bJtzMvaRUihV4cWXHd8X84c3JWhSW3tLp8oVGMhEJHrccYCaA/0w+lF9AngFH+jGWMiwY59RXy4bBszl28nbWM+qtC3U0tundCfiYO7ktKtje38o5yXI4JbcUYbmw+gqutEpLOvqYwxgdqxr4iZy7Yxc9k20rN3owoDurTi9pOTOXtoN5I72/39DYmXQlCsqiUV/+gi0hiIrp7qjDE1yt1XxIfLt/PB0m2kZTvf/Ad2bc1dpw7grCFd6d+5ddARjU+8FII5IvIboLmInAbcArznbyxjTH3IP1DCh8u38d6Srczf4Oz8j+pSsfPvRv/OrYKOaOqBl0JwD3AtsAy4EZgJPOVnKGOMf/YeLOWTFdt5b+k2vsncSXlI6depJXecksw5Q7vZN/8Y5KUQNMcZVOZJABGJc+cV+hnMGFN3CkvK+HRVLu8t2cqcNXmUlIfo0b45N57Ql3OHdWdg19Z2zj+GeSkEnwGnAvvd6ebAJ8AxfoUyxhy58pDyTeZO3l60hY+Wb+dgaTld2jTlqvG9OHdYd4bZrZ7G5aUQNFPViiKAqu4XkRY+ZjLGHIGVW/fx9qIc3l28ldyCYto0a8z5I7pz3vBERvduT5z162Mq8VIIDojISFVdCCAio4CD/sYyxtRG7r4i3l28lTcX5rB6ewFN4oSTjurMhSMSmTCwM82axAUd0UQwL4XgTuB1EdkKCNAVuMzXVMaYGhWVljNr5Q7eyMjhq3V5hBSG90jggfMGcc7Q7rRvGR90RBMlaiwEqpomIgOBo9xZa1S11MvGRWQi8CgQBzylqn+uYp1Lgftx2iYsUdXLPWY3JuaoKkty9vJGxmZmLN7KvqIyurdtxs0n9ePCkUn062S3e5ra89rp3Gigt7v+SBFBVZ+v7gnu3UWPAacBOUCaiMxQ1ZVh6yQDvwaOVdXd1mLZmKrtLSzlncVbeHnBJlZvL6Bp40acObgrl6T2YHzfDtafvzkiXvoaegGnj6HFQLk7W4FqCwFOtxSZqprlbucV4DxgZdg61wOPqepuAFXNrVV6YxowVSU9ezcvz9/EB8u2UVwWYkhiW/5wwWDOHdadNjaSl6kjXo4IUoEUVa1ttxKJwOaw6RxgbKV1BgCIyDc4p4/uV9WPKm9IRG7A6fiOnj171jKGMdFlb2Epby7M4eUFm1iXu59WTRtzSWoSk0f3ZHBi26DjmQbISyFYjnOBeJtPr58MnITTq+mXIjJEVfeEr6Sq04BpAKmpqdbPkWlwVJVFm/fw33nZfLDU+fY/rEcCD140hHOHdbdBXYyvvHy6OgIrRWQBUFwxU1Un1fC8LUCPsOkkd164HGC+e/F5g4isxSkMaR5yGRP1CkvKeHfxVv47L5sVW/fRMj6Oi0clcfnYngzqbt/+Tf3wUgjuP8xtpwHJItIHpwBMBirfEfQOMAV4RkQ64pwqyjrM1zMmamTl7eeFedm8kZFDQVEZA7u25v/OH8z5IxJp1dS+/Zv65eX20TmHs2FVLRORqcDHOOf/p6vqChF5AEhX1RnustNFZCXOhehfququw3k9YyJdeUiZvSaX5+Zm8+XaPJrECWcO7sbV43sxqlc76+7BBEZqugYsIuOAfwJHA/E4O/UDqtrG/3g/lpqaqunp6UG8tDGHpaColNfTc3j2241syi+kS5umXDG2F5PH9KBz62ZBxzMxQkQyVDW1qmVejkH/hXNa53WcO4iuxr3bxxhzaJvzC3nmm428lr6Z/cVljOrVjv838SjOGNSVJjaou4kgnk5GqmqmiMSpajnO+fxFOA3BjDGVZGTn89RXG/h4xXYaiXDO0G789Ng+DOuREHQ0Y6rkpRAUikg8sFhEHsK5jdS+zhgTpjykfLR8O099ncWiTXto27wJN53Yj6vH96ZrWzv9YyKbl0JwFc51ganAXTi3hF7kZyhjokVRaTlvLdzCf75cT/auQnp1aMED5w3i4lFJdu+/iRpe7hrKdh8eBH7vbxxjosP+4jJenJfN019vILegmKFJbXniypGcltLV+vs3UeeQhUBEXlPVS0VkGU7fQj+gqkN9TWZMBNp9oIRnvt3Is99sYF9RGcf278Ajlw3nmH4d7PZPE7WqOyK4w/19Tn0EMSaS5e4r4smvsnhx/iYKS8o5PaULt0zoz3C7AGwagEMWAlXd5nYl/ayqTqjHTMZEjB37inh89npeWrCJsvIQk4Z155YJ/RnQpXXQ0YypM9VeI1DVchEJiUhbVd1bX6GMCVp4ASgPKRePTOKWCf3o1aFl0NGMqXNebmvYDywTkVnAgYqZqnq7b6mMCUjuviL+XakA3DqhPz07tAg6mjG+8VII3nJ/jGmwdu4v5onZ63lhXjZlIeWikYlMnZBsBcDEBC+3jz5XH0GMCcKewhKemJPFc99upLisnAtGJHH7Kf3tFJCJKV6GqkwG/gSkAN81kVTVvj7mMsZX+4vLePqrDTz1VRb7S8qYNKw7d5ySTF8b/N3EIC+nhp4B7gMeASYAP8W6mDBRqqi0nBfmZvP4nPXkHyjhjEFduPu0oziqq90FZGKXl0LQXFU/ExFxWxnfLyIZwL0+ZzOmzpSVh3g9I4dHP13H9n1FHJ/ckV+cfpR1BGcM3gpBsYg0Ata5A81sAez42USFUEiZuXwbf/tkLRt2HmBEzwQeuWw44/t1CDqaMRHDSyG4A2gB3A78L87poZ/4GcqYuvDt+p38aeZqlm3Zy1FdWvPk1amcenRn6wrCmEq8FIJyVd2P057gpz7nMeaIrd6+jz9/uJrZa/Lo3rYZf7tkGOePSLTO4Iw5BC+F4G8i0hV4A3hVVZf7nMmYw7JjXxF//XgNbyzMoXXTxvzmrIFcPb43zZrEBR3NmIjmpR3BBLcQXAr8R0Ta4BSE//M9nTEeFJaU8eSXG3hiznrKQiGuPbYPU0/uT0KL+KCjGRMVPN0GqqrbVfUfwE3AYjzeMSQiE0VkjYhkisg9VSy/RkTyRGSx+3NdrdKbmBYKKW9k5DDhr7N55NO1TBjYiU/vPpH/OSfFioAxteClQdnRwGU4o5LtAl4Ffu7heXHAY8BpQA6QJiIzVHVlpVVfVdWptQ1uYtvizXu4b8YKlmzew7AeCTx2+UhSe7cPOpYxUcnLNYLpwCvAGaq6tRbbHgNkqmoWgIi8ApwHVC4ExniWV1DMQx+t5vWMHDq1bsrDlw7j/OGJNLILwcYcNi/XCMYf5rYTgc1h0znA2CrWu0hETgDWAnep6uYq1jExrrQ8xPNzs/n7rLUUlZVz4wl9ue2UZFo1tXGBjTlSQf8veg94WVWLReRG4Dng5MoricgNwA0APXv2rN+EJnDzsnZx37srWLOjgBMGdOK+c1PoZ30CGVNn/CwEW4AeYdNJ7rzvqOqusMmngIeq2pCqTgOmAaSmpv5o/GTTMO3YV8QfZ67i3cVbSUxozn+uGsXpKV2sQZgxdczPQpAGJItIH5wCMBm4PHwFEemmqtvcyUnAKh/zmChRVh7iubnZPDJrLSXlIW4/uT83n9Sf5vHWHsAYPxyyEIjIe8Ahv32r6qTqNqyqZW7fRB8DccB0VV0hIg8A6ao6A7hdRCYBZUA+cE3t34JpSDKy8/nt28tZvb2AEwd04veTBtG7o40NYIyfRLXqfb2InOg+vBDoCvzXnZ4C7FDVu/yP92Opqamanp4exEsbH+0pLOGPM1fxWnoO3ds2495zUzhjUFc7DWRMHRGRDFVNrWrZIY8IVHWO++S/VXryeyJie2JTJ1SV95Zu44H3VrCnsJQbT+zLHack0yI+6PsYjIkdXv63tRSRvmHtAfoAdqxujtiWPQf53TvL+Xx1LsOS2vLCtWM5uluboGMZE3O8FIK7gNkikgUI0Au40ddUpkELhZQX52fz5w9XE1L43TkpXHNMb+sd1JiAeGlQ9pE7bvFAd9ZqVS32N5ZpqLJ3HeBXby5lXlY+xyd35I8XDKFH+xZBxzImpnnpa6gFcDfQS1WvF5FkETlKVd/3P55pKEIh5fm5G3nwozU0biQ8eNEQLk3tYReDjYkAXgevzwAquprYArwOWCEwnmzZc5C7X13M/A35nDigE3+6cAjdE5oHHcsY4/JSCPqp6mUiMgVAVQvFvsYZj2Yu28Y9by6lPKQ8dNFQLklNsqMAYyKMl0JQIiLNcRuXiUg/wK4RmGodKC7jgfdW8mr6ZoYlteXRySOsYZgxEcpLIbgP+AjoISIvAsdiLYBNNZZv2cvtLy9iw64D3DqhH3eeOoAmcZ7GQDLGBMDLXUOzRGQhMA7n9tE7VHWn78lM1AmFlKe/3sBDH6+mQ8umvHTdOMb36xB0LGNMDbw232wG7HbXTxERVPVL/2KZaJNbUMTPX1vCV+t2csagLjx40VAbLtKYKOHl9tEHcYaqXAGE3NkKWCEwAMxek8vPX1vCgZIy/nDBYC4f09MuCBsTRbwcEZwPHGWNyExlZeUhHp61ln/PXs/Arq15Zco4kru0DjqWMaaWvBSCLKAJdqeQCbNjXxG3vbyIBRvymTKmB/edO4hmTWy8AGOikZdCUAgsFpHPCCsGqnq7b6lMRPtqXR53vrKYg6Xl/P2y4Zw/IjHoSMaYI+ClEMxwf0yMC4WUf36eyd8/W8uAzq157IqR9O9sYwcbE+283D76XH0EMZFt94ES7nptMbPX5HHhiET+cMEQGzrSmAaiuqEqX1PVS0VkGVUMWamqQ31NZiLGks17uOXFheQVFNtdQcY0QNUdEdzh/j6nPoKYyPTKgk3c++4KOrVuyhs3j2doUkLQkYwxday6oSq3ub+z6y+OiRTFZeXcP2MlLy/YxPHJHfnH5BG0a2kNxIxpiLw0KBsH/BM4GogH4oADqmpjCjZQO/YVcdN/M1i0aQ83n9SPX5x+lI0eZkwD5qUnsH8BU4B1QHPgOuAxLxsXkYkiskZEMkXknmrWu0hEVERSvWzX+CcjO5+z//E1a7YX8PgVI/nVxIFWBIxp4Dx1CamqmUCcqpar6jPAxJqeIyJxOAXjTCAFmCIiKVWs1xrnesT82gQ3de/tRTlMmTafVk3jeOfWYzlzSLegIxlj6oGnBmUiEo/TqOwhYBveCsgYIFNVswBE5BXgPGBlpfX+F3gQ+KXn1KZOhULKw7PW8q8vMhnftwOPXznSOowzJoZ42aFfhXNdYCpwAOgBXOTheYnA5rDpHHfed0RkJNBDVT+obkMicoOIpItIel5enoeXNl4dLCln6ssL+dcXmUwe3YPnfjbGioAxMcZLg7KKu4YOAr+vqxcWkUbAw3gY5EZVpwHTAFJTU3/UpsEcntyCIq5/Lp2lW/by27OO5rrj+1j7AGNiUHUNyqpsSFbBQ4OyLThHDxWS3HkVWgODgdnuzqcrMENEJqlqeg3bNkdo3Y4CrnkmjfwDJUy7KpXTUroEHckYE5DqjgiOtCFZGpAsIn1wCsBk4PKKhaq6F+hYMS0is4FfWBHw3zeZO7npvxk0axLHazeOZ0hS26AjGWMCVF2Dsu8akolIV5yLvwqkqer2mjasqmUiMhX4GOcaw3RVXSEiDwDpqmod2QXg9fTN/PqtZfTt1JJnfjqGxITmQUcyxgTMS4Oy64B7gc9xxiz+p4g8oKrTa3quqs4EZlaad+8h1j3JS2BzeFSVv3+6jkc/W8dx/Tvy7ytH0qZZk6BjGWMigJfbR38JjFDVXQAi0gH4FqixEJjIUFoe4tdvLeONjBwuGZXEHy8cQpM4T01IjDExwEsh2AUUhE0XuPNMFCgoKuWWFxfy1bqd3HlqMneckmx3BhljfsBLIcgE5ovIuzjXCM4DlorI3QCq+rCP+cwR2L63iGueWUBm7n4eungol6b2qPlJxpiY46UQrHd/Krzr/rZRyiNYZm4BVz+9gL0HS5l+zWhOGNAp6EjGmAjlpRA8qKpF4TNEpKOq7vQpkzlCGdn5/OzZdJrENeLVG8czONFuDzXGHJqXK4YL3K6oAaenUJyLxSYCfbpyB1c8NZ/2LeN56+ZjrAgYY2rk5YjgCmC62+CrO9ABONnPUObwvJq2id+8vZzB3dsw/ZrRdGjVNOhIxpgo4KWvoWUi8gfgBZw7hk5Q1Rzfk5laeWLOev784WpOGNCJx68YScumXmq8McZ4a1D2NNAPGAoMAN4XkX+qqqfBaYy/VJWHPl7D47PXc+6w7vztkmHEN7Y2AsYY77x8bVwGXKeqCmwQkbE4vYaagJWHlN+9u5yX5m/iirE9eeC8wTaamDGm1rycGvq7iPQSkWRV/RQoAe70P5qpTklZiLtfW8z7S7dx6wRnXGFrKGaMORw1nkMQkeuBN4D/uLOSgHf8DGWqd7CknBteSOf9pdv4zVkD+eUZA60IGGMOm5dTQ7fi9Dw6H0BV14lIZ19TmUMqKCrl2mfTScvO588XDmHymJ5BRzLGRDkvhaBYVUsqvnGKSGOqGbDG+Cf/QAk/mb6AVdv28Y/JIzh3WPegIxljGgAvhWCOiPwGaC4ipwG3AO/5G8tUtn1vEVc+PZ/N+YU8eXUqEwbaQZkxpm54uc/wHiAP5+6hG3HGF/gfP0OZH9qy5yCX/mcu2/cW8fzPxlgRMMbUKS93DYWAJ90fU89ydhcy5cl57Cks5b/XjWV4j4SgIxljGhhrfhrBNuc7RWDfwVJevG4sQ5OsCBhj6p4Vggi1aZdTBPYXl/HS9eOs8zhjjG8890UgIi38DGK+l73rAJOnzeVASRkvXjfWioAxxldeGpQdIyIrgdXu9DAR+beXjYvIRBFZIyKZInJPFctvEpFlIrJYRL4WkZRav4MGxikC8ygsLbciYIypF16OCB4BzsAdp1hVlwAn1PQkEYkDHgPOBFKAKVXs6F9S1SGqOhx4iBjvw6iiCBSVlvPSdeMY1N2KgDHGf55ODanq5kqzyj08bQyQqapZqloCvIIz3nH4dveFTbYkhhuqbdz5fRF48bpxpHRvE3QkY0yM8HKxeLOIHAOoiDQB7gBWeXheIhBeQHKAsZVXEpFbgbuBeA4x4I2I3ADcANCzZ8PrUqHiwnBxWYiXrh/H0d2sCBhj6o+XI4KbcPobSgS2AMPd6Tqhqo+paj/gVxyioZqqTlPVVFVN7dSpYQ3Cvmt/MVdPn89B95qAFQFjTH3zckQgqnrFYWx7C9AjbDrJnXcorwCPH8brRK3CkjJ+9mwa2/YW8dL1VgSMMcHwckTwjYh8IiLXikhtWjSlAcki0kdE4oHJwIzwFUQkOWzybGBdLbYf1UrLQ9z64kKWbdnLvy4fyahe7YOOZIyJUTUWAlUdgHPKZhCwUETeF5ErPTyvDJgKfIxzTeE1VV0hIg+IyCR3takiskJEFuNcJ/jJ4b6RaKKq/PbtZXyxJo//O38Ip6V0CTqSMSaGiTMCpceVRTri3OJ5harG+ZaqGqmpqZqenh7ES9eZh2et5R+freP2U5K5+7QBQccxxsQAEclQ1dSqlnlpUNZGRH4iIh8C3wLbcG4NNYfhrYU5/OOzdVwyKom7Tk2u+QnGGOMzLxeLl+AMTfmAqs71OU+DlrYxn3veXMb4vh34wwVDbHhJY0xE8FII+mptzh+ZKm3aVciNL2SQ1K45j185kvjGnrt5MsYYXx2yEIjI31X1TmCGiPyoEKjqpCqeZqqw92ApP312ASFVnr5mNAkt4oOOZIwx36nuiOAF9/df6yNIQ1UeUqa+tJBN+YW8cO1Y+nRsGXQkY4z5gUMWAlXNcB8OV9VHw5eJyB3AHD+DNRSPfrqWr9bt5MGLhjCub4eg4xhjzI94OVFd1b3919Rxjgbpy7V5/POLTC4ZlcRloxteH0nGmIahumsEU4DLgT4iEt4iuDWQ73ewaLdjXxF3vbqY5M6teOC8wUHHMcaYQ6ruGkFFm4GOwN/C5hcAS/0MFe3KykPc9vIiDpaW8+8rRtI8PpC2d8YY40l11wiygWxgfP3FaRge+XQtCzbk88hlw+jfuXXQcYwxplpeWhaPE5E0EdkvIiUiUi4i+2p6Xqz6et1OHvtiPZNH9+CCEUlBxzHGmBp5uVj8L2AKTs+gzYHrcIagNJXsLy7jV28upW+nltw/aVDQcYwxxhOvQ1VmAnGqWq6qzwAT/Y0VnR78cDVb9x7kLxcPpVkTuy5gjIkOXrqYKHTHE1gsIg/hXEC2/hEqmZe1ixfmZXPtcX1sbAFjTFTxskO/CojDGVvgAM6oYxf5GSraHCwp51dvLqVXhxb84vSjgo5jjDG1UuMRgXv3EMBB4Pf+xolOf/1kDdm7Cnn5+nF2q6gxJupU16BsGXDIXkdVdagviaJMRvZupn+zgavG9WJ8P+tCwhgTfao7Ijin3lJEqdLyEL9+aynd2zbnV2cODDqOMcYclpoalJlqvDA3m7U79jPtqlG0aurlursxxkSeGvdeIlLA96eI4oEmwAFVbeNnsEiXV1DMI7PWcsKATjb4vDEmqtV415CqtlbVNu6OvznOHUP/9rJxEZkoImtEJFNE7qli+d0islJElorIZyLSq9bvICAPfbSaorJy7js3xYacNMZEtVq1B1DHO8AZNa0rInE4LZDPBFKAKSKSUmm1RUCqe+H5DeCh2uQJysJNu3k9I4efHdeHfp1aBR3HGGOOiJdTQxeGTTYCUoEiD9seA2Sqapa7nVeA84CVFSuo6hdh688DrvSw3UCVh5T73l1BlzZNue3k5KDjGGPMEfNyhfPcsMdlwEacHXpNEoHNYdM5wNhq1r8W+NDDdgP1Wvpmlm3Zy6OTh9sFYmNMg+ClQdlP/Q4hIlfiHGmceIjlNwA3APTsGdxIX/uKSnnoo9WM6d2eScO6B5bDGGPqkpdTQ32A272sFbAAAA4pSURBVIDe4eur6qQanroFpzuKCknuvMrbPxX4LXCiqhZXtSFVnQZMA0hNTT1kIze/Pf3VBnYXlvK7c+wCsTGm4fBybuMd4GngPSBUi22nAcluIdkCTMYZ+vI7IjIC+A8wUVVza7HterensITpX2/gjEFdGJLUNug4xhhTZ7wUgiJV/UdtN6yqZSIyFfgYp9O66aq6QkQeANJVdQbwF6AV8Lr7DXuThyONQEz7Mov9JWXcddqAoKMYY0yd8lIIHhWR+4BPgO9O3ajqwpqeqKozgZmV5t0b9vhU71GDs2t/Mc9+u5Gzh3RjYNeYbkdnjGmAvBSCIThdUZ/M96eG1J2OCU/MWU9RaTl3nmpHA8aYhsdLIbgE6KuqJX6HiUS5+4p4fm42549IpH9nazxmjGl4vLQsXg4k+B0kUv179nrKQsodp1jjMWNMw+TliCABWC0iafzwGkFEXtStS1v3HOSl+Zu4ZFQSvTq0DDqOMcb4wkshuM/3FBFq2pdZKMrUk/sHHcUYY3zjpWXxnPoIEmnyD5TwStomzhueSFK7FkHHMcYY39h4BIfw3LcbKSoNcdOJfYOOYowxvvJyRNC64rE4rb7OA8b5GSpohSVlPDd3I6ce3YX+nVvXuL4xxkQz38YjiGavpm1mT2EpN59kRwPGmIbPz/EIolJpeYinvtpAaq92jOrVPug4xhjjOz/HI4hKHyzdxpY9B/n9pEFBRzHGmHoREeMRRApV5Yk560nu3IqTB3YOOo4xxtSLGq8RiMhzIpIQNt1ORKb7GysYs9fmsXp7ATee2I9GjWy8AWNMbPBysXioqu6pmFDV3cAI/yIFZ9qcLLq1bWajjxljYoqXQtBIRNpVTIhIe7xdW4gqa3cUMDdrF1eP701841rdTGWMMVHNyw79b8BcEXndnb4E+IN/kYLxwtxs4hs34rLRPWpe2RhjGhAvF4ufF5F0vh9/4EJVXelvrPpVUFTKWwtzOHdod9q3jA86jjHG1CtPp3jcHX+D2vmHe2vhFg6UlHP1+F5BRzHGmHoX8yfDVZUX5mUzLKktw3rE7LALxpgYFvOFYO76XWTm7ueq8b2DjmKMMYGI+ULw3NyNtGvRhHOGdgs6ijHGBMLXQiAiE0VkjYhkisg9VSw/QUQWikiZiFzsZ5aqbN1zkFkrd3DZ6J40axJX3y9vjDERwbdCICJxwGPAmUAKMEVEUiqttgm4BnjJrxzVeWn+JhS4YmzPIF7eGGMigp8Nw8YAmaqaBSAir+B0Vvfd3UequtFdFvIxR5VKykK8vGATpwzsTI/2NgKZMSZ2+XlqKBHYHDad486rNRG5QUTSRSQ9Ly+vTsJ9uTaPXQdKuNyOBowxMS4qLhar6jRVTVXV1E6dOtXJNj9Yto22zZtwXP+62Z4xxkQrPwvBFiC8v4Ykd17gikrLmbVyB2cM6mL9ChljYp6fe8E0IFlE+ohIPDAZmOHj63k2e00e+4vLOGeo9TJqjDG+FQJVLQOmAh8Dq4DXVHWFiDwgIpMARGS0iOTgdGT3HxFZ4VeecB8s20b7lvEc069DfbycMcZENF+7k1bVmcDMSvPuDXuchnPKqN4cLCnns1U7OH9EIo3j7LSQMcbE3J7wizW5FJaUc84Qa0lsjDEQg4Xg/aVb6diqKWP72mkhY4yBGCsEB4rL+Hx1LmcN6UqcjUlsjDFAjBWCz1bnUlQa4mw7LWSMMd+JqULw/pKtdGnTlNG92wcdxRhjIkbMFIKColJmr83jrCHdaGSnhYwx5jsxUwg+XbWDkrKQNSIzxphKYqYQtG7ahNNTujDChqM0xpgf8LVBWSQ5NaULp6Z0CTqGMcZEnJg5IjDGGFM1KwTGGBPjrBAYY0yMs0JgjDExzgqBMcbEOCsExhgT46wQGGNMjLNCYIwxMU5UNegMtSIieUD2YT69I7CzDuP4LZryRlNWiK680ZQVoitvNGWFI8vbS1U7VbUg6grBkRCRdFVNDTqHV9GUN5qyQnTljaasEF15oykr+JfXTg0ZY0yMs0JgjDExLtYKwbSgA9RSNOWNpqwQXXmjKStEV95oygo+5Y2pawTGGGN+LNaOCIwxxlRihcAYY2JczBQCEZkoImtEJFNE7gk6T2UiMl1EckVkedi89iIyS0TWub/bBZmxgoj0EJEvRGSliKwQkTvc+RGXV0SaicgCEVniZv29O7+PiMx3Pw+vikh80FkriEiciCwSkffd6UjOulFElonIYhFJd+dF3OeggogkiMgbIrJaRFaJyPhIzCsiR7l/04qffSJyp19ZY6IQiEgc8BhwJpACTBGRlGBT/cizwMRK8+4BPlPVZOAzdzoSlAE/V9UUYBxwq/v3jMS8xcDJqjoMGA5MFJFxwIPAI6raH9gNXBtgxsruAFaFTUdyVoAJqjo87P72SPwcVHgU+EhVBwLDcP7OEZdXVde4f9PhwCigEHgbv7KqaoP/AcYDH4dN/xr4ddC5qsjZG1geNr0G6OY+7gasCTrjIXK/C5wW6XmBFsBCYCxO68zGVX0+As6Y5P4HPxl4H5BIzerm2Qh0rDQvIj8HQFtgA+5NMpGeNyzf6cA3fmaNiSMCIBHYHDad486LdF1UdZv7eDsQcYMui0hvYAQwnwjN655qWQzkArOA9cAeVS1zV4mkz8Pfgf8HhNzpDkRuVgAFPhGRDBG5wZ0XkZ8DoA+QBzzjnnp7SkRaErl5K0wGXnYf+5I1VgpB1FPnK0BE3esrIq2AN4E7VXVf+LJIyquq5eocYicBY4CBAUeqkoicA+SqakbQWWrhOFUdiXPa9VYROSF8YSR9DoDGwEjgcVUdARyg0qmVCMuLez1oEvB65WV1mTVWCsEWoEfYdJI7L9LtEJFuAO7v3IDzfEdEmuAUgRdV9S13dsTmBVDVPcAXOKdXEkSksbsoUj4PxwKTRGQj8ArO6aFHicysAKjqFvd3Ls457DFE7ucgB8hR1fnu9Bs4hSFS84JTYBeq6g532pessVII0oBk9+6LeJxDrRkBZ/JiBvAT9/FPcM7FB05EBHgaWKWqD4ctiri8ItJJRBLcx81xrmWswikIF7urRURWVf21qiapam+cz+jnqnoFEZgVQERaikjrisc457KXE4GfAwBV3Q5sFpGj3FmnACuJ0LyuKXx/Wgj8yhr0hZB6vOByFrAW5/zwb4POU0W+l4FtQCnON5drcc4PfwasAz4F2ged0816HM4h6VJgsftzViTmBYYCi9ysy4F73fl9gQVAJs5hd9Ogs1bKfRLwfiRndXMtcX9WVPy/isTPQVjm4UC6+3l4B2gXqXmBlsAuoG3YPF+yWhcTxhgT42Ll1JAxxphDsEJgjDExzgqBMcbEOCsExhgT46wQGGNMjLNCYKKaiMwWEd8HHxeR293eKl/0+7WC5PbOeUvQOUz9skJgYlZYa10vbgFOU6eBV0OWgPNeTQyxQmB8JyK93W/TT7pjAnzitvL9wTd6Eenodq+AiFwjIu+4fa5vFJGpInK321nYPBFpH/YSV7l9ti8XkTHu81uKM8bDAvc554Vtd4aIfI7TMKdy1rvd7SwXkTvdeU/gNJ76UETuqrR+nIj81V1/qYjc5s4/xX3dZW6Opu78jSLyp4r++0VkpIh8LCLrReQmd52TRORLEflAnDE0nhCRRu6yKe42l4vIg2E59ovIH8QZd2GeiHRx53cSkTdFJM39Odadf7+ba7aIZInI7e6m/gz0c/P9RUS6uVkq/r7HH/YHwUSuoFvP2U/D/8HpXrsMGO5OvwZc6T6eDaS6jzsCG93H1+C0pG0NdAL2Aje5yx7B6eiu4vlPuo9PwO3GG/hj2Gsk4LQqb+luN4cqWmTi9Pu+zF2vFU5r2RHuso1U6m7ZnX8zTp81Fd1Etwea4fR2O8Cd93xY3o3AzWHvY2nYe9zhzj8JKMIpPnE4PaZeDHQHNrnrNgY+B853n6PAue7jh4D/cR+/hNMxHEBPnG5BAO4HvgWaun/3XUATftwV+s/5vsVwHNA66M+T/dT9T20OjY05EhtUdbH7OANnh1OTL1S1ACgQkb3Ae+78ZThdR1R4GUBVvxSRNm7fQqfjdOD2C3edZjg7QoBZqppfxesdB7ytqgcAROQt4HicLioO5VTgCXW7iVbVfBEZ5r7fte46zwG34nQxDd/3c7UMaBX2Hosr+kUCFqhqlpvjZTdbKTBbVfPc+S/iFL93gBKc8QvA+fueFpYvxekeCoA24vQaC/CBqhYDxSKSS9VdGqcB08XpZPCdsH9D04BYITD1pTjscTnQ3H1cxvenKJtV85xQ2HSIH352K/eTojgDulykqmvCF4jIWJzuh4MU/j4qv8eK91XVe6pOqapWrFMetp1GwDhVLQpf2S0Mlf9NfrQ/cIvrCcDZwLMi8rCqPl9DFhNl7BqBCdpGnFMy8H0Pm7V1GYCIHAfsVdW9wMfAbW5PqYjICA/b+Qo4X0RauL1pXuDOq84s4MaKC8/utYs1QG8R6e+ucxUwp5bvaYw4veU2wnl/X+N0PHeiey0lDqdnypq2+wlwW8WEiAyvYf0CnFNVFev3wjll9STwFE63zaaBsUJggvZX4GYRWYRzrvpwFLnPf4Lvx/P9X5xz3ktFZIU7XS1VXYgzdvQCnBHXnlLV6k4LgbNz3OS+zhLgcvfb90+B10VkGc43/Sdq+Z7SgH/hdJm9AeeU1TacgVS+wOnxM0NVa+qG+HYg1b2QvRK4qbqVVXUX8I17YfgvONcrlrh/38twxkcwDYz1PmpMhBGRk4BfqOo5QWcxscGOCIwxJsbZEYExxsQ4OyIwxpgYZ4XAGGNinBUCY4yJcVYIjDEmxlkhMMaYGPf/AUCai2XC89BxAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XUUcVgAmu6Eg"
      },
      "source": [
        "### UMAP"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HOPa90IXbyjF",
        "outputId": "530ecf2f-5ce6-44df-85d9-b6ab6d247059"
      },
      "source": [
        "reducer = umap.UMAP(metric=\"braycurtis\", n_neighbors=20, random_state=26, n_components=2) # roughly 1% of the number of original features\n",
        "reducer.fit(X_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "UMAP(a=None, angular_rp_forest=False, b=None, dens_frac=0.0, dens_lambda=0.0,\n",
              "     dens_var_shift=0.1, densmap=False, disconnection_distance=None,\n",
              "     force_approximation_algorithm=False, init='spectral', learning_rate=1.0,\n",
              "     local_connectivity=1.0, low_memory=True, metric='braycurtis',\n",
              "     metric_kwds=None, min_dist=0.1, n_components=2, n_epochs=None, n_jobs=-1,\n",
              "     n_neighbors=20, negative_sample_rate=5, output_dens=False,\n",
              "     output_metric='euclidean', output_metric_kwds=None, random_state=26,\n",
              "     repulsion_strength=1.0, set_op_mix_ratio=1.0, spread=1.0,\n",
              "     target_metric='categorical', target_metric_kwds=None, ...)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p39IbN4bcGfd"
      },
      "source": [
        "X_train_umap = reducer.transform(X_train)\n",
        "X_val_umap = reducer.transform(X_val)\n",
        "X_test_umap = reducer.transform(X_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ax5jkGNPnXbM"
      },
      "source": [
        "---\n",
        "# Model training"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Esvme4Ygx0Y1"
      },
      "source": [
        "## Model 1 - simple neural network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-4a72E8hIbqJ"
      },
      "source": [
        "### Create TensorFlow dataset"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "j6OLRfR1uSmC"
      },
      "source": [
        "#### Dataset with original feature representation data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Gv2e0cJK6V-L"
      },
      "source": [
        "# TensorFlow dataset from CSV\n",
        "def df_to_dataset(dataframe, shuffle=True, batch_size=64):\n",
        "  dataframe = dataframe.copy()\n",
        "  labels = dataframe.pop('cuisine')\n",
        "  ds = tf.data.Dataset.from_tensor_slices((dict(dataframe), labels))\n",
        "  if shuffle:\n",
        "    ds = ds.shuffle(buffer_size=len(dataframe), reshuffle_each_iteration=True)\n",
        "  ds = ds.batch(batch_size)\n",
        "  return ds"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AQ7d-XH6-j3g"
      },
      "source": [
        "feature_columns = []\n",
        "\n",
        "for header in recipes_df.iloc[:,:-1].columns:\n",
        "  feature_columns.append(tf.feature_column.numeric_column(header, 1))"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8_kSxMGq8VHP"
      },
      "source": [
        "# Creating batches of data for mini-batch stochastic gradient descent\n",
        "batch_size = 64\n",
        "train_ds = df_to_dataset(train, batch_size=batch_size)\n",
        "val_ds = df_to_dataset(val, shuffle=False, batch_size=batch_size)\n",
        "test_ds = df_to_dataset(test, shuffle=False, batch_size=batch_size)\n",
        "\n",
        "# train_ds = train_ds.shuffle(296, reshuffle_each_iteration=True)"
      ],
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nQgZQAjgubrI"
      },
      "source": [
        "#### Dataset with PCA/UMAP data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YBOgJ8VzPaT_"
      },
      "source": [
        "SHUFFLE_BUFFER_SIZE = len(train)\n",
        "train_ds = tf.data.Dataset.from_tensor_slices((X_train_umap, tf.one_hot(y_train, 12)))\n",
        "val_ds = tf.data.Dataset.from_tensor_slices((X_val_umap, tf.one_hot(y_val, 12)))\n",
        "test_ds = tf.data.Dataset.from_tensor_slices((X_test_umap, tf.one_hot(y_test, 12)))\n",
        "\n",
        "train_ds = train_ds.shuffle(SHUFFLE_BUFFER_SIZE, reshuffle_each_iteration=True).batch(batch_size)\n",
        "val_ds = val_ds.batch(batch_size)\n",
        "test_ds = test_ds.batch(batch_size)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "C7pxjQ1G2AQo"
      },
      "source": [
        "### Training model on original feature representation data "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CmqX82eXbpAQ"
      },
      "source": [
        "early_stop_callback = tf.keras.callbacks.EarlyStopping(\n",
        "            patience=10,\n",
        "            monitor='val_loss',\n",
        "            verbose=1,\n",
        "            mode='min',\n",
        "            restore_best_weights=True)"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BNsGwu6--ufM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ee34aeed-24aa-4d39-ba27-b72490a8f473"
      },
      "source": [
        "feature_layer = tf.keras.layers.DenseFeatures(feature_columns)\n",
        "\n",
        "model = tf.keras.Sequential([\n",
        "  feature_layer,\n",
        "  layers.Dense(128, activation='relu', kernel_regularizer=tf.keras.regularizers.l2(l2=1e-4)),\n",
        "  layers.Dropout(0.5),\n",
        "  layers.Dense(12, kernel_regularizer=tf.keras.regularizers.l2(l2=1e-4))\n",
        "])\n",
        "\n",
        "model.compile(optimizer='adam',\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "              metrics=tf.metrics.SparseCategoricalAccuracy())\n",
        "\n",
        "# model.summary()\n",
        "\n",
        "model.fit(train_ds,\n",
        "          validation_data=val_ds,\n",
        "          callbacks=[early_stop_callback],\n",
        "          epochs=1000)"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/1000\n",
            "WARNING:tensorflow:Layers in a Sequential model should only have a single input tensor, but we receive a <class 'dict'> input: {'acorn_squash': <tf.Tensor 'ExpandDims_20:0' shape=(None, 1) dtype=int64>, 'adobo': <tf.Tensor 'ExpandDims_21:0' shape=(None, 1) dtype=int64>, 'african_birdseye_chile_pepper': <tf.Tensor 'ExpandDims_22:0' shape=(None, 1) dtype=int64>, 'ale': <tf.Tensor 'ExpandDims_23:0' shape=(None, 1) dtype=int64>, 'aleppo_pepper': <tf.Tensor 'ExpandDims_24:0' shape=(None, 1) dtype=int64>, 'alfalfa_sprouts': <tf.Tensor 'ExpandDims_25:0' shape=(None, 1) dtype=int64>, 'alfredo_sauce': <tf.Tensor 'ExpandDims_26:0' shape=(None, 1) dtype=int64>, 'allspice': <tf.Tensor 'ExpandDims_27:0' shape=(None, 1) dtype=int64>, 'almond': <tf.Tensor 'ExpandDims_28:0' shape=(None, 1) dtype=int64>, 'almond_butter': <tf.Tensor 'ExpandDims_29:0' shape=(None, 1) dtype=int64>, 'amaretti': <tf.Tensor 'ExpandDims_30:0' shape=(None, 1) dtype=int64>, 'amchoor_powder': <tf.Tensor 'ExpandDims_31:0' shape=(None, 1) dtype=int64>, 'ancho_chile_pepper': <tf.Tensor 'ExpandDims_32:0' shape=(None, 1) dtype=int64>, 'anchovy': <tf.Tensor 'ExpandDims_33:0' shape=(None, 1) dtype=int64>, 'anchovy_paste': <tf.Tensor 'ExpandDims_34:0' shape=(None, 1) dtype=int64>, 'andouille_sausage': <tf.Tensor 'ExpandDims_35:0' shape=(None, 1) dtype=int64>, 'anise': <tf.Tensor 'ExpandDims_36:0' shape=(None, 1) dtype=int64>, 'apple': <tf.Tensor 'ExpandDims_37:0' shape=(None, 1) dtype=int64>, 'apple_butter': <tf.Tensor 'ExpandDims_38:0' shape=(None, 1) dtype=int64>, 'apple_juice': <tf.Tensor 'ExpandDims_39:0' shape=(None, 1) dtype=int64>, 'applesauce': <tf.Tensor 'ExpandDims_40:0' shape=(None, 1) dtype=int64>, 'apricot': <tf.Tensor 'ExpandDims_41:0' shape=(None, 1) dtype=int64>, 'arborio_rice': <tf.Tensor 'ExpandDims_42:0' shape=(None, 1) dtype=int64>, 'argan_oil': <tf.Tensor 'ExpandDims_43:0' shape=(None, 1) dtype=int64>, 'arrowroot': <tf.Tensor 'ExpandDims_44:0' shape=(None, 1) dtype=int64>, 'artichoke': <tf.Tensor 'ExpandDims_45:0' shape=(None, 1) dtype=int64>, 'artificial_sweetener': <tf.Tensor 'ExpandDims_46:0' shape=(None, 1) dtype=int64>, 'arugula': <tf.Tensor 'ExpandDims_47:0' shape=(None, 1) dtype=int64>, 'asadero': <tf.Tensor 'ExpandDims_48:0' shape=(None, 1) dtype=int64>, 'asafetida': <tf.Tensor 'ExpandDims_49:0' shape=(None, 1) dtype=int64>, 'asiago_cheese': <tf.Tensor 'ExpandDims_50:0' shape=(None, 1) dtype=int64>, 'asparagus': <tf.Tensor 'ExpandDims_51:0' shape=(None, 1) dtype=int64>, 'assorted_vegetables': <tf.Tensor 'ExpandDims_52:0' shape=(None, 1) dtype=int64>, 'avocado': <tf.Tensor 'ExpandDims_53:0' shape=(None, 1) dtype=int64>, 'azuki_bean': <tf.Tensor 'ExpandDims_54:0' shape=(None, 1) dtype=int64>, 'baby_carrot': <tf.Tensor 'ExpandDims_55:0' shape=(None, 1) dtype=int64>, 'baby_corn': <tf.Tensor 'ExpandDims_56:0' shape=(None, 1) dtype=int64>, 'baby_onions': <tf.Tensor 'ExpandDims_57:0' shape=(None, 1) dtype=int64>, 'baby_spinach': <tf.Tensor 'ExpandDims_58:0' shape=(None, 1) dtype=int64>, 'bacon': <tf.Tensor 'ExpandDims_59:0' shape=(None, 1) dtype=int64>, 'bacon_grease': <tf.Tensor 'ExpandDims_60:0' shape=(None, 1) dtype=int64>, 'bagel': <tf.Tensor 'ExpandDims_61:0' shape=(None, 1) dtype=int64>, 'baguette': <tf.Tensor 'ExpandDims_62:0' shape=(None, 1) dtype=int64>, 'baked_beans': <tf.Tensor 'ExpandDims_63:0' shape=(None, 1) dtype=int64>, 'baking_powder': <tf.Tensor 'ExpandDims_64:0' shape=(None, 1) dtype=int64>, 'baking_soda': <tf.Tensor 'ExpandDims_65:0' shape=(None, 1) dtype=int64>, 'balsamic_vinegar': <tf.Tensor 'ExpandDims_66:0' shape=(None, 1) dtype=int64>, 'bamboo_shoot': <tf.Tensor 'ExpandDims_67:0' shape=(None, 1) dtype=int64>, 'bamboo_skewers': <tf.Tensor 'ExpandDims_68:0' shape=(None, 1) dtype=int64>, 'banana': <tf.Tensor 'ExpandDims_69:0' shape=(None, 1) dtype=int64>, 'barbecue_sauce': <tf.Tensor 'ExpandDims_70:0' shape=(None, 1) dtype=int64>, 'barley': <tf.Tensor 'ExpandDims_71:0' shape=(None, 1) dtype=int64>, 'basil': <tf.Tensor 'ExpandDims_72:0' shape=(None, 1) dtype=int64>, 'basmati_rice': <tf.Tensor 'ExpandDims_73:0' shape=(None, 1) dtype=int64>, 'bass': <tf.Tensor 'ExpandDims_74:0' shape=(None, 1) dtype=int64>, 'bay_leaf': <tf.Tensor 'ExpandDims_75:0' shape=(None, 1) dtype=int64>, 'bean_sprouts': <tf.Tensor 'ExpandDims_76:0' shape=(None, 1) dtype=int64>, 'bean_thread': <tf.Tensor 'ExpandDims_77:0' shape=(None, 1) dtype=int64>, 'beans': <tf.Tensor 'ExpandDims_78:0' shape=(None, 1) dtype=int64>, 'bechamel': <tf.Tensor 'ExpandDims_79:0' shape=(None, 1) dtype=int64>, 'beef': <tf.Tensor 'ExpandDims_80:0' shape=(None, 1) dtype=int64>, 'beef_consomme': <tf.Tensor 'ExpandDims_81:0' shape=(None, 1) dtype=int64>, 'beef_gravy': <tf.Tensor 'ExpandDims_82:0' shape=(None, 1) dtype=int64>, 'beef_round_cuts': <tf.Tensor 'ExpandDims_83:0' shape=(None, 1) dtype=int64>, 'beef_stock': <tf.Tensor 'ExpandDims_84:0' shape=(None, 1) dtype=int64>, 'beer': <tf.Tensor 'ExpandDims_85:0' shape=(None, 1) dtype=int64>, 'beet': <tf.Tensor 'ExpandDims_86:0' shape=(None, 1) dtype=int64>, 'biscuit_mix': <tf.Tensor 'ExpandDims_87:0' shape=(None, 1) dtype=int64>, 'bitter_melon': <tf.Tensor 'ExpandDims_88:0' shape=(None, 1) dtype=int64>, 'black_bean': <tf.Tensor 'ExpandDims_89:0' shape=(None, 1) dtype=int64>, 'black_eyed_pea': <tf.Tensor 'ExpandDims_90:0' shape=(None, 1) dtype=int64>, 'black_forest_ham': <tf.Tensor 'ExpandDims_91:0' shape=(None, 1) dtype=int64>, 'black_mushroom': <tf.Tensor 'ExpandDims_92:0' shape=(None, 1) dtype=int64>, 'black_olive': <tf.Tensor 'ExpandDims_93:0' shape=(None, 1) dtype=int64>, 'black_truffle': <tf.Tensor 'ExpandDims_94:0' shape=(None, 1) dtype=int64>, 'blackberry': <tf.Tensor 'ExpandDims_95:0' shape=(None, 1) dtype=int64>, 'blue_cheese': <tf.Tensor 'ExpandDims_96:0' shape=(None, 1) dtype=int64>, 'bok_choy': <tf.Tensor 'ExpandDims_97:0' shape=(None, 1) dtype=int64>, 'bouillon': <tf.Tensor 'ExpandDims_98:0' shape=(None, 1) dtype=int64>, 'bouquet_garni': <tf.Tensor 'ExpandDims_99:0' shape=(None, 1) dtype=int64>, 'bourbon': <tf.Tensor 'ExpandDims_100:0' shape=(None, 1) dtype=int64>, 'brandy': <tf.Tensor 'ExpandDims_101:0' shape=(None, 1) dtype=int64>, 'bratwursts': <tf.Tensor 'ExpandDims_102:0' shape=(None, 1) dtype=int64>, 'bread': <tf.Tensor 'ExpandDims_103:0' shape=(None, 1) dtype=int64>, 'bread_bun': <tf.Tensor 'ExpandDims_104:0' shape=(None, 1) dtype=int64>, 'breadcrumb': <tf.Tensor 'ExpandDims_105:0' shape=(None, 1) dtype=int64>, 'brick_cheese': <tf.Tensor 'ExpandDims_106:0' shape=(None, 1) dtype=int64>, 'brie_cheese': <tf.Tensor 'ExpandDims_107:0' shape=(None, 1) dtype=int64>, 'broccoli': <tf.Tensor 'ExpandDims_108:0' shape=(None, 1) dtype=int64>, 'broccoli_raab': <tf.Tensor 'ExpandDims_109:0' shape=(None, 1) dtype=int64>, 'broth': <tf.Tensor 'ExpandDims_110:0' shape=(None, 1) dtype=int64>, 'brown_gravy_mix': <tf.Tensor 'ExpandDims_111:0' shape=(None, 1) dtype=int64>, 'brown_onion': <tf.Tensor 'ExpandDims_112:0' shape=(None, 1) dtype=int64>, 'brown_rice': <tf.Tensor 'ExpandDims_113:0' shape=(None, 1) dtype=int64>, 'brown_sugar': <tf.Tensor 'ExpandDims_114:0' shape=(None, 1) dtype=int64>, 'brussels_sprout': <tf.Tensor 'ExpandDims_115:0' shape=(None, 1) dtype=int64>, 'buckwheat': <tf.Tensor 'ExpandDims_116:0' shape=(None, 1) dtype=int64>, 'bulgur_wheat': <tf.Tensor 'ExpandDims_117:0' shape=(None, 1) dtype=int64>, 'burdock_root': <tf.Tensor 'ExpandDims_118:0' shape=(None, 1) dtype=int64>, 'butter': <tf.Tensor 'ExpandDims_119:0' shape=(None, 1) dtype=int64>, 'buttermilk': <tf.Tensor 'ExpandDims_120:0' shape=(None, 1) dtype=int64>, 'butternut_squash': <tf.Tensor 'ExpandDims_121:0' shape=(None, 1) dtype=int64>, 'cabbage': <tf.Tensor 'ExpandDims_122:0' shape=(None, 1) dtype=int64>, 'cajun_seasoning': <tf.Tensor 'ExpandDims_123:0' shape=(None, 1) dtype=int64>, 'calvados': <tf.Tensor 'ExpandDims_124:0' shape=(None, 1) dtype=int64>, 'campbell': <tf.Tensor 'ExpandDims_125:0' shape=(None, 1) dtype=int64>, 'canadian_bacon': <tf.Tensor 'ExpandDims_126:0' shape=(None, 1) dtype=int64>, 'cannellini_bean': <tf.Tensor 'ExpandDims_127:0' shape=(None, 1) dtype=int64>, 'canola_oil': <tf.Tensor 'ExpandDims_128:0' shape=(None, 1) dtype=int64>, 'cantaloupe': <tf.Tensor 'ExpandDims_129:0' shape=(None, 1) dtype=int64>, 'caper': <tf.Tensor 'ExpandDims_130:0' shape=(None, 1) dtype=int64>, 'capsicum': <tf.Tensor 'ExpandDims_131:0' shape=(None, 1) dtype=int64>, 'caraway_seed': <tf.Tensor 'ExpandDims_132:0' shape=(None, 1) dtype=int64>, 'cardamom': <tf.Tensor 'ExpandDims_133:0' shape=(None, 1) dtype=int64>, 'carrot': <tf.Tensor 'ExpandDims_134:0' shape=(None, 1) dtype=int64>, 'cashew_nut': <tf.Tensor 'ExpandDims_135:0' shape=(None, 1) dtype=int64>, 'catfish': <tf.Tensor 'ExpandDims_136:0' shape=(None, 1) dtype=int64>, 'caul': <tf.Tensor 'ExpandDims_137:0' shape=(None, 1) dtype=int64>, 'cauliflower': <tf.Tensor 'ExpandDims_138:0' shape=(None, 1) dtype=int64>, 'cayenne_pepper': <tf.Tensor 'ExpandDims_139:0' shape=(None, 1) dtype=int64>, 'celeriac': <tf.Tensor 'ExpandDims_140:0' shape=(None, 1) dtype=int64>, 'celery': <tf.Tensor 'ExpandDims_141:0' shape=(None, 1) dtype=int64>, 'celery_salt': <tf.Tensor 'ExpandDims_142:0' shape=(None, 1) dtype=int64>, 'celery_seed': <tf.Tensor 'ExpandDims_143:0' shape=(None, 1) dtype=int64>, 'celery_soup': <tf.Tensor 'ExpandDims_144:0' shape=(None, 1) dtype=int64>, 'chanterelle_mushroom': <tf.Tensor 'ExpandDims_145:0' shape=(None, 1) dtype=int64>, 'chard': <tf.Tensor 'ExpandDims_146:0' shape=(None, 1) dtype=int64>, 'chayote': <tf.Tensor 'ExpandDims_147:0' shape=(None, 1) dtype=int64>, 'cheddar_cheese': <tf.Tensor 'ExpandDims_148:0' shape=(None, 1) dtype=int64>, 'cheddar_cheese_soup': <tf.Tensor 'ExpandDims_149:0' shape=(None, 1) dtype=int64>, 'cheese': <tf.Tensor 'ExpandDims_150:0' shape=(None, 1) dtype=int64>, 'cherry': <tf.Tensor 'ExpandDims_151:0' shape=(None, 1) dtype=int64>, 'chervil': <tf.Tensor 'ExpandDims_152:0' shape=(None, 1) dtype=int64>, 'cheshire_cheese': <tf.Tensor 'ExpandDims_153:0' shape=(None, 1) dtype=int64>, 'chestnut': <tf.Tensor 'ExpandDims_154:0' shape=(None, 1) dtype=int64>, 'chestnut_mushroom': <tf.Tensor 'ExpandDims_155:0' shape=(None, 1) dtype=int64>, 'chicken': <tf.Tensor 'ExpandDims_156:0' shape=(None, 1) dtype=int64>, 'chicken_base': <tf.Tensor 'ExpandDims_157:0' shape=(None, 1) dtype=int64>, 'chicken_liver': <tf.Tensor 'ExpandDims_158:0' shape=(None, 1) dtype=int64>, 'chicken_sausage': <tf.Tensor 'ExpandDims_159:0' shape=(None, 1) dtype=int64>, 'chicken_soup': <tf.Tensor 'ExpandDims_160:0' shape=(None, 1) dtype=int64>, 'chicken_stock': <tf.Tensor 'ExpandDims_161:0' shape=(None, 1) dtype=int64>, 'chickpea': <tf.Tensor 'ExpandDims_162:0' shape=(None, 1) dtype=int64>, 'chile_pepper': <tf.Tensor 'ExpandDims_163:0' shape=(None, 1) dtype=int64>, 'chili_oil': <tf.Tensor 'ExpandDims_164:0' shape=(None, 1) dtype=int64>, 'chili_paste': <tf.Tensor 'ExpandDims_165:0' shape=(None, 1) dtype=int64>, 'chili_powder': <tf.Tensor 'ExpandDims_166:0' shape=(None, 1) dtype=int64>, 'chili_sauce': <tf.Tensor 'ExpandDims_167:0' shape=(None, 1) dtype=int64>, 'chinese_cabbage': <tf.Tensor 'ExpandDims_168:0' shape=(None, 1) dtype=int64>, 'chinese_chives': <tf.Tensor 'ExpandDims_169:0' shape=(None, 1) dtype=int64>, 'Chinese_mushroom': <tf.Tensor 'ExpandDims:0' shape=(None, 1) dtype=int64>, 'Chinese_noodles': <tf.Tensor 'ExpandDims_1:0' shape=(None, 1) dtype=int64>, 'chipotle_pepper': <tf.Tensor 'ExpandDims_170:0' shape=(None, 1) dtype=int64>, 'chives': <tf.Tensor 'ExpandDims_171:0' shape=(None, 1) dtype=int64>, 'chocolate': <tf.Tensor 'ExpandDims_172:0' shape=(None, 1) dtype=int64>, 'chutney': <tf.Tensor 'ExpandDims_173:0' shape=(None, 1) dtype=int64>, 'cider': <tf.Tensor 'ExpandDims_174:0' shape=(None, 1) dtype=int64>, 'cider_vinegar': <tf.Tensor 'ExpandDims_175:0' shape=(None, 1) dtype=int64>, 'cilantro': <tf.Tensor 'ExpandDims_176:0' shape=(None, 1) dtype=int64>, 'cinnamon': <tf.Tensor 'ExpandDims_177:0' shape=(None, 1) dtype=int64>, 'clam': <tf.Tensor 'ExpandDims_178:0' shape=(None, 1) dtype=int64>, 'clam_juice': <tf.Tensor 'ExpandDims_179:0' shape=(None, 1) dtype=int64>, 'clarified_butter': <tf.Tensor 'ExpandDims_180:0' shape=(None, 1) dtype=int64>, 'clove': <tf.Tensor 'ExpandDims_181:0' shape=(None, 1) dtype=int64>, 'cocoa_powder': <tf.Tensor 'ExpandDims_182:0' shape=(None, 1) dtype=int64>, 'coconut': <tf.Tensor 'ExpandDims_183:0' shape=(None, 1) dtype=int64>, 'coconut_milk_or_cream': <tf.Tensor 'ExpandDims_184:0' shape=(None, 1) dtype=int64>, 'coconut_oil': <tf.Tensor 'ExpandDims_185:0' shape=(None, 1) dtype=int64>, 'cod': <tf.Tensor 'ExpandDims_186:0' shape=(None, 1) dtype=int64>, 'colby': <tf.Tensor 'ExpandDims_187:0' shape=(None, 1) dtype=int64>, 'coleslaw_mix': <tf.Tensor 'ExpandDims_188:0' shape=(None, 1) dtype=int64>, 'confit': <tf.Tensor 'ExpandDims_189:0' shape=(None, 1) dtype=int64>, 'coriander': <tf.Tensor 'ExpandDims_190:0' shape=(None, 1) dtype=int64>, 'coriander_root': <tf.Tensor 'ExpandDims_191:0' shape=(None, 1) dtype=int64>, 'corn': <tf.Tensor 'ExpandDims_192:0' shape=(None, 1) dtype=int64>, 'corn_chips': <tf.Tensor 'ExpandDims_193:0' shape=(None, 1) dtype=int64>, 'corn_flour': <tf.Tensor 'ExpandDims_194:0' shape=(None, 1) dtype=int64>, 'corn_oil': <tf.Tensor 'ExpandDims_195:0' shape=(None, 1) dtype=int64>, 'cornbread': <tf.Tensor 'ExpandDims_196:0' shape=(None, 1) dtype=int64>, 'corned_beef': <tf.Tensor 'ExpandDims_197:0' shape=(None, 1) dtype=int64>, 'cornish_hen': <tf.Tensor 'ExpandDims_198:0' shape=(None, 1) dtype=int64>, 'cornmeal': <tf.Tensor 'ExpandDims_199:0' shape=(None, 1) dtype=int64>, 'cornstarch': <tf.Tensor 'ExpandDims_200:0' shape=(None, 1) dtype=int64>, 'cottage_cheese': <tf.Tensor 'ExpandDims_201:0' shape=(None, 1) dtype=int64>, 'couscous': <tf.Tensor 'ExpandDims_202:0' shape=(None, 1) dtype=int64>, 'crab': <tf.Tensor 'ExpandDims_203:0' shape=(None, 1) dtype=int64>, 'crab_claw': <tf.Tensor 'ExpandDims_204:0' shape=(None, 1) dtype=int64>, 'cracker': <tf.Tensor 'ExpandDims_205:0' shape=(None, 1) dtype=int64>, 'craisin': <tf.Tensor 'ExpandDims_206:0' shape=(None, 1) dtype=int64>, 'cranberry': <tf.Tensor 'ExpandDims_207:0' shape=(None, 1) dtype=int64>, 'cranberry_bean': <tf.Tensor 'ExpandDims_208:0' shape=(None, 1) dtype=int64>, 'crayfish': <tf.Tensor 'ExpandDims_209:0' shape=(None, 1) dtype=int64>, 'cream': <tf.Tensor 'ExpandDims_210:0' shape=(None, 1) dtype=int64>, 'cream_cheese': <tf.Tensor 'ExpandDims_211:0' shape=(None, 1) dtype=int64>, 'creamed_corn': <tf.Tensor 'ExpandDims_212:0' shape=(None, 1) dtype=int64>, 'creme_fraiche': <tf.Tensor 'ExpandDims_213:0' shape=(None, 1) dtype=int64>, 'cremini_mushroom': <tf.Tensor 'ExpandDims_214:0' shape=(None, 1) dtype=int64>, 'crouton': <tf.Tensor 'ExpandDims_215:0' shape=(None, 1) dtype=int64>, 'crusty_bread': <tf.Tensor 'ExpandDims_216:0' shape=(None, 1) dtype=int64>, 'crystallized_ginger': <tf.Tensor 'ExpandDims_217:0' shape=(None, 1) dtype=int64>, 'cucumber': <tf.Tensor 'ExpandDims_218:0' shape=(None, 1) dtype=int64>, 'cumin': <tf.Tensor 'ExpandDims_219:0' shape=(None, 1) dtype=int64>, 'currant': <tf.Tensor 'ExpandDims_220:0' shape=(None, 1) dtype=int64>, 'curry_leaf': <tf.Tensor 'ExpandDims_221:0' shape=(None, 1) dtype=int64>, 'curry_paste': <tf.Tensor 'ExpandDims_222:0' shape=(None, 1) dtype=int64>, 'curry_powder': <tf.Tensor 'ExpandDims_223:0' shape=(None, 1) dtype=int64>, 'curry_sauce': <tf.Tensor 'ExpandDims_224:0' shape=(None, 1) dtype=int64>, 'dashi': <tf.Tensor 'ExpandDims_225:0' shape=(None, 1) dtype=int64>, 'dashi_stock': <tf.Tensor 'ExpandDims_226:0' shape=(None, 1) dtype=int64>, 'date': <tf.Tensor 'ExpandDims_227:0' shape=(None, 1) dtype=int64>, 'dehydrated_onion': <tf.Tensor 'ExpandDims_228:0' shape=(None, 1) dtype=int64>, 'demi_glace': <tf.Tensor 'ExpandDims_229:0' shape=(None, 1) dtype=int64>, 'dill': <tf.Tensor 'ExpandDims_230:0' shape=(None, 1) dtype=int64>, 'dried_apricot': <tf.Tensor 'ExpandDims_231:0' shape=(None, 1) dtype=int64>, 'dried_chile_pepper': <tf.Tensor 'ExpandDims_232:0' shape=(None, 1) dtype=int64>, 'dripping': <tf.Tensor 'ExpandDims_233:0' shape=(None, 1) dtype=int64>, 'duck': <tf.Tensor 'ExpandDims_234:0' shape=(None, 1) dtype=int64>, 'edamame': <tf.Tensor 'ExpandDims_235:0' shape=(None, 1) dtype=int64>, 'eel': <tf.Tensor 'ExpandDims_236:0' shape=(None, 1) dtype=int64>, 'egg': <tf.Tensor 'ExpandDims_237:0' shape=(None, 1) dtype=int64>, 'egg_noodles': <tf.Tensor 'ExpandDims_238:0' shape=(None, 1) dtype=int64>, 'egg_roll_wraps': <tf.Tensor 'ExpandDims_239:0' shape=(None, 1) dtype=int64>, 'egg_substitute': <tf.Tensor 'ExpandDims_240:0' shape=(None, 1) dtype=int64>, 'eggplant': <tf.Tensor 'ExpandDims_241:0' shape=(None, 1) dtype=int64>, 'emeril': <tf.Tensor 'ExpandDims_242:0' shape=(None, 1) dtype=int64>, 'emmenthaler_cheese': <tf.Tensor 'ExpandDims_243:0' shape=(None, 1) dtype=int64>, 'enchilada_sauce': <tf.Tensor 'ExpandDims_244:0' shape=(None, 1) dtype=int64>, 'endive': <tf.Tensor 'ExpandDims_245:0' shape=(None, 1) dtype=int64>, 'english_muffin': <tf.Tensor 'ExpandDims_246:0' shape=(None, 1) dtype=int64>, 'english_mustard': <tf.Tensor 'ExpandDims_247:0' shape=(None, 1) dtype=int64>, 'enoki_mushroom': <tf.Tensor 'ExpandDims_248:0' shape=(None, 1) dtype=int64>, 'epazote': <tf.Tensor 'ExpandDims_249:0' shape=(None, 1) dtype=int64>, 'evaporated_milk': <tf.Tensor 'ExpandDims_250:0' shape=(None, 1) dtype=int64>, 'fajita_seasoning_mix': <tf.Tensor 'ExpandDims_251:0' shape=(None, 1) dtype=int64>, 'fat-free_refried_beans': <tf.Tensor 'ExpandDims_252:0' shape=(None, 1) dtype=int64>, 'fava_bean': <tf.Tensor 'ExpandDims_253:0' shape=(None, 1) dtype=int64>, 'fennel': <tf.Tensor 'ExpandDims_254:0' shape=(None, 1) dtype=int64>, 'fennel_seeds': <tf.Tensor 'ExpandDims_255:0' shape=(None, 1) dtype=int64>, 'fenugreek': <tf.Tensor 'ExpandDims_256:0' shape=(None, 1) dtype=int64>, 'fermented_black_bean': <tf.Tensor 'ExpandDims_257:0' shape=(None, 1) dtype=int64>, 'feta_cheese': <tf.Tensor 'ExpandDims_258:0' shape=(None, 1) dtype=int64>, 'fig': <tf.Tensor 'ExpandDims_259:0' shape=(None, 1) dtype=int64>, 'filet_mignon': <tf.Tensor 'ExpandDims_260:0' shape=(None, 1) dtype=int64>, 'fire_oil': <tf.Tensor 'ExpandDims_261:0' shape=(None, 1) dtype=int64>, 'fish': <tf.Tensor 'ExpandDims_262:0' shape=(None, 1) dtype=int64>, 'fish_sauce': <tf.Tensor 'ExpandDims_263:0' shape=(None, 1) dtype=int64>, 'fish_stock': <tf.Tensor 'ExpandDims_264:0' shape=(None, 1) dtype=int64>, 'five_spice_powder': <tf.Tensor 'ExpandDims_265:0' shape=(None, 1) dtype=int64>, 'flat_bread': <tf.Tensor 'ExpandDims_266:0' shape=(None, 1) dtype=int64>, 'flax_seed': <tf.Tensor 'ExpandDims_267:0' shape=(None, 1) dtype=int64>, 'flounder': <tf.Tensor 'ExpandDims_268:0' shape=(None, 1) dtype=int64>, 'flour': <tf.Tensor 'ExpandDims_269:0' shape=(None, 1) dtype=int64>, 'fontina_cheese': <tf.Tensor 'ExpandDims_270:0' shape=(None, 1) dtype=int64>, 'French_beans': <tf.Tensor 'ExpandDims_2:0' shape=(None, 1) dtype=int64>, 'French_bread': <tf.Tensor 'ExpandDims_3:0' shape=(None, 1) dtype=int64>, 'French_dressing': <tf.Tensor 'ExpandDims_4:0' shape=(None, 1) dtype=int64>, 'french_style_green_bean': <tf.Tensor 'ExpandDims_271:0' shape=(None, 1) dtype=int64>, 'fresh_button_mushroom': <tf.Tensor 'ExpandDims_272:0' shape=(None, 1) dtype=int64>, 'frozen_edamame': <tf.Tensor 'ExpandDims_273:0' shape=(None, 1) dtype=int64>, 'fusilli': <tf.Tensor 'ExpandDims_274:0' shape=(None, 1) dtype=int64>, 'galangal': <tf.Tensor 'ExpandDims_275:0' shape=(None, 1) dtype=int64>, 'garlic': <tf.Tensor 'ExpandDims_276:0' shape=(None, 1) dtype=int64>, 'garlic_butter': <tf.Tensor 'ExpandDims_277:0' shape=(None, 1) dtype=int64>, 'garlic_pepper_seasoning': <tf.Tensor 'ExpandDims_278:0' shape=(None, 1) dtype=int64>, 'garlic_powder': <tf.Tensor 'ExpandDims_279:0' shape=(None, 1) dtype=int64>, 'garlic_salt': <tf.Tensor 'ExpandDims_280:0' shape=(None, 1) dtype=int64>, 'gelatin': <tf.Tensor 'ExpandDims_281:0' shape=(None, 1) dtype=int64>, 'genoa_salami': <tf.Tensor 'ExpandDims_282:0' shape=(None, 1) dtype=int64>, 'ghee': <tf.Tensor 'ExpandDims_283:0' shape=(None, 1) dtype=int64>, 'ginger': <tf.Tensor 'ExpandDims_284:0' shape=(None, 1) dtype=int64>, 'ginger-garlic_paste': <tf.Tensor 'ExpandDims_285:0' shape=(None, 1) dtype=int64>, 'gingersnap_cookie': <tf.Tensor 'ExpandDims_286:0' shape=(None, 1) dtype=int64>, 'gingersnaps': <tf.Tensor 'ExpandDims_287:0' shape=(None, 1) dtype=int64>, 'gloucester': <tf.Tensor 'ExpandDims_288:0' shape=(None, 1) dtype=int64>, 'gluten': <tf.Tensor 'ExpandDims_289:0' shape=(None, 1) dtype=int64>, 'goat_cheese': <tf.Tensor 'ExpandDims_290:0' shape=(None, 1) dtype=int64>, 'golden_syrup': <tf.Tensor 'ExpandDims_291:0' shape=(None, 1) dtype=int64>, 'goose': <tf.Tensor 'ExpandDims_292:0' shape=(None, 1) dtype=int64>, 'gorgonzola': <tf.Tensor 'ExpandDims_293:0' shape=(None, 1) dtype=int64>, 'gouda': <tf.Tensor 'ExpandDims_294:0' shape=(None, 1) dtype=int64>, 'grand_marnier': <tf.Tensor 'ExpandDims_295:0' shape=(None, 1) dtype=int64>, 'granulated_garlic': <tf.Tensor 'ExpandDims_296:0' shape=(None, 1) dtype=int64>, 'grape': <tf.Tensor 'ExpandDims_297:0' shape=(None, 1) dtype=int64>, 'grapefruit': <tf.Tensor 'ExpandDims_298:0' shape=(None, 1) dtype=int64>, 'grapeseed_oil': <tf.Tensor 'ExpandDims_299:0' shape=(None, 1) dtype=int64>, 'grating_cheese': <tf.Tensor 'ExpandDims_300:0' shape=(None, 1) dtype=int64>, 'gravy': <tf.Tensor 'ExpandDims_301:0' shape=(None, 1) dtype=int64>, 'great_northern_bean': <tf.Tensor 'ExpandDims_302:0' shape=(None, 1) dtype=int64>, 'Greek_salad_dressing': <tf.Tensor 'ExpandDims_5:0' shape=(None, 1) dtype=int64>, 'green_bean': <tf.Tensor 'ExpandDims_303:0' shape=(None, 1) dtype=int64>, 'green_olive': <tf.Tensor 'ExpandDims_304:0' shape=(None, 1) dtype=int64>, 'green_onion': <tf.Tensor 'ExpandDims_305:0' shape=(None, 1) dtype=int64>, 'greens': <tf.Tensor 'ExpandDims_306:0' shape=(None, 1) dtype=int64>, 'ground_meat': <tf.Tensor 'ExpandDims_307:0' shape=(None, 1) dtype=int64>, 'ground_round': <tf.Tensor 'ExpandDims_308:0' shape=(None, 1) dtype=int64>, 'gruyere': <tf.Tensor 'ExpandDims_309:0' shape=(None, 1) dtype=int64>, 'guacamole': <tf.Tensor 'ExpandDims_310:0' shape=(None, 1) dtype=int64>, 'Guinness_stout': <tf.Tensor 'ExpandDims_6:0' shape=(None, 1) dtype=int64>, 'gumbo': <tf.Tensor 'ExpandDims_311:0' shape=(None, 1) dtype=int64>, 'habanero_chile': <tf.Tensor 'ExpandDims_312:0' shape=(None, 1) dtype=int64>, 'haddock': <tf.Tensor 'ExpandDims_313:0' shape=(None, 1) dtype=int64>, 'hake_fillets': <tf.Tensor 'ExpandDims_314:0' shape=(None, 1) dtype=int64>, 'half_and_half': <tf.Tensor 'ExpandDims_315:0' shape=(None, 1) dtype=int64>, 'halibut': <tf.Tensor 'ExpandDims_316:0' shape=(None, 1) dtype=int64>, 'ham': <tf.Tensor 'ExpandDims_317:0' shape=(None, 1) dtype=int64>, 'ham_hock': <tf.Tensor 'ExpandDims_318:0' shape=(None, 1) dtype=int64>, 'hamburger': <tf.Tensor 'ExpandDims_319:0' shape=(None, 1) dtype=int64>, 'harissa': <tf.Tensor 'ExpandDims_320:0' shape=(None, 1) dtype=int64>, 'hash_browns': <tf.Tensor 'ExpandDims_321:0' shape=(None, 1) dtype=int64>, 'havarti': <tf.Tensor 'ExpandDims_322:0' shape=(None, 1) dtype=int64>, 'hazelnut': <tf.Tensor 'ExpandDims_323:0' shape=(None, 1) dtype=int64>, 'heavy_cream': <tf.Tensor 'ExpandDims_324:0' shape=(None, 1) dtype=int64>, 'herb': <tf.Tensor 'ExpandDims_325:0' shape=(None, 1) dtype=int64>, 'herbes_de_provence': <tf.Tensor 'ExpandDims_326:0' shape=(None, 1) dtype=int64>, 'hoagie_rolls': <tf.Tensor 'ExpandDims_327:0' shape=(None, 1) dtype=int64>, 'hoisin_sauce': <tf.Tensor 'ExpandDims_328:0' shape=(None, 1) dtype=int64>, 'hokkien_noodles': <tf.Tensor 'ExpandDims_329:0' shape=(None, 1) dtype=int64>, 'hominy': <tf.Tensor 'ExpandDims_330:0' shape=(None, 1) dtype=int64>, 'honey': <tf.Tensor 'ExpandDims_331:0' shape=(None, 1) dtype=int64>, 'honey_mustard': <tf.Tensor 'ExpandDims_332:0' shape=(None, 1) dtype=int64>, 'horseradish': <tf.Tensor 'ExpandDims_333:0' shape=(None, 1) dtype=int64>, 'horseradish_sauce': <tf.Tensor 'ExpandDims_334:0' shape=(None, 1) dtype=int64>, 'hot_bean_paste': <tf.Tensor 'ExpandDims_335:0' shape=(None, 1) dtype=int64>, 'hot_pepper_sauce': <tf.Tensor 'ExpandDims_336:0' shape=(None, 1) dtype=int64>, 'hummus': <tf.Tensor 'ExpandDims_337:0' shape=(None, 1) dtype=int64>, 'ice_water': <tf.Tensor 'ExpandDims_338:0' shape=(None, 1) dtype=int64>, 'Italian_bread': <tf.Tensor 'ExpandDims_7:0' shape=(None, 1) dtype=int64>, 'Italian_breadcrumbs': <tf.Tensor 'ExpandDims_8:0' shape=(None, 1) dtype=int64>, 'Italian_rolls': <tf.Tensor 'ExpandDims_9:0' shape=(None, 1) dtype=int64>, 'Italian_salad_dressing': <tf.Tensor 'ExpandDims_10:0' shape=(None, 1) dtype=int64>, 'Italian_spices': <tf.Tensor 'ExpandDims_11:0' shape=(None, 1) dtype=int64>, 'jack_cheese': <tf.Tensor 'ExpandDims_339:0' shape=(None, 1) dtype=int64>, 'jaggery': <tf.Tensor 'ExpandDims_340:0' shape=(None, 1) dtype=int64>, 'jalapeno_pepper': <tf.Tensor 'ExpandDims_341:0' shape=(None, 1) dtype=int64>, 'Japanese_breadcrumbs': <tf.Tensor 'ExpandDims_12:0' shape=(None, 1) dtype=int64>, 'jasmine_rice': <tf.Tensor 'ExpandDims_342:0' shape=(None, 1) dtype=int64>, 'jicama': <tf.Tensor 'ExpandDims_343:0' shape=(None, 1) dtype=int64>, 'juniper_berries': <tf.Tensor 'ExpandDims_344:0' shape=(None, 1) dtype=int64>, 'kaffir_lime': <tf.Tensor 'ExpandDims_345:0' shape=(None, 1) dtype=int64>, 'kalamata_olive': <tf.Tensor 'ExpandDims_346:0' shape=(None, 1) dtype=int64>, 'kale': <tf.Tensor 'ExpandDims_347:0' shape=(None, 1) dtype=int64>, 'kamaboko': <tf.Tensor 'ExpandDims_348:0' shape=(None, 1) dtype=int64>, 'ketchup': <tf.Tensor 'ExpandDims_349:0' shape=(None, 1) dtype=int64>, 'ketjap_manis': <tf.Tensor 'ExpandDims_350:0' shape=(None, 1) dtype=int64>, 'kidney_bean': <tf.Tensor 'ExpandDims_351:0' shape=(None, 1) dtype=int64>, 'kielbasa': <tf.Tensor 'ExpandDims_352:0' shape=(None, 1) dtype=int64>, 'kohlrabi': <tf.Tensor 'ExpandDims_353:0' shape=(None, 1) dtype=int64>, 'kumquat': <tf.Tensor 'ExpandDims_354:0' shape=(None, 1) dtype=int64>, 'lamb': <tf.Tensor 'ExpandDims_355:0' shape=(None, 1) dtype=int64>, 'lambs_kidney': <tf.Tensor 'ExpandDims_356:0' shape=(None, 1) dtype=int64>, 'lancashire_cheese': <tf.Tensor 'ExpandDims_357:0' shape=(None, 1) dtype=int64>, 'lard': <tf.Tensor 'ExpandDims_358:0' shape=(None, 1) dtype=int64>, 'lardon': <tf.Tensor 'ExpandDims_359:0' shape=(None, 1) dtype=int64>, 'lasagna_noodles': <tf.Tensor 'ExpandDims_360:0' shape=(None, 1) dtype=int64>, 'leek': <tf.Tensor 'ExpandDims_361:0' shape=(None, 1) dtype=int64>, 'lemon': <tf.Tensor 'ExpandDims_362:0' shape=(None, 1) dtype=int64>, 'lemon_grass': <tf.Tensor 'ExpandDims_363:0' shape=(None, 1) dtype=int64>, 'lemon_juice': <tf.Tensor 'ExpandDims_364:0' shape=(None, 1) dtype=int64>, 'lentil': <tf.Tensor 'ExpandDims_365:0' shape=(None, 1) dtype=int64>, 'lettuce': <tf.Tensor 'ExpandDims_366:0' shape=(None, 1) dtype=int64>, 'lima_bean': <tf.Tensor 'ExpandDims_367:0' shape=(None, 1) dtype=int64>, 'lime': <tf.Tensor 'ExpandDims_368:0' shape=(None, 1) dtype=int64>, 'liver': <tf.Tensor 'ExpandDims_369:0' shape=(None, 1) dtype=int64>, 'lobster': <tf.Tensor 'ExpandDims_370:0' shape=(None, 1) dtype=int64>, 'long_grain_and_wild_rice_blend': <tf.Tensor 'ExpandDims_371:0' shape=(None, 1) dtype=int64>, 'lotus': <tf.Tensor 'ExpandDims_372:0' shape=(None, 1) dtype=int64>, 'macaroni': <tf.Tensor 'ExpandDims_373:0' shape=(None, 1) dtype=int64>, 'mace': <tf.Tensor 'ExpandDims_374:0' shape=(None, 1) dtype=int64>, 'mackerel': <tf.Tensor 'ExpandDims_375:0' shape=(None, 1) dtype=int64>, 'madeira': <tf.Tensor 'ExpandDims_376:0' shape=(None, 1) dtype=int64>, 'mahi_mahi': <tf.Tensor 'ExpandDims_377:0' shape=(None, 1) dtype=int64>, 'malt_vinegar': <tf.Tensor 'ExpandDims_378:0' shape=(None, 1) dtype=int64>, 'mandarin_orange': <tf.Tensor 'ExpandDims_379:0' shape=(None, 1) dtype=int64>, 'mange_tout': <tf.Tensor 'ExpandDims_380:0' shape=(None, 1) dtype=int64>, 'mango': <tf.Tensor 'ExpandDims_381:0' shape=(None, 1) dtype=int64>, 'manicotti': <tf.Tensor 'ExpandDims_382:0' shape=(None, 1) dtype=int64>, 'manzanilla_olive': <tf.Tensor 'ExpandDims_383:0' shape=(None, 1) dtype=int64>, 'maple_syrup': <tf.Tensor 'ExpandDims_384:0' shape=(None, 1) dtype=int64>, 'margarine': <tf.Tensor 'ExpandDims_385:0' shape=(None, 1) dtype=int64>, 'marinara_sauce': <tf.Tensor 'ExpandDims_386:0' shape=(None, 1) dtype=int64>, 'marjoram': <tf.Tensor 'ExpandDims_387:0' shape=(None, 1) dtype=int64>, 'marmalade': <tf.Tensor 'ExpandDims_388:0' shape=(None, 1) dtype=int64>, 'marmite': <tf.Tensor 'ExpandDims_389:0' shape=(None, 1) dtype=int64>, 'marsala': <tf.Tensor 'ExpandDims_390:0' shape=(None, 1) dtype=int64>, 'masa_harina': <tf.Tensor 'ExpandDims_391:0' shape=(None, 1) dtype=int64>, 'masala': <tf.Tensor 'ExpandDims_392:0' shape=(None, 1) dtype=int64>, 'mascarpone': <tf.Tensor 'ExpandDims_393:0' shape=(None, 1) dtype=int64>, 'masoor_dal': <tf.Tensor 'ExpandDims_394:0' shape=(None, 1) dtype=int64>, 'mayonnaise': <tf.Tensor 'ExpandDims_395:0' shape=(None, 1) dtype=int64>, 'meat_stock': <tf.Tensor 'ExpandDims_396:0' shape=(None, 1) dtype=int64>, 'meatloaf_mix': <tf.Tensor 'ExpandDims_397:0' shape=(None, 1) dtype=int64>, 'Mexican_blend_cheese': <tf.Tensor 'ExpandDims_13:0' shape=(None, 1) dtype=int64>, 'mexicorn': <tf.Tensor 'ExpandDims_398:0' shape=(None, 1) dtype=int64>, 'milk': <tf.Tensor 'ExpandDims_399:0' shape=(None, 1) dtype=int64>, 'mint': <tf.Tensor 'ExpandDims_400:0' shape=(None, 1) dtype=int64>, 'miracle_whip_dressing': <tf.Tensor 'ExpandDims_401:0' shape=(None, 1) dtype=int64>, 'miso': <tf.Tensor 'ExpandDims_402:0' shape=(None, 1) dtype=int64>, 'mixed_spice': <tf.Tensor 'ExpandDims_403:0' shape=(None, 1) dtype=int64>, 'mizithra_cheese': <tf.Tensor 'ExpandDims_404:0' shape=(None, 1) dtype=int64>, 'molasses': <tf.Tensor 'ExpandDims_405:0' shape=(None, 1) dtype=int64>, 'monosodium_glutamate': <tf.Tensor 'ExpandDims_406:0' shape=(None, 1) dtype=int64>, 'moroccan_seasoning': <tf.Tensor 'ExpandDims_407:0' shape=(None, 1) dtype=int64>, 'mortadella': <tf.Tensor 'ExpandDims_408:0' shape=(None, 1) dtype=int64>, 'mozzarella': <tf.Tensor 'ExpandDims_409:0' shape=(None, 1) dtype=int64>, 'mung_beans': <tf.Tensor 'ExpandDims_410:0' shape=(None, 1) dtype=int64>, 'muscovado_sugar': <tf.Tensor 'ExpandDims_411:0' shape=(None, 1) dtype=int64>, 'mushroom': <tf.Tensor 'ExpandDims_412:0' shape=(None, 1) dtype=int64>, 'mushroom_soup': <tf.Tensor 'ExpandDims_413:0' shape=(None, 1) dtype=int64>, 'mussel': <tf.Tensor 'ExpandDims_414:0' shape=(None, 1) dtype=int64>, 'mustard_oil': <tf.Tensor 'ExpandDims_415:0' shape=(None, 1) dtype=int64>, 'mustard_prepared': <tf.Tensor 'ExpandDims_416:0' shape=(None, 1) dtype=int64>, 'mustard_seed_and_powder': <tf.Tensor 'ExpandDims_417:0' shape=(None, 1) dtype=int64>, 'navy_bean': <tf.Tensor 'ExpandDims_418:0' shape=(None, 1) dtype=int64>, 'nectarine': <tf.Tensor 'ExpandDims_419:0' shape=(None, 1) dtype=int64>, 'nigella_seeds': <tf.Tensor 'ExpandDims_420:0' shape=(None, 1) dtype=int64>, 'nori': <tf.Tensor 'ExpandDims_421:0' shape=(None, 1) dtype=int64>, 'nutmeg': <tf.Tensor 'ExpandDims_422:0' shape=(None, 1) dtype=int64>, 'oatmeal': <tf.Tensor 'ExpandDims_423:0' shape=(None, 1) dtype=int64>, 'okra': <tf.Tensor 'ExpandDims_424:0' shape=(None, 1) dtype=int64>, 'Old_Bay_Seasoning': <tf.Tensor 'ExpandDims_14:0' shape=(None, 1) dtype=int64>, 'olive': <tf.Tensor 'ExpandDims_425:0' shape=(None, 1) dtype=int64>, 'olive_oil': <tf.Tensor 'ExpandDims_426:0' shape=(None, 1) dtype=int64>, 'onion': <tf.Tensor 'ExpandDims_427:0' shape=(None, 1) dtype=int64>, 'onion_powder': <tf.Tensor 'ExpandDims_428:0' shape=(None, 1) dtype=int64>, 'onion_soup': <tf.Tensor 'ExpandDims_429:0' shape=(None, 1) dtype=int64>, 'orange': <tf.Tensor 'ExpandDims_430:0' shape=(None, 1) dtype=int64>, 'orange_juice': <tf.Tensor 'ExpandDims_431:0' shape=(None, 1) dtype=int64>, 'orange_rind': <tf.Tensor 'ExpandDims_432:0' shape=(None, 1) dtype=int64>, 'oregano': <tf.Tensor 'ExpandDims_433:0' shape=(None, 1) dtype=int64>, 'oxtails': <tf.Tensor 'ExpandDims_434:0' shape=(None, 1) dtype=int64>, 'oyster_sauce': <tf.Tensor 'ExpandDims_435:0' shape=(None, 1) dtype=int64>, 'oysters': <tf.Tensor 'ExpandDims_436:0' shape=(None, 1) dtype=int64>, 'Pace_Picante_Sauce': <tf.Tensor 'ExpandDims_15:0' shape=(None, 1) dtype=int64>, 'palm_sugar': <tf.Tensor 'ExpandDims_437:0' shape=(None, 1) dtype=int64>, 'pancetta': <tf.Tensor 'ExpandDims_438:0' shape=(None, 1) dtype=int64>, 'panir': <tf.Tensor 'ExpandDims_439:0' shape=(None, 1) dtype=int64>, 'panko': <tf.Tensor 'ExpandDims_440:0' shape=(None, 1) dtype=int64>, 'papaya': <tf.Tensor 'ExpandDims_441:0' shape=(None, 1) dtype=int64>, 'paprika': <tf.Tensor 'ExpandDims_442:0' shape=(None, 1) dtype=int64>, 'parchment_paper': <tf.Tensor 'ExpandDims_443:0' shape=(None, 1) dtype=int64>, 'parmesan_cheese': <tf.Tensor 'ExpandDims_444:0' shape=(None, 1) dtype=int64>, 'parsley': <tf.Tensor 'ExpandDims_445:0' shape=(None, 1) dtype=int64>, 'parsnip': <tf.Tensor 'ExpandDims_446:0' shape=(None, 1) dtype=int64>, 'passata': <tf.Tensor 'ExpandDims_447:0' shape=(None, 1) dtype=int64>, 'pasta': <tf.Tensor 'ExpandDims_448:0' shape=(None, 1) dtype=int64>, 'pasta_sauce': <tf.Tensor 'ExpandDims_449:0' shape=(None, 1) dtype=int64>, 'pastis': <tf.Tensor 'ExpandDims_450:0' shape=(None, 1) dtype=int64>, 'pastry_dough': <tf.Tensor 'ExpandDims_451:0' shape=(None, 1) dtype=int64>, 'pea': <tf.Tensor 'ExpandDims_452:0' shape=(None, 1) dtype=int64>, 'peanut': <tf.Tensor 'ExpandDims_453:0' shape=(None, 1) dtype=int64>, 'peanut_butter': <tf.Tensor 'ExpandDims_454:0' shape=(None, 1) dtype=int64>, 'peanut_oil': <tf.Tensor 'ExpandDims_455:0' shape=(None, 1) dtype=int64>, 'peanut_sauce': <tf.Tensor 'ExpandDims_456:0' shape=(None, 1) dtype=int64>, 'pear': <tf.Tensor 'ExpandDims_457:0' shape=(None, 1) dtype=int64>, 'pearl_onion': <tf.Tensor 'ExpandDims_458:0' shape=(None, 1) dtype=int64>, 'pecan': <tf.Tensor 'ExpandDims_459:0' shape=(None, 1) dtype=int64>, 'pecorino_cheese': <tf.Tensor 'ExpandDims_460:0' shape=(None, 1) dtype=int64>, 'penne_rigate': <tf.Tensor 'ExpandDims_461:0' shape=(None, 1) dtype=int64>, 'pepper': <tf.Tensor 'ExpandDims_462:0' shape=(None, 1) dtype=int64>, 'pepperoncini_pepper': <tf.Tensor 'ExpandDims_463:0' shape=(None, 1) dtype=int64>, 'pepperoni': <tf.Tensor 'ExpandDims_464:0' shape=(None, 1) dtype=int64>, 'pesto': <tf.Tensor 'ExpandDims_465:0' shape=(None, 1) dtype=int64>, 'phyllo_dough': <tf.Tensor 'ExpandDims_466:0' shape=(None, 1) dtype=int64>, 'picante_sauce': <tf.Tensor 'ExpandDims_467:0' shape=(None, 1) dtype=int64>, 'pickle': <tf.Tensor 'ExpandDims_468:0' shape=(None, 1) dtype=int64>, 'pickling_spices': <tf.Tensor 'ExpandDims_469:0' shape=(None, 1) dtype=int64>, 'pie_crust': <tf.Tensor 'ExpandDims_470:0' shape=(None, 1) dtype=int64>, 'pierogi': <tf.Tensor 'ExpandDims_471:0' shape=(None, 1) dtype=int64>, 'pimiento': <tf.Tensor 'ExpandDims_472:0' shape=(None, 1) dtype=int64>, 'pine_nuts': <tf.Tensor 'ExpandDims_473:0' shape=(None, 1) dtype=int64>, 'pineapple': <tf.Tensor 'ExpandDims_474:0' shape=(None, 1) dtype=int64>, 'pink_bean': <tf.Tensor 'ExpandDims_475:0' shape=(None, 1) dtype=int64>, 'pinto_bean': <tf.Tensor 'ExpandDims_476:0' shape=(None, 1) dtype=int64>, 'piquillo_pepper': <tf.Tensor 'ExpandDims_477:0' shape=(None, 1) dtype=int64>, 'pistachio': <tf.Tensor 'ExpandDims_478:0' shape=(None, 1) dtype=int64>, 'pita_bread': <tf.Tensor 'ExpandDims_479:0' shape=(None, 1) dtype=int64>, 'pitted_prunes': <tf.Tensor 'ExpandDims_480:0' shape=(None, 1) dtype=int64>, 'pizza_crust': <tf.Tensor 'ExpandDims_481:0' shape=(None, 1) dtype=int64>, 'pizza_dough': <tf.Tensor 'ExpandDims_482:0' shape=(None, 1) dtype=int64>, 'pizza_sauce': <tf.Tensor 'ExpandDims_483:0' shape=(None, 1) dtype=int64>, 'plum': <tf.Tensor 'ExpandDims_484:0' shape=(None, 1) dtype=int64>, 'plum_tomato': <tf.Tensor 'ExpandDims_485:0' shape=(None, 1) dtype=int64>, 'poblano_pepper': <tf.Tensor 'ExpandDims_486:0' shape=(None, 1) dtype=int64>, 'polenta': <tf.Tensor 'ExpandDims_487:0' shape=(None, 1) dtype=int64>, 'pomegranate': <tf.Tensor 'ExpandDims_488:0' shape=(None, 1) dtype=int64>, 'ponzu': <tf.Tensor 'ExpandDims_489:0' shape=(None, 1) dtype=int64>, 'poppy_seed': <tf.Tensor 'ExpandDims_490:0' shape=(None, 1) dtype=int64>, 'porcino': <tf.Tensor 'ExpandDims_491:0' shape=(None, 1) dtype=int64>, 'pork': <tf.Tensor 'ExpandDims_492:0' shape=(None, 1) dtype=int64>, 'portabella': <tf.Tensor 'ExpandDims_493:0' shape=(None, 1) dtype=int64>, 'potato': <tf.Tensor 'ExpandDims_494:0' shape=(None, 1) dtype=int64>, 'potato_starch': <tf.Tensor 'ExpandDims_495:0' shape=(None, 1) dtype=int64>, 'poultry_seasoning': <tf.Tensor 'ExpandDims_496:0' shape=(None, 1) dtype=int64>, 'powdered_sugar': <tf.Tensor 'ExpandDims_497:0' shape=(None, 1) dtype=int64>, 'prawn': <tf.Tensor 'ExpandDims_498:0' shape=(None, 1) dtype=int64>, 'prepared_horseradish': <tf.Tensor 'ExpandDims_499:0' shape=(None, 1) dtype=int64>, 'preserved_lemon': <tf.Tensor 'ExpandDims_500:0' shape=(None, 1) dtype=int64>, 'processed_cheese': <tf.Tensor 'ExpandDims_501:0' shape=(None, 1) dtype=int64>, 'prosciutto': <tf.Tensor 'ExpandDims_502:0' shape=(None, 1) dtype=int64>, 'provolone': <tf.Tensor 'ExpandDims_503:0' shape=(None, 1) dtype=int64>, 'prune': <tf.Tensor 'ExpandDims_504:0' shape=(None, 1) dtype=int64>, 'puff_pastry': <tf.Tensor 'ExpandDims_505:0' shape=(None, 1) dtype=int64>, 'pumpkin': <tf.Tensor 'ExpandDims_506:0' shape=(None, 1) dtype=int64>, 'pumpkin_pie_spice': <tf.Tensor 'ExpandDims_507:0' shape=(None, 1) dtype=int64>, 'quark': <tf.Tensor 'ExpandDims_508:0' shape=(None, 1) dtype=int64>, 'queso': <tf.Tensor 'ExpandDims_509:0' shape=(None, 1) dtype=int64>, 'quinoa': <tf.Tensor 'ExpandDims_510:0' shape=(None, 1) dtype=int64>, 'rabbit': <tf.Tensor 'ExpandDims_511:0' shape=(None, 1) dtype=int64>, 'radicchio': <tf.Tensor 'ExpandDims_512:0' shape=(None, 1) dtype=int64>, 'radish': <tf.Tensor 'ExpandDims_513:0' shape=(None, 1) dtype=int64>, 'raisin': <tf.Tensor 'ExpandDims_514:0' shape=(None, 1) dtype=int64>, 'ramen_noodles': <tf.Tensor 'ExpandDims_515:0' shape=(None, 1) dtype=int64>, 'Ranch_salad_dressing': <tf.Tensor 'ExpandDims_16:0' shape=(None, 1) dtype=int64>, 'rapeseed_oil': <tf.Tensor 'ExpandDims_516:0' shape=(None, 1) dtype=int64>, 'ras_el_hanout': <tf.Tensor 'ExpandDims_517:0' shape=(None, 1) dtype=int64>, 'raw_sugar': <tf.Tensor 'ExpandDims_518:0' shape=(None, 1) dtype=int64>, 'red_beans': <tf.Tensor 'ExpandDims_519:0' shape=(None, 1) dtype=int64>, 'red_chile_powder': <tf.Tensor 'ExpandDims_520:0' shape=(None, 1) dtype=int64>, 'red_food_coloring': <tf.Tensor 'ExpandDims_521:0' shape=(None, 1) dtype=int64>, 'red_pepper_flakes': <tf.Tensor 'ExpandDims_522:0' shape=(None, 1) dtype=int64>, 'red_pepper_sauce': <tf.Tensor 'ExpandDims_523:0' shape=(None, 1) dtype=int64>, 'red_snapper': <tf.Tensor 'ExpandDims_524:0' shape=(None, 1) dtype=int64>, 'refried_beans': <tf.Tensor 'ExpandDims_525:0' shape=(None, 1) dtype=int64>, 'rhubarb': <tf.Tensor 'ExpandDims_526:0' shape=(None, 1) dtype=int64>, 'rice': <tf.Tensor 'ExpandDims_527:0' shape=(None, 1) dtype=int64>, 'rice_flour': <tf.Tensor 'ExpandDims_528:0' shape=(None, 1) dtype=int64>, 'rice_noodles': <tf.Tensor 'ExpandDims_529:0' shape=(None, 1) dtype=int64>, 'rice_vermicelli': <tf.Tensor 'ExpandDims_530:0' shape=(None, 1) dtype=int64>, 'rice_vinegar': <tf.Tensor 'ExpandDims_531:0' shape=(None, 1) dtype=int64>, 'rice_wine': <tf.Tensor 'ExpandDims_532:0' shape=(None, 1) dtype=int64>, 'ricotta_cheese': <tf.Tensor 'ExpandDims_533:0' shape=(None, 1) dtype=int64>, 'roll': <tf.Tensor 'ExpandDims_534:0' shape=(None, 1) dtype=int64>, 'romaine_lettuce': <tf.Tensor 'ExpandDims_535:0' shape=(None, 1) dtype=int64>, 'romano_cheese': <tf.Tensor 'ExpandDims_536:0' shape=(None, 1) dtype=int64>, 'rose_water': <tf.Tensor 'ExpandDims_537:0' shape=(None, 1) dtype=int64>, 'rosemary': <tf.Tensor 'ExpandDims_538:0' shape=(None, 1) dtype=int64>, 'rotel_tomatoes': <tf.Tensor 'ExpandDims_539:0' shape=(None, 1) dtype=int64>, 'rum': <tf.Tensor 'ExpandDims_540:0' shape=(None, 1) dtype=int64>, 'rutabaga': <tf.Tensor 'ExpandDims_541:0' shape=(None, 1) dtype=int64>, 'rye_bread': <tf.Tensor 'ExpandDims_542:0' shape=(None, 1) dtype=int64>, 'saffron': <tf.Tensor 'ExpandDims_543:0' shape=(None, 1) dtype=int64>, 'sage': <tf.Tensor 'ExpandDims_544:0' shape=(None, 1) dtype=int64>, 'salmon': <tf.Tensor 'ExpandDims_545:0' shape=(None, 1) dtype=int64>, 'salsa': <tf.Tensor 'ExpandDims_546:0' shape=(None, 1) dtype=int64>, 'salsify': <tf.Tensor 'ExpandDims_547:0' shape=(None, 1) dtype=int64>, 'salt': <tf.Tensor 'ExpandDims_548:0' shape=(None, 1) dtype=int64>, 'salt_pork': <tf.Tensor 'ExpandDims_549:0' shape=(None, 1) dtype=int64>, 'sambal': <tf.Tensor 'ExpandDims_550:0' shape=(None, 1) dtype=int64>, 'sardine': <tf.Tensor 'ExpandDims_551:0' shape=(None, 1) dtype=int64>, 'satay': <tf.Tensor 'ExpandDims_552:0' shape=(None, 1) dtype=int64>, 'sauerkraut': <tf.Tensor 'ExpandDims_553:0' shape=(None, 1) dtype=int64>, 'sausage': <tf.Tensor 'ExpandDims_554:0' shape=(None, 1) dtype=int64>, 'savory': <tf.Tensor 'ExpandDims_555:0' shape=(None, 1) dtype=int64>, 'sazon': <tf.Tensor 'ExpandDims_556:0' shape=(None, 1) dtype=int64>, 'sazon_goya': <tf.Tensor 'ExpandDims_557:0' shape=(None, 1) dtype=int64>, 'scallop': <tf.Tensor 'ExpandDims_558:0' shape=(None, 1) dtype=int64>, 'screwpine_leaves': <tf.Tensor 'ExpandDims_559:0' shape=(None, 1) dtype=int64>, 'sea_cucumber': <tf.Tensor 'ExpandDims_560:0' shape=(None, 1) dtype=int64>, 'seafood': <tf.Tensor 'ExpandDims_561:0' shape=(None, 1) dtype=int64>, 'seitan': <tf.Tensor 'ExpandDims_562:0' shape=(None, 1) dtype=int64>, 'semolina': <tf.Tensor 'ExpandDims_563:0' shape=(None, 1) dtype=int64>, 'serrano_pepper': <tf.Tensor 'ExpandDims_564:0' shape=(None, 1) dtype=int64>, 'sesame_oil': <tf.Tensor 'ExpandDims_565:0' shape=(None, 1) dtype=int64>, 'sesame_seed': <tf.Tensor 'ExpandDims_566:0' shape=(None, 1) dtype=int64>, 'shallot': <tf.Tensor 'ExpandDims_567:0' shape=(None, 1) dtype=int64>, 'sherry': <tf.Tensor 'ExpandDims_568:0' shape=(None, 1) dtype=int64>, 'shimeji_mushroom': <tf.Tensor 'ExpandDims_569:0' shape=(None, 1) dtype=int64>, 'shirataki_noodles': <tf.Tensor 'ExpandDims_570:0' shape=(None, 1) dtype=int64>, 'shitake': <tf.Tensor 'ExpandDims_571:0' shape=(None, 1) dtype=int64>, 'shortcrust_pastry': <tf.Tensor 'ExpandDims_572:0' shape=(None, 1) dtype=int64>, 'shortening': <tf.Tensor 'ExpandDims_573:0' shape=(None, 1) dtype=int64>, 'shoyu': <tf.Tensor 'ExpandDims_574:0' shape=(None, 1) dtype=int64>, 'shrimp': <tf.Tensor 'ExpandDims_575:0' shape=(None, 1) dtype=int64>, 'shrimp_paste': <tf.Tensor 'ExpandDims_576:0' shape=(None, 1) dtype=int64>, 'sliced_pepperoni': <tf.Tensor 'ExpandDims_577:0' shape=(None, 1) dtype=int64>, 'smen': <tf.Tensor 'ExpandDims_578:0' shape=(None, 1) dtype=int64>, 'smoke_seasoning': <tf.Tensor 'ExpandDims_579:0' shape=(None, 1) dtype=int64>, 'smoked_salmon': <tf.Tensor 'ExpandDims_580:0' shape=(None, 1) dtype=int64>, 'snap_pea': <tf.Tensor 'ExpandDims_581:0' shape=(None, 1) dtype=int64>, 'snow_pea': <tf.Tensor 'ExpandDims_582:0' shape=(None, 1) dtype=int64>, 'soba_noodles': <tf.Tensor 'ExpandDims_583:0' shape=(None, 1) dtype=int64>, 'sofrito_sauce': <tf.Tensor 'ExpandDims_584:0' shape=(None, 1) dtype=int64>, 'sole': <tf.Tensor 'ExpandDims_585:0' shape=(None, 1) dtype=int64>, 'sorrel': <tf.Tensor 'ExpandDims_586:0' shape=(None, 1) dtype=int64>, 'sour_cream': <tf.Tensor 'ExpandDims_587:0' shape=(None, 1) dtype=int64>, 'sourdough_bread': <tf.Tensor 'ExpandDims_588:0' shape=(None, 1) dtype=int64>, 'soy_sauce': <tf.Tensor 'ExpandDims_589:0' shape=(None, 1) dtype=int64>, 'soybean': <tf.Tensor 'ExpandDims_590:0' shape=(None, 1) dtype=int64>, 'soymilk': <tf.Tensor 'ExpandDims_591:0' shape=(None, 1) dtype=int64>, 'spaetzle': <tf.Tensor 'ExpandDims_592:0' shape=(None, 1) dtype=int64>, 'spaghetti_squash': <tf.Tensor 'ExpandDims_593:0' shape=(None, 1) dtype=int64>, 'spanish_cheese': <tf.Tensor 'ExpandDims_594:0' shape=(None, 1) dtype=int64>, 'spanish_chorizo': <tf.Tensor 'ExpandDims_595:0' shape=(None, 1) dtype=int64>, 'spareribs': <tf.Tensor 'ExpandDims_596:0' shape=(None, 1) dtype=int64>, 'spearmint': <tf.Tensor 'ExpandDims_597:0' shape=(None, 1) dtype=int64>, 'specialty_cheese': <tf.Tensor 'ExpandDims_598:0' shape=(None, 1) dtype=int64>, 'spinach': <tf.Tensor 'ExpandDims_599:0' shape=(None, 1) dtype=int64>, 'squash': <tf.Tensor 'ExpandDims_600:0' shape=(None, 1) dtype=int64>, 'squid': <tf.Tensor 'ExpandDims_601:0' shape=(None, 1) dtype=int64>, 'sriracha_hot_sauce': <tf.Tensor 'ExpandDims_602:0' shape=(None, 1) dtype=int64>, 'star_anise': <tf.Tensor 'ExpandDims_603:0' shape=(None, 1) dtype=int64>, 'steaks': <tf.Tensor 'ExpandDims_604:0' shape=(None, 1) dtype=int64>, 'stilton': <tf.Tensor 'ExpandDims_605:0' shape=(None, 1) dtype=int64>, 'stir_fry_vegetables': <tf.Tensor 'ExpandDims_606:0' shape=(None, 1) dtype=int64>, 'stock': <tf.Tensor 'ExpandDims_607:0' shape=(None, 1) dtype=int64>, 'straw_mushroom': <tf.Tensor 'ExpandDims_608:0' shape=(None, 1) dtype=int64>, 'strawberry': <tf.Tensor 'ExpandDims_609:0' shape=(None, 1) dtype=int64>, 'stuffing': <tf.Tensor 'ExpandDims_610:0' shape=(None, 1) dtype=int64>, 'suet': <tf.Tensor 'ExpandDims_611:0' shape=(None, 1) dtype=int64>, 'sugar': <tf.Tensor 'ExpandDims_612:0' shape=(None, 1) dtype=int64>, 'sultana': <tf.Tensor 'ExpandDims_613:0' shape=(None, 1) dtype=int64>, 'sumac': <tf.Tensor 'ExpandDims_614:0' shape=(None, 1) dtype=int64>, 'sunflower_oil': <tf.Tensor 'ExpandDims_615:0' shape=(None, 1) dtype=int64>, 'sunflower_seed': <tf.Tensor 'ExpandDims_616:0' shape=(None, 1) dtype=int64>, 'sushi_rice': <tf.Tensor 'ExpandDims_617:0' shape=(None, 1) dtype=int64>, 'sweet_pepper': <tf.Tensor 'ExpandDims_618:0' shape=(None, 1) dtype=int64>, 'sweet_potato': <tf.Tensor 'ExpandDims_619:0' shape=(None, 1) dtype=int64>, 'swiss_cheese': <tf.Tensor 'ExpandDims_620:0' shape=(None, 1) dtype=int64>, 'swordfish': <tf.Tensor 'ExpandDims_621:0' shape=(None, 1) dtype=int64>, 'szechuan_peppercorn': <tf.Tensor 'ExpandDims_622:0' shape=(None, 1) dtype=int64>, 'tabasco_sauce': <tf.Tensor 'ExpandDims_623:0' shape=(None, 1) dtype=int64>, 'taco_seasoning': <tf.Tensor 'ExpandDims_624:0' shape=(None, 1) dtype=int64>, 'tahini': <tf.Tensor 'ExpandDims_625:0' shape=(None, 1) dtype=int64>, 'tamarind': <tf.Tensor 'ExpandDims_626:0' shape=(None, 1) dtype=int64>, 'tapioca': <tf.Tensor 'ExpandDims_627:0' shape=(None, 1) dtype=int64>, 'tarragon': <tf.Tensor 'ExpandDims_628:0' shape=(None, 1) dtype=int64>, 'tatsoi': <tf.Tensor 'ExpandDims_629:0' shape=(None, 1) dtype=int64>, 'tea': <tf.Tensor 'ExpandDims_630:0' shape=(None, 1) dtype=int64>, 'tempeh': <tf.Tensor 'ExpandDims_631:0' shape=(None, 1) dtype=int64>, 'tequila': <tf.Tensor 'ExpandDims_632:0' shape=(None, 1) dtype=int64>, 'teriyaki_marinade': <tf.Tensor 'ExpandDims_633:0' shape=(None, 1) dtype=int64>, 'teriyaki_sauce': <tf.Tensor 'ExpandDims_634:0' shape=(None, 1) dtype=int64>, 'Thai_chile_pepper': <tf.Tensor 'ExpandDims_17:0' shape=(None, 1) dtype=int64>, 'Thai_curry_paste': <tf.Tensor 'ExpandDims_18:0' shape=(None, 1) dtype=int64>, 'thousand_island_dressing': <tf.Tensor 'ExpandDims_635:0' shape=(None, 1) dtype=int64>, 'thyme': <tf.Tensor 'ExpandDims_636:0' shape=(None, 1) dtype=int64>, 'tilapia_fillets': <tf.Tensor 'ExpandDims_637:0' shape=(None, 1) dtype=int64>, 'tofu': <tf.Tensor 'ExpandDims_638:0' shape=(None, 1) dtype=int64>, 'togarashi': <tf.Tensor 'ExpandDims_639:0' shape=(None, 1) dtype=int64>, 'tomatillos': <tf.Tensor 'ExpandDims_640:0' shape=(None, 1) dtype=int64>, 'tomato': <tf.Tensor 'ExpandDims_641:0' shape=(None, 1) dtype=int64>, 'tomato_juice': <tf.Tensor 'ExpandDims_642:0' shape=(None, 1) dtype=int64>, 'tomato_paste': <tf.Tensor 'ExpandDims_643:0' shape=(None, 1) dtype=int64>, 'tomato_puree': <tf.Tensor 'ExpandDims_644:0' shape=(None, 1) dtype=int64>, 'tomato_sauce': <tf.Tensor 'ExpandDims_645:0' shape=(None, 1) dtype=int64>, 'tomato_soup': <tf.Tensor 'ExpandDims_646:0' shape=(None, 1) dtype=int64>, 'toor_dal': <tf.Tensor 'ExpandDims_647:0' shape=(None, 1) dtype=int64>, 'tortilla': <tf.Tensor 'ExpandDims_648:0' shape=(None, 1) dtype=int64>, 'trout': <tf.Tensor 'ExpandDims_649:0' shape=(None, 1) dtype=int64>, 'truffle': <tf.Tensor 'ExpandDims_650:0' shape=(None, 1) dtype=int64>, 'truffle_oil': <tf.Tensor 'ExpandDims_651:0' shape=(None, 1) dtype=int64>, 'tuna': <tf.Tensor 'ExpandDims_652:0' shape=(None, 1) dtype=int64>, 'turbinado_sugar': <tf.Tensor 'ExpandDims_653:0' shape=(None, 1) dtype=int64>, 'turkey': <tf.Tensor 'ExpandDims_654:0' shape=(None, 1) dtype=int64>, 'turmeric': <tf.Tensor 'ExpandDims_655:0' shape=(None, 1) dtype=int64>, 'turnip': <tf.Tensor 'ExpandDims_656:0' shape=(None, 1) dtype=int64>, 'udon_noodles': <tf.Tensor 'ExpandDims_657:0' shape=(None, 1) dtype=int64>, 'umeboshi': <tf.Tensor 'ExpandDims_658:0' shape=(None, 1) dtype=int64>, 'unsweetened_chocolate': <tf.Tensor 'ExpandDims_659:0' shape=(None, 1) dtype=int64>, 'urad_dal': <tf.Tensor 'ExpandDims_660:0' shape=(None, 1) dtype=int64>, 'vanilla': <tf.Tensor 'ExpandDims_661:0' shape=(None, 1) dtype=int64>, 'veal': <tf.Tensor 'ExpandDims_662:0' shape=(None, 1) dtype=int64>, 'vegan_brand_names': <tf.Tensor 'ExpandDims_663:0' shape=(None, 1) dtype=int64>, 'vegetable_juice': <tf.Tensor 'ExpandDims_664:0' shape=(None, 1) dtype=int64>, 'vegetable_oil': <tf.Tensor 'ExpandDims_665:0' shape=(None, 1) dtype=int64>, 'vegetable_stock': <tf.Tensor 'ExpandDims_666:0' shape=(None, 1) dtype=int64>, 'vegetarian_brand_names': <tf.Tensor 'ExpandDims_667:0' shape=(None, 1) dtype=int64>, 'velveeta': <tf.Tensor 'ExpandDims_668:0' shape=(None, 1) dtype=int64>, 'venison': <tf.Tensor 'ExpandDims_669:0' shape=(None, 1) dtype=int64>, 'vermicelli': <tf.Tensor 'ExpandDims_670:0' shape=(None, 1) dtype=int64>, 'vermouth': <tf.Tensor 'ExpandDims_671:0' shape=(None, 1) dtype=int64>, 'Vidalia_onion': <tf.Tensor 'ExpandDims_19:0' shape=(None, 1) dtype=int64>, 'vinegar': <tf.Tensor 'ExpandDims_672:0' shape=(None, 1) dtype=int64>, 'vital_wheat_gluten': <tf.Tensor 'ExpandDims_673:0' shape=(None, 1) dtype=int64>, 'vodka': <tf.Tensor 'ExpandDims_674:0' shape=(None, 1) dtype=int64>, 'walnut': <tf.Tensor 'ExpandDims_675:0' shape=(None, 1) dtype=int64>, 'wasabi': <tf.Tensor 'ExpandDims_676:0' shape=(None, 1) dtype=int64>, 'washed_rind_cheese': <tf.Tensor 'ExpandDims_677:0' shape=(None, 1) dtype=int64>, 'water': <tf.Tensor 'ExpandDims_678:0' shape=(None, 1) dtype=int64>, 'water_chestnut': <tf.Tensor 'ExpandDims_679:0' shape=(None, 1) dtype=int64>, 'watercress': <tf.Tensor 'ExpandDims_680:0' shape=(None, 1) dtype=int64>, 'watermelon': <tf.Tensor 'ExpandDims_681:0' shape=(None, 1) dtype=int64>, 'wheat_bread': <tf.Tensor 'ExpandDims_682:0' shape=(None, 1) dtype=int64>, 'whipping_cream': <tf.Tensor 'ExpandDims_683:0' shape=(None, 1) dtype=int64>, 'white_bean': <tf.Tensor 'ExpandDims_684:0' shape=(None, 1) dtype=int64>, 'white_bread': <tf.Tensor 'ExpandDims_685:0' shape=(None, 1) dtype=int64>, 'white_cheese': <tf.Tensor 'ExpandDims_686:0' shape=(None, 1) dtype=int64>, 'white_fish': <tf.Tensor 'ExpandDims_687:0' shape=(None, 1) dtype=int64>, 'white_onion': <tf.Tensor 'ExpandDims_688:0' shape=(None, 1) dtype=int64>, 'white_peppercorns': <tf.Tensor 'ExpandDims_689:0' shape=(None, 1) dtype=int64>, 'white_sesame_seeds': <tf.Tensor 'ExpandDims_690:0' shape=(None, 1) dtype=int64>, 'whole_grain_mustard': <tf.Tensor 'ExpandDims_691:0' shape=(None, 1) dtype=int64>, 'whole_milk': <tf.Tensor 'ExpandDims_692:0' shape=(None, 1) dtype=int64>, 'wild_rice': <tf.Tensor 'ExpandDims_693:0' shape=(None, 1) dtype=int64>, 'wine': <tf.Tensor 'ExpandDims_694:0' shape=(None, 1) dtype=int64>, 'wine_vinegar': <tf.Tensor 'ExpandDims_695:0' shape=(None, 1) dtype=int64>, 'winter_squash': <tf.Tensor 'ExpandDims_696:0' shape=(None, 1) dtype=int64>, 'worcestershire_sauce': <tf.Tensor 'ExpandDims_697:0' shape=(None, 1) dtype=int64>, 'xanthan_gum': <tf.Tensor 'ExpandDims_698:0' shape=(None, 1) dtype=int64>, 'yams': <tf.Tensor 'ExpandDims_699:0' shape=(None, 1) dtype=int64>, 'yeast': <tf.Tensor 'ExpandDims_700:0' shape=(None, 1) dtype=int64>, 'yellow_curry_paste': <tf.Tensor 'ExpandDims_701:0' shape=(None, 1) dtype=int64>, 'yellow_food_coloring': <tf.Tensor 'ExpandDims_702:0' shape=(None, 1) dtype=int64>, 'yellow_split_pea': <tf.Tensor 'ExpandDims_703:0' shape=(None, 1) dtype=int64>, 'yellow_squash': <tf.Tensor 'ExpandDims_704:0' shape=(None, 1) dtype=int64>, 'yogurt': <tf.Tensor 'ExpandDims_705:0' shape=(None, 1) dtype=int64>, 'zaatar': <tf.Tensor 'ExpandDims_706:0' shape=(None, 1) dtype=int64>, 'zest': <tf.Tensor 'ExpandDims_707:0' shape=(None, 1) dtype=int64>, 'zucchini': <tf.Tensor 'ExpandDims_708:0' shape=(None, 1) dtype=int64>}\n",
            "Consider rewriting this model with the Functional API.\n",
            "WARNING:tensorflow:Layers in a Sequential model should only have a single input tensor, but we receive a <class 'dict'> input: {'acorn_squash': <tf.Tensor 'ExpandDims_20:0' shape=(None, 1) dtype=int64>, 'adobo': <tf.Tensor 'ExpandDims_21:0' shape=(None, 1) dtype=int64>, 'african_birdseye_chile_pepper': <tf.Tensor 'ExpandDims_22:0' shape=(None, 1) dtype=int64>, 'ale': <tf.Tensor 'ExpandDims_23:0' shape=(None, 1) dtype=int64>, 'aleppo_pepper': <tf.Tensor 'ExpandDims_24:0' shape=(None, 1) dtype=int64>, 'alfalfa_sprouts': <tf.Tensor 'ExpandDims_25:0' shape=(None, 1) dtype=int64>, 'alfredo_sauce': <tf.Tensor 'ExpandDims_26:0' shape=(None, 1) dtype=int64>, 'allspice': <tf.Tensor 'ExpandDims_27:0' shape=(None, 1) dtype=int64>, 'almond': <tf.Tensor 'ExpandDims_28:0' shape=(None, 1) dtype=int64>, 'almond_butter': <tf.Tensor 'ExpandDims_29:0' shape=(None, 1) dtype=int64>, 'amaretti': <tf.Tensor 'ExpandDims_30:0' shape=(None, 1) dtype=int64>, 'amchoor_powder': <tf.Tensor 'ExpandDims_31:0' shape=(None, 1) dtype=int64>, 'ancho_chile_pepper': <tf.Tensor 'ExpandDims_32:0' shape=(None, 1) dtype=int64>, 'anchovy': <tf.Tensor 'ExpandDims_33:0' shape=(None, 1) dtype=int64>, 'anchovy_paste': <tf.Tensor 'ExpandDims_34:0' shape=(None, 1) dtype=int64>, 'andouille_sausage': <tf.Tensor 'ExpandDims_35:0' shape=(None, 1) dtype=int64>, 'anise': <tf.Tensor 'ExpandDims_36:0' shape=(None, 1) dtype=int64>, 'apple': <tf.Tensor 'ExpandDims_37:0' shape=(None, 1) dtype=int64>, 'apple_butter': <tf.Tensor 'ExpandDims_38:0' shape=(None, 1) dtype=int64>, 'apple_juice': <tf.Tensor 'ExpandDims_39:0' shape=(None, 1) dtype=int64>, 'applesauce': <tf.Tensor 'ExpandDims_40:0' shape=(None, 1) dtype=int64>, 'apricot': <tf.Tensor 'ExpandDims_41:0' shape=(None, 1) dtype=int64>, 'arborio_rice': <tf.Tensor 'ExpandDims_42:0' shape=(None, 1) dtype=int64>, 'argan_oil': <tf.Tensor 'ExpandDims_43:0' shape=(None, 1) dtype=int64>, 'arrowroot': <tf.Tensor 'ExpandDims_44:0' shape=(None, 1) dtype=int64>, 'artichoke': <tf.Tensor 'ExpandDims_45:0' shape=(None, 1) dtype=int64>, 'artificial_sweetener': <tf.Tensor 'ExpandDims_46:0' shape=(None, 1) dtype=int64>, 'arugula': <tf.Tensor 'ExpandDims_47:0' shape=(None, 1) dtype=int64>, 'asadero': <tf.Tensor 'ExpandDims_48:0' shape=(None, 1) dtype=int64>, 'asafetida': <tf.Tensor 'ExpandDims_49:0' shape=(None, 1) dtype=int64>, 'asiago_cheese': <tf.Tensor 'ExpandDims_50:0' shape=(None, 1) dtype=int64>, 'asparagus': <tf.Tensor 'ExpandDims_51:0' shape=(None, 1) dtype=int64>, 'assorted_vegetables': <tf.Tensor 'ExpandDims_52:0' shape=(None, 1) dtype=int64>, 'avocado': <tf.Tensor 'ExpandDims_53:0' shape=(None, 1) dtype=int64>, 'azuki_bean': <tf.Tensor 'ExpandDims_54:0' shape=(None, 1) dtype=int64>, 'baby_carrot': <tf.Tensor 'ExpandDims_55:0' shape=(None, 1) dtype=int64>, 'baby_corn': <tf.Tensor 'ExpandDims_56:0' shape=(None, 1) dtype=int64>, 'baby_onions': <tf.Tensor 'ExpandDims_57:0' shape=(None, 1) dtype=int64>, 'baby_spinach': <tf.Tensor 'ExpandDims_58:0' shape=(None, 1) dtype=int64>, 'bacon': <tf.Tensor 'ExpandDims_59:0' shape=(None, 1) dtype=int64>, 'bacon_grease': <tf.Tensor 'ExpandDims_60:0' shape=(None, 1) dtype=int64>, 'bagel': <tf.Tensor 'ExpandDims_61:0' shape=(None, 1) dtype=int64>, 'baguette': <tf.Tensor 'ExpandDims_62:0' shape=(None, 1) dtype=int64>, 'baked_beans': <tf.Tensor 'ExpandDims_63:0' shape=(None, 1) dtype=int64>, 'baking_powder': <tf.Tensor 'ExpandDims_64:0' shape=(None, 1) dtype=int64>, 'baking_soda': <tf.Tensor 'ExpandDims_65:0' shape=(None, 1) dtype=int64>, 'balsamic_vinegar': <tf.Tensor 'ExpandDims_66:0' shape=(None, 1) dtype=int64>, 'bamboo_shoot': <tf.Tensor 'ExpandDims_67:0' shape=(None, 1) dtype=int64>, 'bamboo_skewers': <tf.Tensor 'ExpandDims_68:0' shape=(None, 1) dtype=int64>, 'banana': <tf.Tensor 'ExpandDims_69:0' shape=(None, 1) dtype=int64>, 'barbecue_sauce': <tf.Tensor 'ExpandDims_70:0' shape=(None, 1) dtype=int64>, 'barley': <tf.Tensor 'ExpandDims_71:0' shape=(None, 1) dtype=int64>, 'basil': <tf.Tensor 'ExpandDims_72:0' shape=(None, 1) dtype=int64>, 'basmati_rice': <tf.Tensor 'ExpandDims_73:0' shape=(None, 1) dtype=int64>, 'bass': <tf.Tensor 'ExpandDims_74:0' shape=(None, 1) dtype=int64>, 'bay_leaf': <tf.Tensor 'ExpandDims_75:0' shape=(None, 1) dtype=int64>, 'bean_sprouts': <tf.Tensor 'ExpandDims_76:0' shape=(None, 1) dtype=int64>, 'bean_thread': <tf.Tensor 'ExpandDims_77:0' shape=(None, 1) dtype=int64>, 'beans': <tf.Tensor 'ExpandDims_78:0' shape=(None, 1) dtype=int64>, 'bechamel': <tf.Tensor 'ExpandDims_79:0' shape=(None, 1) dtype=int64>, 'beef': <tf.Tensor 'ExpandDims_80:0' shape=(None, 1) dtype=int64>, 'beef_consomme': <tf.Tensor 'ExpandDims_81:0' shape=(None, 1) dtype=int64>, 'beef_gravy': <tf.Tensor 'ExpandDims_82:0' shape=(None, 1) dtype=int64>, 'beef_round_cuts': <tf.Tensor 'ExpandDims_83:0' shape=(None, 1) dtype=int64>, 'beef_stock': <tf.Tensor 'ExpandDims_84:0' shape=(None, 1) dtype=int64>, 'beer': <tf.Tensor 'ExpandDims_85:0' shape=(None, 1) dtype=int64>, 'beet': <tf.Tensor 'ExpandDims_86:0' shape=(None, 1) dtype=int64>, 'biscuit_mix': <tf.Tensor 'ExpandDims_87:0' shape=(None, 1) dtype=int64>, 'bitter_melon': <tf.Tensor 'ExpandDims_88:0' shape=(None, 1) dtype=int64>, 'black_bean': <tf.Tensor 'ExpandDims_89:0' shape=(None, 1) dtype=int64>, 'black_eyed_pea': <tf.Tensor 'ExpandDims_90:0' shape=(None, 1) dtype=int64>, 'black_forest_ham': <tf.Tensor 'ExpandDims_91:0' shape=(None, 1) dtype=int64>, 'black_mushroom': <tf.Tensor 'ExpandDims_92:0' shape=(None, 1) dtype=int64>, 'black_olive': <tf.Tensor 'ExpandDims_93:0' shape=(None, 1) dtype=int64>, 'black_truffle': <tf.Tensor 'ExpandDims_94:0' shape=(None, 1) dtype=int64>, 'blackberry': <tf.Tensor 'ExpandDims_95:0' shape=(None, 1) dtype=int64>, 'blue_cheese': <tf.Tensor 'ExpandDims_96:0' shape=(None, 1) dtype=int64>, 'bok_choy': <tf.Tensor 'ExpandDims_97:0' shape=(None, 1) dtype=int64>, 'bouillon': <tf.Tensor 'ExpandDims_98:0' shape=(None, 1) dtype=int64>, 'bouquet_garni': <tf.Tensor 'ExpandDims_99:0' shape=(None, 1) dtype=int64>, 'bourbon': <tf.Tensor 'ExpandDims_100:0' shape=(None, 1) dtype=int64>, 'brandy': <tf.Tensor 'ExpandDims_101:0' shape=(None, 1) dtype=int64>, 'bratwursts': <tf.Tensor 'ExpandDims_102:0' shape=(None, 1) dtype=int64>, 'bread': <tf.Tensor 'ExpandDims_103:0' shape=(None, 1) dtype=int64>, 'bread_bun': <tf.Tensor 'ExpandDims_104:0' shape=(None, 1) dtype=int64>, 'breadcrumb': <tf.Tensor 'ExpandDims_105:0' shape=(None, 1) dtype=int64>, 'brick_cheese': <tf.Tensor 'ExpandDims_106:0' shape=(None, 1) dtype=int64>, 'brie_cheese': <tf.Tensor 'ExpandDims_107:0' shape=(None, 1) dtype=int64>, 'broccoli': <tf.Tensor 'ExpandDims_108:0' shape=(None, 1) dtype=int64>, 'broccoli_raab': <tf.Tensor 'ExpandDims_109:0' shape=(None, 1) dtype=int64>, 'broth': <tf.Tensor 'ExpandDims_110:0' shape=(None, 1) dtype=int64>, 'brown_gravy_mix': <tf.Tensor 'ExpandDims_111:0' shape=(None, 1) dtype=int64>, 'brown_onion': <tf.Tensor 'ExpandDims_112:0' shape=(None, 1) dtype=int64>, 'brown_rice': <tf.Tensor 'ExpandDims_113:0' shape=(None, 1) dtype=int64>, 'brown_sugar': <tf.Tensor 'ExpandDims_114:0' shape=(None, 1) dtype=int64>, 'brussels_sprout': <tf.Tensor 'ExpandDims_115:0' shape=(None, 1) dtype=int64>, 'buckwheat': <tf.Tensor 'ExpandDims_116:0' shape=(None, 1) dtype=int64>, 'bulgur_wheat': <tf.Tensor 'ExpandDims_117:0' shape=(None, 1) dtype=int64>, 'burdock_root': <tf.Tensor 'ExpandDims_118:0' shape=(None, 1) dtype=int64>, 'butter': <tf.Tensor 'ExpandDims_119:0' shape=(None, 1) dtype=int64>, 'buttermilk': <tf.Tensor 'ExpandDims_120:0' shape=(None, 1) dtype=int64>, 'butternut_squash': <tf.Tensor 'ExpandDims_121:0' shape=(None, 1) dtype=int64>, 'cabbage': <tf.Tensor 'ExpandDims_122:0' shape=(None, 1) dtype=int64>, 'cajun_seasoning': <tf.Tensor 'ExpandDims_123:0' shape=(None, 1) dtype=int64>, 'calvados': <tf.Tensor 'ExpandDims_124:0' shape=(None, 1) dtype=int64>, 'campbell': <tf.Tensor 'ExpandDims_125:0' shape=(None, 1) dtype=int64>, 'canadian_bacon': <tf.Tensor 'ExpandDims_126:0' shape=(None, 1) dtype=int64>, 'cannellini_bean': <tf.Tensor 'ExpandDims_127:0' shape=(None, 1) dtype=int64>, 'canola_oil': <tf.Tensor 'ExpandDims_128:0' shape=(None, 1) dtype=int64>, 'cantaloupe': <tf.Tensor 'ExpandDims_129:0' shape=(None, 1) dtype=int64>, 'caper': <tf.Tensor 'ExpandDims_130:0' shape=(None, 1) dtype=int64>, 'capsicum': <tf.Tensor 'ExpandDims_131:0' shape=(None, 1) dtype=int64>, 'caraway_seed': <tf.Tensor 'ExpandDims_132:0' shape=(None, 1) dtype=int64>, 'cardamom': <tf.Tensor 'ExpandDims_133:0' shape=(None, 1) dtype=int64>, 'carrot': <tf.Tensor 'ExpandDims_134:0' shape=(None, 1) dtype=int64>, 'cashew_nut': <tf.Tensor 'ExpandDims_135:0' shape=(None, 1) dtype=int64>, 'catfish': <tf.Tensor 'ExpandDims_136:0' shape=(None, 1) dtype=int64>, 'caul': <tf.Tensor 'ExpandDims_137:0' shape=(None, 1) dtype=int64>, 'cauliflower': <tf.Tensor 'ExpandDims_138:0' shape=(None, 1) dtype=int64>, 'cayenne_pepper': <tf.Tensor 'ExpandDims_139:0' shape=(None, 1) dtype=int64>, 'celeriac': <tf.Tensor 'ExpandDims_140:0' shape=(None, 1) dtype=int64>, 'celery': <tf.Tensor 'ExpandDims_141:0' shape=(None, 1) dtype=int64>, 'celery_salt': <tf.Tensor 'ExpandDims_142:0' shape=(None, 1) dtype=int64>, 'celery_seed': <tf.Tensor 'ExpandDims_143:0' shape=(None, 1) dtype=int64>, 'celery_soup': <tf.Tensor 'ExpandDims_144:0' shape=(None, 1) dtype=int64>, 'chanterelle_mushroom': <tf.Tensor 'ExpandDims_145:0' shape=(None, 1) dtype=int64>, 'chard': <tf.Tensor 'ExpandDims_146:0' shape=(None, 1) dtype=int64>, 'chayote': <tf.Tensor 'ExpandDims_147:0' shape=(None, 1) dtype=int64>, 'cheddar_cheese': <tf.Tensor 'ExpandDims_148:0' shape=(None, 1) dtype=int64>, 'cheddar_cheese_soup': <tf.Tensor 'ExpandDims_149:0' shape=(None, 1) dtype=int64>, 'cheese': <tf.Tensor 'ExpandDims_150:0' shape=(None, 1) dtype=int64>, 'cherry': <tf.Tensor 'ExpandDims_151:0' shape=(None, 1) dtype=int64>, 'chervil': <tf.Tensor 'ExpandDims_152:0' shape=(None, 1) dtype=int64>, 'cheshire_cheese': <tf.Tensor 'ExpandDims_153:0' shape=(None, 1) dtype=int64>, 'chestnut': <tf.Tensor 'ExpandDims_154:0' shape=(None, 1) dtype=int64>, 'chestnut_mushroom': <tf.Tensor 'ExpandDims_155:0' shape=(None, 1) dtype=int64>, 'chicken': <tf.Tensor 'ExpandDims_156:0' shape=(None, 1) dtype=int64>, 'chicken_base': <tf.Tensor 'ExpandDims_157:0' shape=(None, 1) dtype=int64>, 'chicken_liver': <tf.Tensor 'ExpandDims_158:0' shape=(None, 1) dtype=int64>, 'chicken_sausage': <tf.Tensor 'ExpandDims_159:0' shape=(None, 1) dtype=int64>, 'chicken_soup': <tf.Tensor 'ExpandDims_160:0' shape=(None, 1) dtype=int64>, 'chicken_stock': <tf.Tensor 'ExpandDims_161:0' shape=(None, 1) dtype=int64>, 'chickpea': <tf.Tensor 'ExpandDims_162:0' shape=(None, 1) dtype=int64>, 'chile_pepper': <tf.Tensor 'ExpandDims_163:0' shape=(None, 1) dtype=int64>, 'chili_oil': <tf.Tensor 'ExpandDims_164:0' shape=(None, 1) dtype=int64>, 'chili_paste': <tf.Tensor 'ExpandDims_165:0' shape=(None, 1) dtype=int64>, 'chili_powder': <tf.Tensor 'ExpandDims_166:0' shape=(None, 1) dtype=int64>, 'chili_sauce': <tf.Tensor 'ExpandDims_167:0' shape=(None, 1) dtype=int64>, 'chinese_cabbage': <tf.Tensor 'ExpandDims_168:0' shape=(None, 1) dtype=int64>, 'chinese_chives': <tf.Tensor 'ExpandDims_169:0' shape=(None, 1) dtype=int64>, 'Chinese_mushroom': <tf.Tensor 'ExpandDims:0' shape=(None, 1) dtype=int64>, 'Chinese_noodles': <tf.Tensor 'ExpandDims_1:0' shape=(None, 1) dtype=int64>, 'chipotle_pepper': <tf.Tensor 'ExpandDims_170:0' shape=(None, 1) dtype=int64>, 'chives': <tf.Tensor 'ExpandDims_171:0' shape=(None, 1) dtype=int64>, 'chocolate': <tf.Tensor 'ExpandDims_172:0' shape=(None, 1) dtype=int64>, 'chutney': <tf.Tensor 'ExpandDims_173:0' shape=(None, 1) dtype=int64>, 'cider': <tf.Tensor 'ExpandDims_174:0' shape=(None, 1) dtype=int64>, 'cider_vinegar': <tf.Tensor 'ExpandDims_175:0' shape=(None, 1) dtype=int64>, 'cilantro': <tf.Tensor 'ExpandDims_176:0' shape=(None, 1) dtype=int64>, 'cinnamon': <tf.Tensor 'ExpandDims_177:0' shape=(None, 1) dtype=int64>, 'clam': <tf.Tensor 'ExpandDims_178:0' shape=(None, 1) dtype=int64>, 'clam_juice': <tf.Tensor 'ExpandDims_179:0' shape=(None, 1) dtype=int64>, 'clarified_butter': <tf.Tensor 'ExpandDims_180:0' shape=(None, 1) dtype=int64>, 'clove': <tf.Tensor 'ExpandDims_181:0' shape=(None, 1) dtype=int64>, 'cocoa_powder': <tf.Tensor 'ExpandDims_182:0' shape=(None, 1) dtype=int64>, 'coconut': <tf.Tensor 'ExpandDims_183:0' shape=(None, 1) dtype=int64>, 'coconut_milk_or_cream': <tf.Tensor 'ExpandDims_184:0' shape=(None, 1) dtype=int64>, 'coconut_oil': <tf.Tensor 'ExpandDims_185:0' shape=(None, 1) dtype=int64>, 'cod': <tf.Tensor 'ExpandDims_186:0' shape=(None, 1) dtype=int64>, 'colby': <tf.Tensor 'ExpandDims_187:0' shape=(None, 1) dtype=int64>, 'coleslaw_mix': <tf.Tensor 'ExpandDims_188:0' shape=(None, 1) dtype=int64>, 'confit': <tf.Tensor 'ExpandDims_189:0' shape=(None, 1) dtype=int64>, 'coriander': <tf.Tensor 'ExpandDims_190:0' shape=(None, 1) dtype=int64>, 'coriander_root': <tf.Tensor 'ExpandDims_191:0' shape=(None, 1) dtype=int64>, 'corn': <tf.Tensor 'ExpandDims_192:0' shape=(None, 1) dtype=int64>, 'corn_chips': <tf.Tensor 'ExpandDims_193:0' shape=(None, 1) dtype=int64>, 'corn_flour': <tf.Tensor 'ExpandDims_194:0' shape=(None, 1) dtype=int64>, 'corn_oil': <tf.Tensor 'ExpandDims_195:0' shape=(None, 1) dtype=int64>, 'cornbread': <tf.Tensor 'ExpandDims_196:0' shape=(None, 1) dtype=int64>, 'corned_beef': <tf.Tensor 'ExpandDims_197:0' shape=(None, 1) dtype=int64>, 'cornish_hen': <tf.Tensor 'ExpandDims_198:0' shape=(None, 1) dtype=int64>, 'cornmeal': <tf.Tensor 'ExpandDims_199:0' shape=(None, 1) dtype=int64>, 'cornstarch': <tf.Tensor 'ExpandDims_200:0' shape=(None, 1) dtype=int64>, 'cottage_cheese': <tf.Tensor 'ExpandDims_201:0' shape=(None, 1) dtype=int64>, 'couscous': <tf.Tensor 'ExpandDims_202:0' shape=(None, 1) dtype=int64>, 'crab': <tf.Tensor 'ExpandDims_203:0' shape=(None, 1) dtype=int64>, 'crab_claw': <tf.Tensor 'ExpandDims_204:0' shape=(None, 1) dtype=int64>, 'cracker': <tf.Tensor 'ExpandDims_205:0' shape=(None, 1) dtype=int64>, 'craisin': <tf.Tensor 'ExpandDims_206:0' shape=(None, 1) dtype=int64>, 'cranberry': <tf.Tensor 'ExpandDims_207:0' shape=(None, 1) dtype=int64>, 'cranberry_bean': <tf.Tensor 'ExpandDims_208:0' shape=(None, 1) dtype=int64>, 'crayfish': <tf.Tensor 'ExpandDims_209:0' shape=(None, 1) dtype=int64>, 'cream': <tf.Tensor 'ExpandDims_210:0' shape=(None, 1) dtype=int64>, 'cream_cheese': <tf.Tensor 'ExpandDims_211:0' shape=(None, 1) dtype=int64>, 'creamed_corn': <tf.Tensor 'ExpandDims_212:0' shape=(None, 1) dtype=int64>, 'creme_fraiche': <tf.Tensor 'ExpandDims_213:0' shape=(None, 1) dtype=int64>, 'cremini_mushroom': <tf.Tensor 'ExpandDims_214:0' shape=(None, 1) dtype=int64>, 'crouton': <tf.Tensor 'ExpandDims_215:0' shape=(None, 1) dtype=int64>, 'crusty_bread': <tf.Tensor 'ExpandDims_216:0' shape=(None, 1) dtype=int64>, 'crystallized_ginger': <tf.Tensor 'ExpandDims_217:0' shape=(None, 1) dtype=int64>, 'cucumber': <tf.Tensor 'ExpandDims_218:0' shape=(None, 1) dtype=int64>, 'cumin': <tf.Tensor 'ExpandDims_219:0' shape=(None, 1) dtype=int64>, 'currant': <tf.Tensor 'ExpandDims_220:0' shape=(None, 1) dtype=int64>, 'curry_leaf': <tf.Tensor 'ExpandDims_221:0' shape=(None, 1) dtype=int64>, 'curry_paste': <tf.Tensor 'ExpandDims_222:0' shape=(None, 1) dtype=int64>, 'curry_powder': <tf.Tensor 'ExpandDims_223:0' shape=(None, 1) dtype=int64>, 'curry_sauce': <tf.Tensor 'ExpandDims_224:0' shape=(None, 1) dtype=int64>, 'dashi': <tf.Tensor 'ExpandDims_225:0' shape=(None, 1) dtype=int64>, 'dashi_stock': <tf.Tensor 'ExpandDims_226:0' shape=(None, 1) dtype=int64>, 'date': <tf.Tensor 'ExpandDims_227:0' shape=(None, 1) dtype=int64>, 'dehydrated_onion': <tf.Tensor 'ExpandDims_228:0' shape=(None, 1) dtype=int64>, 'demi_glace': <tf.Tensor 'ExpandDims_229:0' shape=(None, 1) dtype=int64>, 'dill': <tf.Tensor 'ExpandDims_230:0' shape=(None, 1) dtype=int64>, 'dried_apricot': <tf.Tensor 'ExpandDims_231:0' shape=(None, 1) dtype=int64>, 'dried_chile_pepper': <tf.Tensor 'ExpandDims_232:0' shape=(None, 1) dtype=int64>, 'dripping': <tf.Tensor 'ExpandDims_233:0' shape=(None, 1) dtype=int64>, 'duck': <tf.Tensor 'ExpandDims_234:0' shape=(None, 1) dtype=int64>, 'edamame': <tf.Tensor 'ExpandDims_235:0' shape=(None, 1) dtype=int64>, 'eel': <tf.Tensor 'ExpandDims_236:0' shape=(None, 1) dtype=int64>, 'egg': <tf.Tensor 'ExpandDims_237:0' shape=(None, 1) dtype=int64>, 'egg_noodles': <tf.Tensor 'ExpandDims_238:0' shape=(None, 1) dtype=int64>, 'egg_roll_wraps': <tf.Tensor 'ExpandDims_239:0' shape=(None, 1) dtype=int64>, 'egg_substitute': <tf.Tensor 'ExpandDims_240:0' shape=(None, 1) dtype=int64>, 'eggplant': <tf.Tensor 'ExpandDims_241:0' shape=(None, 1) dtype=int64>, 'emeril': <tf.Tensor 'ExpandDims_242:0' shape=(None, 1) dtype=int64>, 'emmenthaler_cheese': <tf.Tensor 'ExpandDims_243:0' shape=(None, 1) dtype=int64>, 'enchilada_sauce': <tf.Tensor 'ExpandDims_244:0' shape=(None, 1) dtype=int64>, 'endive': <tf.Tensor 'ExpandDims_245:0' shape=(None, 1) dtype=int64>, 'english_muffin': <tf.Tensor 'ExpandDims_246:0' shape=(None, 1) dtype=int64>, 'english_mustard': <tf.Tensor 'ExpandDims_247:0' shape=(None, 1) dtype=int64>, 'enoki_mushroom': <tf.Tensor 'ExpandDims_248:0' shape=(None, 1) dtype=int64>, 'epazote': <tf.Tensor 'ExpandDims_249:0' shape=(None, 1) dtype=int64>, 'evaporated_milk': <tf.Tensor 'ExpandDims_250:0' shape=(None, 1) dtype=int64>, 'fajita_seasoning_mix': <tf.Tensor 'ExpandDims_251:0' shape=(None, 1) dtype=int64>, 'fat-free_refried_beans': <tf.Tensor 'ExpandDims_252:0' shape=(None, 1) dtype=int64>, 'fava_bean': <tf.Tensor 'ExpandDims_253:0' shape=(None, 1) dtype=int64>, 'fennel': <tf.Tensor 'ExpandDims_254:0' shape=(None, 1) dtype=int64>, 'fennel_seeds': <tf.Tensor 'ExpandDims_255:0' shape=(None, 1) dtype=int64>, 'fenugreek': <tf.Tensor 'ExpandDims_256:0' shape=(None, 1) dtype=int64>, 'fermented_black_bean': <tf.Tensor 'ExpandDims_257:0' shape=(None, 1) dtype=int64>, 'feta_cheese': <tf.Tensor 'ExpandDims_258:0' shape=(None, 1) dtype=int64>, 'fig': <tf.Tensor 'ExpandDims_259:0' shape=(None, 1) dtype=int64>, 'filet_mignon': <tf.Tensor 'ExpandDims_260:0' shape=(None, 1) dtype=int64>, 'fire_oil': <tf.Tensor 'ExpandDims_261:0' shape=(None, 1) dtype=int64>, 'fish': <tf.Tensor 'ExpandDims_262:0' shape=(None, 1) dtype=int64>, 'fish_sauce': <tf.Tensor 'ExpandDims_263:0' shape=(None, 1) dtype=int64>, 'fish_stock': <tf.Tensor 'ExpandDims_264:0' shape=(None, 1) dtype=int64>, 'five_spice_powder': <tf.Tensor 'ExpandDims_265:0' shape=(None, 1) dtype=int64>, 'flat_bread': <tf.Tensor 'ExpandDims_266:0' shape=(None, 1) dtype=int64>, 'flax_seed': <tf.Tensor 'ExpandDims_267:0' shape=(None, 1) dtype=int64>, 'flounder': <tf.Tensor 'ExpandDims_268:0' shape=(None, 1) dtype=int64>, 'flour': <tf.Tensor 'ExpandDims_269:0' shape=(None, 1) dtype=int64>, 'fontina_cheese': <tf.Tensor 'ExpandDims_270:0' shape=(None, 1) dtype=int64>, 'French_beans': <tf.Tensor 'ExpandDims_2:0' shape=(None, 1) dtype=int64>, 'French_bread': <tf.Tensor 'ExpandDims_3:0' shape=(None, 1) dtype=int64>, 'French_dressing': <tf.Tensor 'ExpandDims_4:0' shape=(None, 1) dtype=int64>, 'french_style_green_bean': <tf.Tensor 'ExpandDims_271:0' shape=(None, 1) dtype=int64>, 'fresh_button_mushroom': <tf.Tensor 'ExpandDims_272:0' shape=(None, 1) dtype=int64>, 'frozen_edamame': <tf.Tensor 'ExpandDims_273:0' shape=(None, 1) dtype=int64>, 'fusilli': <tf.Tensor 'ExpandDims_274:0' shape=(None, 1) dtype=int64>, 'galangal': <tf.Tensor 'ExpandDims_275:0' shape=(None, 1) dtype=int64>, 'garlic': <tf.Tensor 'ExpandDims_276:0' shape=(None, 1) dtype=int64>, 'garlic_butter': <tf.Tensor 'ExpandDims_277:0' shape=(None, 1) dtype=int64>, 'garlic_pepper_seasoning': <tf.Tensor 'ExpandDims_278:0' shape=(None, 1) dtype=int64>, 'garlic_powder': <tf.Tensor 'ExpandDims_279:0' shape=(None, 1) dtype=int64>, 'garlic_salt': <tf.Tensor 'ExpandDims_280:0' shape=(None, 1) dtype=int64>, 'gelatin': <tf.Tensor 'ExpandDims_281:0' shape=(None, 1) dtype=int64>, 'genoa_salami': <tf.Tensor 'ExpandDims_282:0' shape=(None, 1) dtype=int64>, 'ghee': <tf.Tensor 'ExpandDims_283:0' shape=(None, 1) dtype=int64>, 'ginger': <tf.Tensor 'ExpandDims_284:0' shape=(None, 1) dtype=int64>, 'ginger-garlic_paste': <tf.Tensor 'ExpandDims_285:0' shape=(None, 1) dtype=int64>, 'gingersnap_cookie': <tf.Tensor 'ExpandDims_286:0' shape=(None, 1) dtype=int64>, 'gingersnaps': <tf.Tensor 'ExpandDims_287:0' shape=(None, 1) dtype=int64>, 'gloucester': <tf.Tensor 'ExpandDims_288:0' shape=(None, 1) dtype=int64>, 'gluten': <tf.Tensor 'ExpandDims_289:0' shape=(None, 1) dtype=int64>, 'goat_cheese': <tf.Tensor 'ExpandDims_290:0' shape=(None, 1) dtype=int64>, 'golden_syrup': <tf.Tensor 'ExpandDims_291:0' shape=(None, 1) dtype=int64>, 'goose': <tf.Tensor 'ExpandDims_292:0' shape=(None, 1) dtype=int64>, 'gorgonzola': <tf.Tensor 'ExpandDims_293:0' shape=(None, 1) dtype=int64>, 'gouda': <tf.Tensor 'ExpandDims_294:0' shape=(None, 1) dtype=int64>, 'grand_marnier': <tf.Tensor 'ExpandDims_295:0' shape=(None, 1) dtype=int64>, 'granulated_garlic': <tf.Tensor 'ExpandDims_296:0' shape=(None, 1) dtype=int64>, 'grape': <tf.Tensor 'ExpandDims_297:0' shape=(None, 1) dtype=int64>, 'grapefruit': <tf.Tensor 'ExpandDims_298:0' shape=(None, 1) dtype=int64>, 'grapeseed_oil': <tf.Tensor 'ExpandDims_299:0' shape=(None, 1) dtype=int64>, 'grating_cheese': <tf.Tensor 'ExpandDims_300:0' shape=(None, 1) dtype=int64>, 'gravy': <tf.Tensor 'ExpandDims_301:0' shape=(None, 1) dtype=int64>, 'great_northern_bean': <tf.Tensor 'ExpandDims_302:0' shape=(None, 1) dtype=int64>, 'Greek_salad_dressing': <tf.Tensor 'ExpandDims_5:0' shape=(None, 1) dtype=int64>, 'green_bean': <tf.Tensor 'ExpandDims_303:0' shape=(None, 1) dtype=int64>, 'green_olive': <tf.Tensor 'ExpandDims_304:0' shape=(None, 1) dtype=int64>, 'green_onion': <tf.Tensor 'ExpandDims_305:0' shape=(None, 1) dtype=int64>, 'greens': <tf.Tensor 'ExpandDims_306:0' shape=(None, 1) dtype=int64>, 'ground_meat': <tf.Tensor 'ExpandDims_307:0' shape=(None, 1) dtype=int64>, 'ground_round': <tf.Tensor 'ExpandDims_308:0' shape=(None, 1) dtype=int64>, 'gruyere': <tf.Tensor 'ExpandDims_309:0' shape=(None, 1) dtype=int64>, 'guacamole': <tf.Tensor 'ExpandDims_310:0' shape=(None, 1) dtype=int64>, 'Guinness_stout': <tf.Tensor 'ExpandDims_6:0' shape=(None, 1) dtype=int64>, 'gumbo': <tf.Tensor 'ExpandDims_311:0' shape=(None, 1) dtype=int64>, 'habanero_chile': <tf.Tensor 'ExpandDims_312:0' shape=(None, 1) dtype=int64>, 'haddock': <tf.Tensor 'ExpandDims_313:0' shape=(None, 1) dtype=int64>, 'hake_fillets': <tf.Tensor 'ExpandDims_314:0' shape=(None, 1) dtype=int64>, 'half_and_half': <tf.Tensor 'ExpandDims_315:0' shape=(None, 1) dtype=int64>, 'halibut': <tf.Tensor 'ExpandDims_316:0' shape=(None, 1) dtype=int64>, 'ham': <tf.Tensor 'ExpandDims_317:0' shape=(None, 1) dtype=int64>, 'ham_hock': <tf.Tensor 'ExpandDims_318:0' shape=(None, 1) dtype=int64>, 'hamburger': <tf.Tensor 'ExpandDims_319:0' shape=(None, 1) dtype=int64>, 'harissa': <tf.Tensor 'ExpandDims_320:0' shape=(None, 1) dtype=int64>, 'hash_browns': <tf.Tensor 'ExpandDims_321:0' shape=(None, 1) dtype=int64>, 'havarti': <tf.Tensor 'ExpandDims_322:0' shape=(None, 1) dtype=int64>, 'hazelnut': <tf.Tensor 'ExpandDims_323:0' shape=(None, 1) dtype=int64>, 'heavy_cream': <tf.Tensor 'ExpandDims_324:0' shape=(None, 1) dtype=int64>, 'herb': <tf.Tensor 'ExpandDims_325:0' shape=(None, 1) dtype=int64>, 'herbes_de_provence': <tf.Tensor 'ExpandDims_326:0' shape=(None, 1) dtype=int64>, 'hoagie_rolls': <tf.Tensor 'ExpandDims_327:0' shape=(None, 1) dtype=int64>, 'hoisin_sauce': <tf.Tensor 'ExpandDims_328:0' shape=(None, 1) dtype=int64>, 'hokkien_noodles': <tf.Tensor 'ExpandDims_329:0' shape=(None, 1) dtype=int64>, 'hominy': <tf.Tensor 'ExpandDims_330:0' shape=(None, 1) dtype=int64>, 'honey': <tf.Tensor 'ExpandDims_331:0' shape=(None, 1) dtype=int64>, 'honey_mustard': <tf.Tensor 'ExpandDims_332:0' shape=(None, 1) dtype=int64>, 'horseradish': <tf.Tensor 'ExpandDims_333:0' shape=(None, 1) dtype=int64>, 'horseradish_sauce': <tf.Tensor 'ExpandDims_334:0' shape=(None, 1) dtype=int64>, 'hot_bean_paste': <tf.Tensor 'ExpandDims_335:0' shape=(None, 1) dtype=int64>, 'hot_pepper_sauce': <tf.Tensor 'ExpandDims_336:0' shape=(None, 1) dtype=int64>, 'hummus': <tf.Tensor 'ExpandDims_337:0' shape=(None, 1) dtype=int64>, 'ice_water': <tf.Tensor 'ExpandDims_338:0' shape=(None, 1) dtype=int64>, 'Italian_bread': <tf.Tensor 'ExpandDims_7:0' shape=(None, 1) dtype=int64>, 'Italian_breadcrumbs': <tf.Tensor 'ExpandDims_8:0' shape=(None, 1) dtype=int64>, 'Italian_rolls': <tf.Tensor 'ExpandDims_9:0' shape=(None, 1) dtype=int64>, 'Italian_salad_dressing': <tf.Tensor 'ExpandDims_10:0' shape=(None, 1) dtype=int64>, 'Italian_spices': <tf.Tensor 'ExpandDims_11:0' shape=(None, 1) dtype=int64>, 'jack_cheese': <tf.Tensor 'ExpandDims_339:0' shape=(None, 1) dtype=int64>, 'jaggery': <tf.Tensor 'ExpandDims_340:0' shape=(None, 1) dtype=int64>, 'jalapeno_pepper': <tf.Tensor 'ExpandDims_341:0' shape=(None, 1) dtype=int64>, 'Japanese_breadcrumbs': <tf.Tensor 'ExpandDims_12:0' shape=(None, 1) dtype=int64>, 'jasmine_rice': <tf.Tensor 'ExpandDims_342:0' shape=(None, 1) dtype=int64>, 'jicama': <tf.Tensor 'ExpandDims_343:0' shape=(None, 1) dtype=int64>, 'juniper_berries': <tf.Tensor 'ExpandDims_344:0' shape=(None, 1) dtype=int64>, 'kaffir_lime': <tf.Tensor 'ExpandDims_345:0' shape=(None, 1) dtype=int64>, 'kalamata_olive': <tf.Tensor 'ExpandDims_346:0' shape=(None, 1) dtype=int64>, 'kale': <tf.Tensor 'ExpandDims_347:0' shape=(None, 1) dtype=int64>, 'kamaboko': <tf.Tensor 'ExpandDims_348:0' shape=(None, 1) dtype=int64>, 'ketchup': <tf.Tensor 'ExpandDims_349:0' shape=(None, 1) dtype=int64>, 'ketjap_manis': <tf.Tensor 'ExpandDims_350:0' shape=(None, 1) dtype=int64>, 'kidney_bean': <tf.Tensor 'ExpandDims_351:0' shape=(None, 1) dtype=int64>, 'kielbasa': <tf.Tensor 'ExpandDims_352:0' shape=(None, 1) dtype=int64>, 'kohlrabi': <tf.Tensor 'ExpandDims_353:0' shape=(None, 1) dtype=int64>, 'kumquat': <tf.Tensor 'ExpandDims_354:0' shape=(None, 1) dtype=int64>, 'lamb': <tf.Tensor 'ExpandDims_355:0' shape=(None, 1) dtype=int64>, 'lambs_kidney': <tf.Tensor 'ExpandDims_356:0' shape=(None, 1) dtype=int64>, 'lancashire_cheese': <tf.Tensor 'ExpandDims_357:0' shape=(None, 1) dtype=int64>, 'lard': <tf.Tensor 'ExpandDims_358:0' shape=(None, 1) dtype=int64>, 'lardon': <tf.Tensor 'ExpandDims_359:0' shape=(None, 1) dtype=int64>, 'lasagna_noodles': <tf.Tensor 'ExpandDims_360:0' shape=(None, 1) dtype=int64>, 'leek': <tf.Tensor 'ExpandDims_361:0' shape=(None, 1) dtype=int64>, 'lemon': <tf.Tensor 'ExpandDims_362:0' shape=(None, 1) dtype=int64>, 'lemon_grass': <tf.Tensor 'ExpandDims_363:0' shape=(None, 1) dtype=int64>, 'lemon_juice': <tf.Tensor 'ExpandDims_364:0' shape=(None, 1) dtype=int64>, 'lentil': <tf.Tensor 'ExpandDims_365:0' shape=(None, 1) dtype=int64>, 'lettuce': <tf.Tensor 'ExpandDims_366:0' shape=(None, 1) dtype=int64>, 'lima_bean': <tf.Tensor 'ExpandDims_367:0' shape=(None, 1) dtype=int64>, 'lime': <tf.Tensor 'ExpandDims_368:0' shape=(None, 1) dtype=int64>, 'liver': <tf.Tensor 'ExpandDims_369:0' shape=(None, 1) dtype=int64>, 'lobster': <tf.Tensor 'ExpandDims_370:0' shape=(None, 1) dtype=int64>, 'long_grain_and_wild_rice_blend': <tf.Tensor 'ExpandDims_371:0' shape=(None, 1) dtype=int64>, 'lotus': <tf.Tensor 'ExpandDims_372:0' shape=(None, 1) dtype=int64>, 'macaroni': <tf.Tensor 'ExpandDims_373:0' shape=(None, 1) dtype=int64>, 'mace': <tf.Tensor 'ExpandDims_374:0' shape=(None, 1) dtype=int64>, 'mackerel': <tf.Tensor 'ExpandDims_375:0' shape=(None, 1) dtype=int64>, 'madeira': <tf.Tensor 'ExpandDims_376:0' shape=(None, 1) dtype=int64>, 'mahi_mahi': <tf.Tensor 'ExpandDims_377:0' shape=(None, 1) dtype=int64>, 'malt_vinegar': <tf.Tensor 'ExpandDims_378:0' shape=(None, 1) dtype=int64>, 'mandarin_orange': <tf.Tensor 'ExpandDims_379:0' shape=(None, 1) dtype=int64>, 'mange_tout': <tf.Tensor 'ExpandDims_380:0' shape=(None, 1) dtype=int64>, 'mango': <tf.Tensor 'ExpandDims_381:0' shape=(None, 1) dtype=int64>, 'manicotti': <tf.Tensor 'ExpandDims_382:0' shape=(None, 1) dtype=int64>, 'manzanilla_olive': <tf.Tensor 'ExpandDims_383:0' shape=(None, 1) dtype=int64>, 'maple_syrup': <tf.Tensor 'ExpandDims_384:0' shape=(None, 1) dtype=int64>, 'margarine': <tf.Tensor 'ExpandDims_385:0' shape=(None, 1) dtype=int64>, 'marinara_sauce': <tf.Tensor 'ExpandDims_386:0' shape=(None, 1) dtype=int64>, 'marjoram': <tf.Tensor 'ExpandDims_387:0' shape=(None, 1) dtype=int64>, 'marmalade': <tf.Tensor 'ExpandDims_388:0' shape=(None, 1) dtype=int64>, 'marmite': <tf.Tensor 'ExpandDims_389:0' shape=(None, 1) dtype=int64>, 'marsala': <tf.Tensor 'ExpandDims_390:0' shape=(None, 1) dtype=int64>, 'masa_harina': <tf.Tensor 'ExpandDims_391:0' shape=(None, 1) dtype=int64>, 'masala': <tf.Tensor 'ExpandDims_392:0' shape=(None, 1) dtype=int64>, 'mascarpone': <tf.Tensor 'ExpandDims_393:0' shape=(None, 1) dtype=int64>, 'masoor_dal': <tf.Tensor 'ExpandDims_394:0' shape=(None, 1) dtype=int64>, 'mayonnaise': <tf.Tensor 'ExpandDims_395:0' shape=(None, 1) dtype=int64>, 'meat_stock': <tf.Tensor 'ExpandDims_396:0' shape=(None, 1) dtype=int64>, 'meatloaf_mix': <tf.Tensor 'ExpandDims_397:0' shape=(None, 1) dtype=int64>, 'Mexican_blend_cheese': <tf.Tensor 'ExpandDims_13:0' shape=(None, 1) dtype=int64>, 'mexicorn': <tf.Tensor 'ExpandDims_398:0' shape=(None, 1) dtype=int64>, 'milk': <tf.Tensor 'ExpandDims_399:0' shape=(None, 1) dtype=int64>, 'mint': <tf.Tensor 'ExpandDims_400:0' shape=(None, 1) dtype=int64>, 'miracle_whip_dressing': <tf.Tensor 'ExpandDims_401:0' shape=(None, 1) dtype=int64>, 'miso': <tf.Tensor 'ExpandDims_402:0' shape=(None, 1) dtype=int64>, 'mixed_spice': <tf.Tensor 'ExpandDims_403:0' shape=(None, 1) dtype=int64>, 'mizithra_cheese': <tf.Tensor 'ExpandDims_404:0' shape=(None, 1) dtype=int64>, 'molasses': <tf.Tensor 'ExpandDims_405:0' shape=(None, 1) dtype=int64>, 'monosodium_glutamate': <tf.Tensor 'ExpandDims_406:0' shape=(None, 1) dtype=int64>, 'moroccan_seasoning': <tf.Tensor 'ExpandDims_407:0' shape=(None, 1) dtype=int64>, 'mortadella': <tf.Tensor 'ExpandDims_408:0' shape=(None, 1) dtype=int64>, 'mozzarella': <tf.Tensor 'ExpandDims_409:0' shape=(None, 1) dtype=int64>, 'mung_beans': <tf.Tensor 'ExpandDims_410:0' shape=(None, 1) dtype=int64>, 'muscovado_sugar': <tf.Tensor 'ExpandDims_411:0' shape=(None, 1) dtype=int64>, 'mushroom': <tf.Tensor 'ExpandDims_412:0' shape=(None, 1) dtype=int64>, 'mushroom_soup': <tf.Tensor 'ExpandDims_413:0' shape=(None, 1) dtype=int64>, 'mussel': <tf.Tensor 'ExpandDims_414:0' shape=(None, 1) dtype=int64>, 'mustard_oil': <tf.Tensor 'ExpandDims_415:0' shape=(None, 1) dtype=int64>, 'mustard_prepared': <tf.Tensor 'ExpandDims_416:0' shape=(None, 1) dtype=int64>, 'mustard_seed_and_powder': <tf.Tensor 'ExpandDims_417:0' shape=(None, 1) dtype=int64>, 'navy_bean': <tf.Tensor 'ExpandDims_418:0' shape=(None, 1) dtype=int64>, 'nectarine': <tf.Tensor 'ExpandDims_419:0' shape=(None, 1) dtype=int64>, 'nigella_seeds': <tf.Tensor 'ExpandDims_420:0' shape=(None, 1) dtype=int64>, 'nori': <tf.Tensor 'ExpandDims_421:0' shape=(None, 1) dtype=int64>, 'nutmeg': <tf.Tensor 'ExpandDims_422:0' shape=(None, 1) dtype=int64>, 'oatmeal': <tf.Tensor 'ExpandDims_423:0' shape=(None, 1) dtype=int64>, 'okra': <tf.Tensor 'ExpandDims_424:0' shape=(None, 1) dtype=int64>, 'Old_Bay_Seasoning': <tf.Tensor 'ExpandDims_14:0' shape=(None, 1) dtype=int64>, 'olive': <tf.Tensor 'ExpandDims_425:0' shape=(None, 1) dtype=int64>, 'olive_oil': <tf.Tensor 'ExpandDims_426:0' shape=(None, 1) dtype=int64>, 'onion': <tf.Tensor 'ExpandDims_427:0' shape=(None, 1) dtype=int64>, 'onion_powder': <tf.Tensor 'ExpandDims_428:0' shape=(None, 1) dtype=int64>, 'onion_soup': <tf.Tensor 'ExpandDims_429:0' shape=(None, 1) dtype=int64>, 'orange': <tf.Tensor 'ExpandDims_430:0' shape=(None, 1) dtype=int64>, 'orange_juice': <tf.Tensor 'ExpandDims_431:0' shape=(None, 1) dtype=int64>, 'orange_rind': <tf.Tensor 'ExpandDims_432:0' shape=(None, 1) dtype=int64>, 'oregano': <tf.Tensor 'ExpandDims_433:0' shape=(None, 1) dtype=int64>, 'oxtails': <tf.Tensor 'ExpandDims_434:0' shape=(None, 1) dtype=int64>, 'oyster_sauce': <tf.Tensor 'ExpandDims_435:0' shape=(None, 1) dtype=int64>, 'oysters': <tf.Tensor 'ExpandDims_436:0' shape=(None, 1) dtype=int64>, 'Pace_Picante_Sauce': <tf.Tensor 'ExpandDims_15:0' shape=(None, 1) dtype=int64>, 'palm_sugar': <tf.Tensor 'ExpandDims_437:0' shape=(None, 1) dtype=int64>, 'pancetta': <tf.Tensor 'ExpandDims_438:0' shape=(None, 1) dtype=int64>, 'panir': <tf.Tensor 'ExpandDims_439:0' shape=(None, 1) dtype=int64>, 'panko': <tf.Tensor 'ExpandDims_440:0' shape=(None, 1) dtype=int64>, 'papaya': <tf.Tensor 'ExpandDims_441:0' shape=(None, 1) dtype=int64>, 'paprika': <tf.Tensor 'ExpandDims_442:0' shape=(None, 1) dtype=int64>, 'parchment_paper': <tf.Tensor 'ExpandDims_443:0' shape=(None, 1) dtype=int64>, 'parmesan_cheese': <tf.Tensor 'ExpandDims_444:0' shape=(None, 1) dtype=int64>, 'parsley': <tf.Tensor 'ExpandDims_445:0' shape=(None, 1) dtype=int64>, 'parsnip': <tf.Tensor 'ExpandDims_446:0' shape=(None, 1) dtype=int64>, 'passata': <tf.Tensor 'ExpandDims_447:0' shape=(None, 1) dtype=int64>, 'pasta': <tf.Tensor 'ExpandDims_448:0' shape=(None, 1) dtype=int64>, 'pasta_sauce': <tf.Tensor 'ExpandDims_449:0' shape=(None, 1) dtype=int64>, 'pastis': <tf.Tensor 'ExpandDims_450:0' shape=(None, 1) dtype=int64>, 'pastry_dough': <tf.Tensor 'ExpandDims_451:0' shape=(None, 1) dtype=int64>, 'pea': <tf.Tensor 'ExpandDims_452:0' shape=(None, 1) dtype=int64>, 'peanut': <tf.Tensor 'ExpandDims_453:0' shape=(None, 1) dtype=int64>, 'peanut_butter': <tf.Tensor 'ExpandDims_454:0' shape=(None, 1) dtype=int64>, 'peanut_oil': <tf.Tensor 'ExpandDims_455:0' shape=(None, 1) dtype=int64>, 'peanut_sauce': <tf.Tensor 'ExpandDims_456:0' shape=(None, 1) dtype=int64>, 'pear': <tf.Tensor 'ExpandDims_457:0' shape=(None, 1) dtype=int64>, 'pearl_onion': <tf.Tensor 'ExpandDims_458:0' shape=(None, 1) dtype=int64>, 'pecan': <tf.Tensor 'ExpandDims_459:0' shape=(None, 1) dtype=int64>, 'pecorino_cheese': <tf.Tensor 'ExpandDims_460:0' shape=(None, 1) dtype=int64>, 'penne_rigate': <tf.Tensor 'ExpandDims_461:0' shape=(None, 1) dtype=int64>, 'pepper': <tf.Tensor 'ExpandDims_462:0' shape=(None, 1) dtype=int64>, 'pepperoncini_pepper': <tf.Tensor 'ExpandDims_463:0' shape=(None, 1) dtype=int64>, 'pepperoni': <tf.Tensor 'ExpandDims_464:0' shape=(None, 1) dtype=int64>, 'pesto': <tf.Tensor 'ExpandDims_465:0' shape=(None, 1) dtype=int64>, 'phyllo_dough': <tf.Tensor 'ExpandDims_466:0' shape=(None, 1) dtype=int64>, 'picante_sauce': <tf.Tensor 'ExpandDims_467:0' shape=(None, 1) dtype=int64>, 'pickle': <tf.Tensor 'ExpandDims_468:0' shape=(None, 1) dtype=int64>, 'pickling_spices': <tf.Tensor 'ExpandDims_469:0' shape=(None, 1) dtype=int64>, 'pie_crust': <tf.Tensor 'ExpandDims_470:0' shape=(None, 1) dtype=int64>, 'pierogi': <tf.Tensor 'ExpandDims_471:0' shape=(None, 1) dtype=int64>, 'pimiento': <tf.Tensor 'ExpandDims_472:0' shape=(None, 1) dtype=int64>, 'pine_nuts': <tf.Tensor 'ExpandDims_473:0' shape=(None, 1) dtype=int64>, 'pineapple': <tf.Tensor 'ExpandDims_474:0' shape=(None, 1) dtype=int64>, 'pink_bean': <tf.Tensor 'ExpandDims_475:0' shape=(None, 1) dtype=int64>, 'pinto_bean': <tf.Tensor 'ExpandDims_476:0' shape=(None, 1) dtype=int64>, 'piquillo_pepper': <tf.Tensor 'ExpandDims_477:0' shape=(None, 1) dtype=int64>, 'pistachio': <tf.Tensor 'ExpandDims_478:0' shape=(None, 1) dtype=int64>, 'pita_bread': <tf.Tensor 'ExpandDims_479:0' shape=(None, 1) dtype=int64>, 'pitted_prunes': <tf.Tensor 'ExpandDims_480:0' shape=(None, 1) dtype=int64>, 'pizza_crust': <tf.Tensor 'ExpandDims_481:0' shape=(None, 1) dtype=int64>, 'pizza_dough': <tf.Tensor 'ExpandDims_482:0' shape=(None, 1) dtype=int64>, 'pizza_sauce': <tf.Tensor 'ExpandDims_483:0' shape=(None, 1) dtype=int64>, 'plum': <tf.Tensor 'ExpandDims_484:0' shape=(None, 1) dtype=int64>, 'plum_tomato': <tf.Tensor 'ExpandDims_485:0' shape=(None, 1) dtype=int64>, 'poblano_pepper': <tf.Tensor 'ExpandDims_486:0' shape=(None, 1) dtype=int64>, 'polenta': <tf.Tensor 'ExpandDims_487:0' shape=(None, 1) dtype=int64>, 'pomegranate': <tf.Tensor 'ExpandDims_488:0' shape=(None, 1) dtype=int64>, 'ponzu': <tf.Tensor 'ExpandDims_489:0' shape=(None, 1) dtype=int64>, 'poppy_seed': <tf.Tensor 'ExpandDims_490:0' shape=(None, 1) dtype=int64>, 'porcino': <tf.Tensor 'ExpandDims_491:0' shape=(None, 1) dtype=int64>, 'pork': <tf.Tensor 'ExpandDims_492:0' shape=(None, 1) dtype=int64>, 'portabella': <tf.Tensor 'ExpandDims_493:0' shape=(None, 1) dtype=int64>, 'potato': <tf.Tensor 'ExpandDims_494:0' shape=(None, 1) dtype=int64>, 'potato_starch': <tf.Tensor 'ExpandDims_495:0' shape=(None, 1) dtype=int64>, 'poultry_seasoning': <tf.Tensor 'ExpandDims_496:0' shape=(None, 1) dtype=int64>, 'powdered_sugar': <tf.Tensor 'ExpandDims_497:0' shape=(None, 1) dtype=int64>, 'prawn': <tf.Tensor 'ExpandDims_498:0' shape=(None, 1) dtype=int64>, 'prepared_horseradish': <tf.Tensor 'ExpandDims_499:0' shape=(None, 1) dtype=int64>, 'preserved_lemon': <tf.Tensor 'ExpandDims_500:0' shape=(None, 1) dtype=int64>, 'processed_cheese': <tf.Tensor 'ExpandDims_501:0' shape=(None, 1) dtype=int64>, 'prosciutto': <tf.Tensor 'ExpandDims_502:0' shape=(None, 1) dtype=int64>, 'provolone': <tf.Tensor 'ExpandDims_503:0' shape=(None, 1) dtype=int64>, 'prune': <tf.Tensor 'ExpandDims_504:0' shape=(None, 1) dtype=int64>, 'puff_pastry': <tf.Tensor 'ExpandDims_505:0' shape=(None, 1) dtype=int64>, 'pumpkin': <tf.Tensor 'ExpandDims_506:0' shape=(None, 1) dtype=int64>, 'pumpkin_pie_spice': <tf.Tensor 'ExpandDims_507:0' shape=(None, 1) dtype=int64>, 'quark': <tf.Tensor 'ExpandDims_508:0' shape=(None, 1) dtype=int64>, 'queso': <tf.Tensor 'ExpandDims_509:0' shape=(None, 1) dtype=int64>, 'quinoa': <tf.Tensor 'ExpandDims_510:0' shape=(None, 1) dtype=int64>, 'rabbit': <tf.Tensor 'ExpandDims_511:0' shape=(None, 1) dtype=int64>, 'radicchio': <tf.Tensor 'ExpandDims_512:0' shape=(None, 1) dtype=int64>, 'radish': <tf.Tensor 'ExpandDims_513:0' shape=(None, 1) dtype=int64>, 'raisin': <tf.Tensor 'ExpandDims_514:0' shape=(None, 1) dtype=int64>, 'ramen_noodles': <tf.Tensor 'ExpandDims_515:0' shape=(None, 1) dtype=int64>, 'Ranch_salad_dressing': <tf.Tensor 'ExpandDims_16:0' shape=(None, 1) dtype=int64>, 'rapeseed_oil': <tf.Tensor 'ExpandDims_516:0' shape=(None, 1) dtype=int64>, 'ras_el_hanout': <tf.Tensor 'ExpandDims_517:0' shape=(None, 1) dtype=int64>, 'raw_sugar': <tf.Tensor 'ExpandDims_518:0' shape=(None, 1) dtype=int64>, 'red_beans': <tf.Tensor 'ExpandDims_519:0' shape=(None, 1) dtype=int64>, 'red_chile_powder': <tf.Tensor 'ExpandDims_520:0' shape=(None, 1) dtype=int64>, 'red_food_coloring': <tf.Tensor 'ExpandDims_521:0' shape=(None, 1) dtype=int64>, 'red_pepper_flakes': <tf.Tensor 'ExpandDims_522:0' shape=(None, 1) dtype=int64>, 'red_pepper_sauce': <tf.Tensor 'ExpandDims_523:0' shape=(None, 1) dtype=int64>, 'red_snapper': <tf.Tensor 'ExpandDims_524:0' shape=(None, 1) dtype=int64>, 'refried_beans': <tf.Tensor 'ExpandDims_525:0' shape=(None, 1) dtype=int64>, 'rhubarb': <tf.Tensor 'ExpandDims_526:0' shape=(None, 1) dtype=int64>, 'rice': <tf.Tensor 'ExpandDims_527:0' shape=(None, 1) dtype=int64>, 'rice_flour': <tf.Tensor 'ExpandDims_528:0' shape=(None, 1) dtype=int64>, 'rice_noodles': <tf.Tensor 'ExpandDims_529:0' shape=(None, 1) dtype=int64>, 'rice_vermicelli': <tf.Tensor 'ExpandDims_530:0' shape=(None, 1) dtype=int64>, 'rice_vinegar': <tf.Tensor 'ExpandDims_531:0' shape=(None, 1) dtype=int64>, 'rice_wine': <tf.Tensor 'ExpandDims_532:0' shape=(None, 1) dtype=int64>, 'ricotta_cheese': <tf.Tensor 'ExpandDims_533:0' shape=(None, 1) dtype=int64>, 'roll': <tf.Tensor 'ExpandDims_534:0' shape=(None, 1) dtype=int64>, 'romaine_lettuce': <tf.Tensor 'ExpandDims_535:0' shape=(None, 1) dtype=int64>, 'romano_cheese': <tf.Tensor 'ExpandDims_536:0' shape=(None, 1) dtype=int64>, 'rose_water': <tf.Tensor 'ExpandDims_537:0' shape=(None, 1) dtype=int64>, 'rosemary': <tf.Tensor 'ExpandDims_538:0' shape=(None, 1) dtype=int64>, 'rotel_tomatoes': <tf.Tensor 'ExpandDims_539:0' shape=(None, 1) dtype=int64>, 'rum': <tf.Tensor 'ExpandDims_540:0' shape=(None, 1) dtype=int64>, 'rutabaga': <tf.Tensor 'ExpandDims_541:0' shape=(None, 1) dtype=int64>, 'rye_bread': <tf.Tensor 'ExpandDims_542:0' shape=(None, 1) dtype=int64>, 'saffron': <tf.Tensor 'ExpandDims_543:0' shape=(None, 1) dtype=int64>, 'sage': <tf.Tensor 'ExpandDims_544:0' shape=(None, 1) dtype=int64>, 'salmon': <tf.Tensor 'ExpandDims_545:0' shape=(None, 1) dtype=int64>, 'salsa': <tf.Tensor 'ExpandDims_546:0' shape=(None, 1) dtype=int64>, 'salsify': <tf.Tensor 'ExpandDims_547:0' shape=(None, 1) dtype=int64>, 'salt': <tf.Tensor 'ExpandDims_548:0' shape=(None, 1) dtype=int64>, 'salt_pork': <tf.Tensor 'ExpandDims_549:0' shape=(None, 1) dtype=int64>, 'sambal': <tf.Tensor 'ExpandDims_550:0' shape=(None, 1) dtype=int64>, 'sardine': <tf.Tensor 'ExpandDims_551:0' shape=(None, 1) dtype=int64>, 'satay': <tf.Tensor 'ExpandDims_552:0' shape=(None, 1) dtype=int64>, 'sauerkraut': <tf.Tensor 'ExpandDims_553:0' shape=(None, 1) dtype=int64>, 'sausage': <tf.Tensor 'ExpandDims_554:0' shape=(None, 1) dtype=int64>, 'savory': <tf.Tensor 'ExpandDims_555:0' shape=(None, 1) dtype=int64>, 'sazon': <tf.Tensor 'ExpandDims_556:0' shape=(None, 1) dtype=int64>, 'sazon_goya': <tf.Tensor 'ExpandDims_557:0' shape=(None, 1) dtype=int64>, 'scallop': <tf.Tensor 'ExpandDims_558:0' shape=(None, 1) dtype=int64>, 'screwpine_leaves': <tf.Tensor 'ExpandDims_559:0' shape=(None, 1) dtype=int64>, 'sea_cucumber': <tf.Tensor 'ExpandDims_560:0' shape=(None, 1) dtype=int64>, 'seafood': <tf.Tensor 'ExpandDims_561:0' shape=(None, 1) dtype=int64>, 'seitan': <tf.Tensor 'ExpandDims_562:0' shape=(None, 1) dtype=int64>, 'semolina': <tf.Tensor 'ExpandDims_563:0' shape=(None, 1) dtype=int64>, 'serrano_pepper': <tf.Tensor 'ExpandDims_564:0' shape=(None, 1) dtype=int64>, 'sesame_oil': <tf.Tensor 'ExpandDims_565:0' shape=(None, 1) dtype=int64>, 'sesame_seed': <tf.Tensor 'ExpandDims_566:0' shape=(None, 1) dtype=int64>, 'shallot': <tf.Tensor 'ExpandDims_567:0' shape=(None, 1) dtype=int64>, 'sherry': <tf.Tensor 'ExpandDims_568:0' shape=(None, 1) dtype=int64>, 'shimeji_mushroom': <tf.Tensor 'ExpandDims_569:0' shape=(None, 1) dtype=int64>, 'shirataki_noodles': <tf.Tensor 'ExpandDims_570:0' shape=(None, 1) dtype=int64>, 'shitake': <tf.Tensor 'ExpandDims_571:0' shape=(None, 1) dtype=int64>, 'shortcrust_pastry': <tf.Tensor 'ExpandDims_572:0' shape=(None, 1) dtype=int64>, 'shortening': <tf.Tensor 'ExpandDims_573:0' shape=(None, 1) dtype=int64>, 'shoyu': <tf.Tensor 'ExpandDims_574:0' shape=(None, 1) dtype=int64>, 'shrimp': <tf.Tensor 'ExpandDims_575:0' shape=(None, 1) dtype=int64>, 'shrimp_paste': <tf.Tensor 'ExpandDims_576:0' shape=(None, 1) dtype=int64>, 'sliced_pepperoni': <tf.Tensor 'ExpandDims_577:0' shape=(None, 1) dtype=int64>, 'smen': <tf.Tensor 'ExpandDims_578:0' shape=(None, 1) dtype=int64>, 'smoke_seasoning': <tf.Tensor 'ExpandDims_579:0' shape=(None, 1) dtype=int64>, 'smoked_salmon': <tf.Tensor 'ExpandDims_580:0' shape=(None, 1) dtype=int64>, 'snap_pea': <tf.Tensor 'ExpandDims_581:0' shape=(None, 1) dtype=int64>, 'snow_pea': <tf.Tensor 'ExpandDims_582:0' shape=(None, 1) dtype=int64>, 'soba_noodles': <tf.Tensor 'ExpandDims_583:0' shape=(None, 1) dtype=int64>, 'sofrito_sauce': <tf.Tensor 'ExpandDims_584:0' shape=(None, 1) dtype=int64>, 'sole': <tf.Tensor 'ExpandDims_585:0' shape=(None, 1) dtype=int64>, 'sorrel': <tf.Tensor 'ExpandDims_586:0' shape=(None, 1) dtype=int64>, 'sour_cream': <tf.Tensor 'ExpandDims_587:0' shape=(None, 1) dtype=int64>, 'sourdough_bread': <tf.Tensor 'ExpandDims_588:0' shape=(None, 1) dtype=int64>, 'soy_sauce': <tf.Tensor 'ExpandDims_589:0' shape=(None, 1) dtype=int64>, 'soybean': <tf.Tensor 'ExpandDims_590:0' shape=(None, 1) dtype=int64>, 'soymilk': <tf.Tensor 'ExpandDims_591:0' shape=(None, 1) dtype=int64>, 'spaetzle': <tf.Tensor 'ExpandDims_592:0' shape=(None, 1) dtype=int64>, 'spaghetti_squash': <tf.Tensor 'ExpandDims_593:0' shape=(None, 1) dtype=int64>, 'spanish_cheese': <tf.Tensor 'ExpandDims_594:0' shape=(None, 1) dtype=int64>, 'spanish_chorizo': <tf.Tensor 'ExpandDims_595:0' shape=(None, 1) dtype=int64>, 'spareribs': <tf.Tensor 'ExpandDims_596:0' shape=(None, 1) dtype=int64>, 'spearmint': <tf.Tensor 'ExpandDims_597:0' shape=(None, 1) dtype=int64>, 'specialty_cheese': <tf.Tensor 'ExpandDims_598:0' shape=(None, 1) dtype=int64>, 'spinach': <tf.Tensor 'ExpandDims_599:0' shape=(None, 1) dtype=int64>, 'squash': <tf.Tensor 'ExpandDims_600:0' shape=(None, 1) dtype=int64>, 'squid': <tf.Tensor 'ExpandDims_601:0' shape=(None, 1) dtype=int64>, 'sriracha_hot_sauce': <tf.Tensor 'ExpandDims_602:0' shape=(None, 1) dtype=int64>, 'star_anise': <tf.Tensor 'ExpandDims_603:0' shape=(None, 1) dtype=int64>, 'steaks': <tf.Tensor 'ExpandDims_604:0' shape=(None, 1) dtype=int64>, 'stilton': <tf.Tensor 'ExpandDims_605:0' shape=(None, 1) dtype=int64>, 'stir_fry_vegetables': <tf.Tensor 'ExpandDims_606:0' shape=(None, 1) dtype=int64>, 'stock': <tf.Tensor 'ExpandDims_607:0' shape=(None, 1) dtype=int64>, 'straw_mushroom': <tf.Tensor 'ExpandDims_608:0' shape=(None, 1) dtype=int64>, 'strawberry': <tf.Tensor 'ExpandDims_609:0' shape=(None, 1) dtype=int64>, 'stuffing': <tf.Tensor 'ExpandDims_610:0' shape=(None, 1) dtype=int64>, 'suet': <tf.Tensor 'ExpandDims_611:0' shape=(None, 1) dtype=int64>, 'sugar': <tf.Tensor 'ExpandDims_612:0' shape=(None, 1) dtype=int64>, 'sultana': <tf.Tensor 'ExpandDims_613:0' shape=(None, 1) dtype=int64>, 'sumac': <tf.Tensor 'ExpandDims_614:0' shape=(None, 1) dtype=int64>, 'sunflower_oil': <tf.Tensor 'ExpandDims_615:0' shape=(None, 1) dtype=int64>, 'sunflower_seed': <tf.Tensor 'ExpandDims_616:0' shape=(None, 1) dtype=int64>, 'sushi_rice': <tf.Tensor 'ExpandDims_617:0' shape=(None, 1) dtype=int64>, 'sweet_pepper': <tf.Tensor 'ExpandDims_618:0' shape=(None, 1) dtype=int64>, 'sweet_potato': <tf.Tensor 'ExpandDims_619:0' shape=(None, 1) dtype=int64>, 'swiss_cheese': <tf.Tensor 'ExpandDims_620:0' shape=(None, 1) dtype=int64>, 'swordfish': <tf.Tensor 'ExpandDims_621:0' shape=(None, 1) dtype=int64>, 'szechuan_peppercorn': <tf.Tensor 'ExpandDims_622:0' shape=(None, 1) dtype=int64>, 'tabasco_sauce': <tf.Tensor 'ExpandDims_623:0' shape=(None, 1) dtype=int64>, 'taco_seasoning': <tf.Tensor 'ExpandDims_624:0' shape=(None, 1) dtype=int64>, 'tahini': <tf.Tensor 'ExpandDims_625:0' shape=(None, 1) dtype=int64>, 'tamarind': <tf.Tensor 'ExpandDims_626:0' shape=(None, 1) dtype=int64>, 'tapioca': <tf.Tensor 'ExpandDims_627:0' shape=(None, 1) dtype=int64>, 'tarragon': <tf.Tensor 'ExpandDims_628:0' shape=(None, 1) dtype=int64>, 'tatsoi': <tf.Tensor 'ExpandDims_629:0' shape=(None, 1) dtype=int64>, 'tea': <tf.Tensor 'ExpandDims_630:0' shape=(None, 1) dtype=int64>, 'tempeh': <tf.Tensor 'ExpandDims_631:0' shape=(None, 1) dtype=int64>, 'tequila': <tf.Tensor 'ExpandDims_632:0' shape=(None, 1) dtype=int64>, 'teriyaki_marinade': <tf.Tensor 'ExpandDims_633:0' shape=(None, 1) dtype=int64>, 'teriyaki_sauce': <tf.Tensor 'ExpandDims_634:0' shape=(None, 1) dtype=int64>, 'Thai_chile_pepper': <tf.Tensor 'ExpandDims_17:0' shape=(None, 1) dtype=int64>, 'Thai_curry_paste': <tf.Tensor 'ExpandDims_18:0' shape=(None, 1) dtype=int64>, 'thousand_island_dressing': <tf.Tensor 'ExpandDims_635:0' shape=(None, 1) dtype=int64>, 'thyme': <tf.Tensor 'ExpandDims_636:0' shape=(None, 1) dtype=int64>, 'tilapia_fillets': <tf.Tensor 'ExpandDims_637:0' shape=(None, 1) dtype=int64>, 'tofu': <tf.Tensor 'ExpandDims_638:0' shape=(None, 1) dtype=int64>, 'togarashi': <tf.Tensor 'ExpandDims_639:0' shape=(None, 1) dtype=int64>, 'tomatillos': <tf.Tensor 'ExpandDims_640:0' shape=(None, 1) dtype=int64>, 'tomato': <tf.Tensor 'ExpandDims_641:0' shape=(None, 1) dtype=int64>, 'tomato_juice': <tf.Tensor 'ExpandDims_642:0' shape=(None, 1) dtype=int64>, 'tomato_paste': <tf.Tensor 'ExpandDims_643:0' shape=(None, 1) dtype=int64>, 'tomato_puree': <tf.Tensor 'ExpandDims_644:0' shape=(None, 1) dtype=int64>, 'tomato_sauce': <tf.Tensor 'ExpandDims_645:0' shape=(None, 1) dtype=int64>, 'tomato_soup': <tf.Tensor 'ExpandDims_646:0' shape=(None, 1) dtype=int64>, 'toor_dal': <tf.Tensor 'ExpandDims_647:0' shape=(None, 1) dtype=int64>, 'tortilla': <tf.Tensor 'ExpandDims_648:0' shape=(None, 1) dtype=int64>, 'trout': <tf.Tensor 'ExpandDims_649:0' shape=(None, 1) dtype=int64>, 'truffle': <tf.Tensor 'ExpandDims_650:0' shape=(None, 1) dtype=int64>, 'truffle_oil': <tf.Tensor 'ExpandDims_651:0' shape=(None, 1) dtype=int64>, 'tuna': <tf.Tensor 'ExpandDims_652:0' shape=(None, 1) dtype=int64>, 'turbinado_sugar': <tf.Tensor 'ExpandDims_653:0' shape=(None, 1) dtype=int64>, 'turkey': <tf.Tensor 'ExpandDims_654:0' shape=(None, 1) dtype=int64>, 'turmeric': <tf.Tensor 'ExpandDims_655:0' shape=(None, 1) dtype=int64>, 'turnip': <tf.Tensor 'ExpandDims_656:0' shape=(None, 1) dtype=int64>, 'udon_noodles': <tf.Tensor 'ExpandDims_657:0' shape=(None, 1) dtype=int64>, 'umeboshi': <tf.Tensor 'ExpandDims_658:0' shape=(None, 1) dtype=int64>, 'unsweetened_chocolate': <tf.Tensor 'ExpandDims_659:0' shape=(None, 1) dtype=int64>, 'urad_dal': <tf.Tensor 'ExpandDims_660:0' shape=(None, 1) dtype=int64>, 'vanilla': <tf.Tensor 'ExpandDims_661:0' shape=(None, 1) dtype=int64>, 'veal': <tf.Tensor 'ExpandDims_662:0' shape=(None, 1) dtype=int64>, 'vegan_brand_names': <tf.Tensor 'ExpandDims_663:0' shape=(None, 1) dtype=int64>, 'vegetable_juice': <tf.Tensor 'ExpandDims_664:0' shape=(None, 1) dtype=int64>, 'vegetable_oil': <tf.Tensor 'ExpandDims_665:0' shape=(None, 1) dtype=int64>, 'vegetable_stock': <tf.Tensor 'ExpandDims_666:0' shape=(None, 1) dtype=int64>, 'vegetarian_brand_names': <tf.Tensor 'ExpandDims_667:0' shape=(None, 1) dtype=int64>, 'velveeta': <tf.Tensor 'ExpandDims_668:0' shape=(None, 1) dtype=int64>, 'venison': <tf.Tensor 'ExpandDims_669:0' shape=(None, 1) dtype=int64>, 'vermicelli': <tf.Tensor 'ExpandDims_670:0' shape=(None, 1) dtype=int64>, 'vermouth': <tf.Tensor 'ExpandDims_671:0' shape=(None, 1) dtype=int64>, 'Vidalia_onion': <tf.Tensor 'ExpandDims_19:0' shape=(None, 1) dtype=int64>, 'vinegar': <tf.Tensor 'ExpandDims_672:0' shape=(None, 1) dtype=int64>, 'vital_wheat_gluten': <tf.Tensor 'ExpandDims_673:0' shape=(None, 1) dtype=int64>, 'vodka': <tf.Tensor 'ExpandDims_674:0' shape=(None, 1) dtype=int64>, 'walnut': <tf.Tensor 'ExpandDims_675:0' shape=(None, 1) dtype=int64>, 'wasabi': <tf.Tensor 'ExpandDims_676:0' shape=(None, 1) dtype=int64>, 'washed_rind_cheese': <tf.Tensor 'ExpandDims_677:0' shape=(None, 1) dtype=int64>, 'water': <tf.Tensor 'ExpandDims_678:0' shape=(None, 1) dtype=int64>, 'water_chestnut': <tf.Tensor 'ExpandDims_679:0' shape=(None, 1) dtype=int64>, 'watercress': <tf.Tensor 'ExpandDims_680:0' shape=(None, 1) dtype=int64>, 'watermelon': <tf.Tensor 'ExpandDims_681:0' shape=(None, 1) dtype=int64>, 'wheat_bread': <tf.Tensor 'ExpandDims_682:0' shape=(None, 1) dtype=int64>, 'whipping_cream': <tf.Tensor 'ExpandDims_683:0' shape=(None, 1) dtype=int64>, 'white_bean': <tf.Tensor 'ExpandDims_684:0' shape=(None, 1) dtype=int64>, 'white_bread': <tf.Tensor 'ExpandDims_685:0' shape=(None, 1) dtype=int64>, 'white_cheese': <tf.Tensor 'ExpandDims_686:0' shape=(None, 1) dtype=int64>, 'white_fish': <tf.Tensor 'ExpandDims_687:0' shape=(None, 1) dtype=int64>, 'white_onion': <tf.Tensor 'ExpandDims_688:0' shape=(None, 1) dtype=int64>, 'white_peppercorns': <tf.Tensor 'ExpandDims_689:0' shape=(None, 1) dtype=int64>, 'white_sesame_seeds': <tf.Tensor 'ExpandDims_690:0' shape=(None, 1) dtype=int64>, 'whole_grain_mustard': <tf.Tensor 'ExpandDims_691:0' shape=(None, 1) dtype=int64>, 'whole_milk': <tf.Tensor 'ExpandDims_692:0' shape=(None, 1) dtype=int64>, 'wild_rice': <tf.Tensor 'ExpandDims_693:0' shape=(None, 1) dtype=int64>, 'wine': <tf.Tensor 'ExpandDims_694:0' shape=(None, 1) dtype=int64>, 'wine_vinegar': <tf.Tensor 'ExpandDims_695:0' shape=(None, 1) dtype=int64>, 'winter_squash': <tf.Tensor 'ExpandDims_696:0' shape=(None, 1) dtype=int64>, 'worcestershire_sauce': <tf.Tensor 'ExpandDims_697:0' shape=(None, 1) dtype=int64>, 'xanthan_gum': <tf.Tensor 'ExpandDims_698:0' shape=(None, 1) dtype=int64>, 'yams': <tf.Tensor 'ExpandDims_699:0' shape=(None, 1) dtype=int64>, 'yeast': <tf.Tensor 'ExpandDims_700:0' shape=(None, 1) dtype=int64>, 'yellow_curry_paste': <tf.Tensor 'ExpandDims_701:0' shape=(None, 1) dtype=int64>, 'yellow_food_coloring': <tf.Tensor 'ExpandDims_702:0' shape=(None, 1) dtype=int64>, 'yellow_split_pea': <tf.Tensor 'ExpandDims_703:0' shape=(None, 1) dtype=int64>, 'yellow_squash': <tf.Tensor 'ExpandDims_704:0' shape=(None, 1) dtype=int64>, 'yogurt': <tf.Tensor 'ExpandDims_705:0' shape=(None, 1) dtype=int64>, 'zaatar': <tf.Tensor 'ExpandDims_706:0' shape=(None, 1) dtype=int64>, 'zest': <tf.Tensor 'ExpandDims_707:0' shape=(None, 1) dtype=int64>, 'zucchini': <tf.Tensor 'ExpandDims_708:0' shape=(None, 1) dtype=int64>}\n",
            "Consider rewriting this model with the Functional API.\n",
            "47/47 [==============================] - ETA: 0s - loss: 2.4205 - sparse_categorical_accuracy: 0.1869WARNING:tensorflow:Layers in a Sequential model should only have a single input tensor, but we receive a <class 'dict'> input: {'acorn_squash': <tf.Tensor 'ExpandDims_20:0' shape=(None, 1) dtype=int64>, 'adobo': <tf.Tensor 'ExpandDims_21:0' shape=(None, 1) dtype=int64>, 'african_birdseye_chile_pepper': <tf.Tensor 'ExpandDims_22:0' shape=(None, 1) dtype=int64>, 'ale': <tf.Tensor 'ExpandDims_23:0' shape=(None, 1) dtype=int64>, 'aleppo_pepper': <tf.Tensor 'ExpandDims_24:0' shape=(None, 1) dtype=int64>, 'alfalfa_sprouts': <tf.Tensor 'ExpandDims_25:0' shape=(None, 1) dtype=int64>, 'alfredo_sauce': <tf.Tensor 'ExpandDims_26:0' shape=(None, 1) dtype=int64>, 'allspice': <tf.Tensor 'ExpandDims_27:0' shape=(None, 1) dtype=int64>, 'almond': <tf.Tensor 'ExpandDims_28:0' shape=(None, 1) dtype=int64>, 'almond_butter': <tf.Tensor 'ExpandDims_29:0' shape=(None, 1) dtype=int64>, 'amaretti': <tf.Tensor 'ExpandDims_30:0' shape=(None, 1) dtype=int64>, 'amchoor_powder': <tf.Tensor 'ExpandDims_31:0' shape=(None, 1) dtype=int64>, 'ancho_chile_pepper': <tf.Tensor 'ExpandDims_32:0' shape=(None, 1) dtype=int64>, 'anchovy': <tf.Tensor 'ExpandDims_33:0' shape=(None, 1) dtype=int64>, 'anchovy_paste': <tf.Tensor 'ExpandDims_34:0' shape=(None, 1) dtype=int64>, 'andouille_sausage': <tf.Tensor 'ExpandDims_35:0' shape=(None, 1) dtype=int64>, 'anise': <tf.Tensor 'ExpandDims_36:0' shape=(None, 1) dtype=int64>, 'apple': <tf.Tensor 'ExpandDims_37:0' shape=(None, 1) dtype=int64>, 'apple_butter': <tf.Tensor 'ExpandDims_38:0' shape=(None, 1) dtype=int64>, 'apple_juice': <tf.Tensor 'ExpandDims_39:0' shape=(None, 1) dtype=int64>, 'applesauce': <tf.Tensor 'ExpandDims_40:0' shape=(None, 1) dtype=int64>, 'apricot': <tf.Tensor 'ExpandDims_41:0' shape=(None, 1) dtype=int64>, 'arborio_rice': <tf.Tensor 'ExpandDims_42:0' shape=(None, 1) dtype=int64>, 'argan_oil': <tf.Tensor 'ExpandDims_43:0' shape=(None, 1) dtype=int64>, 'arrowroot': <tf.Tensor 'ExpandDims_44:0' shape=(None, 1) dtype=int64>, 'artichoke': <tf.Tensor 'ExpandDims_45:0' shape=(None, 1) dtype=int64>, 'artificial_sweetener': <tf.Tensor 'ExpandDims_46:0' shape=(None, 1) dtype=int64>, 'arugula': <tf.Tensor 'ExpandDims_47:0' shape=(None, 1) dtype=int64>, 'asadero': <tf.Tensor 'ExpandDims_48:0' shape=(None, 1) dtype=int64>, 'asafetida': <tf.Tensor 'ExpandDims_49:0' shape=(None, 1) dtype=int64>, 'asiago_cheese': <tf.Tensor 'ExpandDims_50:0' shape=(None, 1) dtype=int64>, 'asparagus': <tf.Tensor 'ExpandDims_51:0' shape=(None, 1) dtype=int64>, 'assorted_vegetables': <tf.Tensor 'ExpandDims_52:0' shape=(None, 1) dtype=int64>, 'avocado': <tf.Tensor 'ExpandDims_53:0' shape=(None, 1) dtype=int64>, 'azuki_bean': <tf.Tensor 'ExpandDims_54:0' shape=(None, 1) dtype=int64>, 'baby_carrot': <tf.Tensor 'ExpandDims_55:0' shape=(None, 1) dtype=int64>, 'baby_corn': <tf.Tensor 'ExpandDims_56:0' shape=(None, 1) dtype=int64>, 'baby_onions': <tf.Tensor 'ExpandDims_57:0' shape=(None, 1) dtype=int64>, 'baby_spinach': <tf.Tensor 'ExpandDims_58:0' shape=(None, 1) dtype=int64>, 'bacon': <tf.Tensor 'ExpandDims_59:0' shape=(None, 1) dtype=int64>, 'bacon_grease': <tf.Tensor 'ExpandDims_60:0' shape=(None, 1) dtype=int64>, 'bagel': <tf.Tensor 'ExpandDims_61:0' shape=(None, 1) dtype=int64>, 'baguette': <tf.Tensor 'ExpandDims_62:0' shape=(None, 1) dtype=int64>, 'baked_beans': <tf.Tensor 'ExpandDims_63:0' shape=(None, 1) dtype=int64>, 'baking_powder': <tf.Tensor 'ExpandDims_64:0' shape=(None, 1) dtype=int64>, 'baking_soda': <tf.Tensor 'ExpandDims_65:0' shape=(None, 1) dtype=int64>, 'balsamic_vinegar': <tf.Tensor 'ExpandDims_66:0' shape=(None, 1) dtype=int64>, 'bamboo_shoot': <tf.Tensor 'ExpandDims_67:0' shape=(None, 1) dtype=int64>, 'bamboo_skewers': <tf.Tensor 'ExpandDims_68:0' shape=(None, 1) dtype=int64>, 'banana': <tf.Tensor 'ExpandDims_69:0' shape=(None, 1) dtype=int64>, 'barbecue_sauce': <tf.Tensor 'ExpandDims_70:0' shape=(None, 1) dtype=int64>, 'barley': <tf.Tensor 'ExpandDims_71:0' shape=(None, 1) dtype=int64>, 'basil': <tf.Tensor 'ExpandDims_72:0' shape=(None, 1) dtype=int64>, 'basmati_rice': <tf.Tensor 'ExpandDims_73:0' shape=(None, 1) dtype=int64>, 'bass': <tf.Tensor 'ExpandDims_74:0' shape=(None, 1) dtype=int64>, 'bay_leaf': <tf.Tensor 'ExpandDims_75:0' shape=(None, 1) dtype=int64>, 'bean_sprouts': <tf.Tensor 'ExpandDims_76:0' shape=(None, 1) dtype=int64>, 'bean_thread': <tf.Tensor 'ExpandDims_77:0' shape=(None, 1) dtype=int64>, 'beans': <tf.Tensor 'ExpandDims_78:0' shape=(None, 1) dtype=int64>, 'bechamel': <tf.Tensor 'ExpandDims_79:0' shape=(None, 1) dtype=int64>, 'beef': <tf.Tensor 'ExpandDims_80:0' shape=(None, 1) dtype=int64>, 'beef_consomme': <tf.Tensor 'ExpandDims_81:0' shape=(None, 1) dtype=int64>, 'beef_gravy': <tf.Tensor 'ExpandDims_82:0' shape=(None, 1) dtype=int64>, 'beef_round_cuts': <tf.Tensor 'ExpandDims_83:0' shape=(None, 1) dtype=int64>, 'beef_stock': <tf.Tensor 'ExpandDims_84:0' shape=(None, 1) dtype=int64>, 'beer': <tf.Tensor 'ExpandDims_85:0' shape=(None, 1) dtype=int64>, 'beet': <tf.Tensor 'ExpandDims_86:0' shape=(None, 1) dtype=int64>, 'biscuit_mix': <tf.Tensor 'ExpandDims_87:0' shape=(None, 1) dtype=int64>, 'bitter_melon': <tf.Tensor 'ExpandDims_88:0' shape=(None, 1) dtype=int64>, 'black_bean': <tf.Tensor 'ExpandDims_89:0' shape=(None, 1) dtype=int64>, 'black_eyed_pea': <tf.Tensor 'ExpandDims_90:0' shape=(None, 1) dtype=int64>, 'black_forest_ham': <tf.Tensor 'ExpandDims_91:0' shape=(None, 1) dtype=int64>, 'black_mushroom': <tf.Tensor 'ExpandDims_92:0' shape=(None, 1) dtype=int64>, 'black_olive': <tf.Tensor 'ExpandDims_93:0' shape=(None, 1) dtype=int64>, 'black_truffle': <tf.Tensor 'ExpandDims_94:0' shape=(None, 1) dtype=int64>, 'blackberry': <tf.Tensor 'ExpandDims_95:0' shape=(None, 1) dtype=int64>, 'blue_cheese': <tf.Tensor 'ExpandDims_96:0' shape=(None, 1) dtype=int64>, 'bok_choy': <tf.Tensor 'ExpandDims_97:0' shape=(None, 1) dtype=int64>, 'bouillon': <tf.Tensor 'ExpandDims_98:0' shape=(None, 1) dtype=int64>, 'bouquet_garni': <tf.Tensor 'ExpandDims_99:0' shape=(None, 1) dtype=int64>, 'bourbon': <tf.Tensor 'ExpandDims_100:0' shape=(None, 1) dtype=int64>, 'brandy': <tf.Tensor 'ExpandDims_101:0' shape=(None, 1) dtype=int64>, 'bratwursts': <tf.Tensor 'ExpandDims_102:0' shape=(None, 1) dtype=int64>, 'bread': <tf.Tensor 'ExpandDims_103:0' shape=(None, 1) dtype=int64>, 'bread_bun': <tf.Tensor 'ExpandDims_104:0' shape=(None, 1) dtype=int64>, 'breadcrumb': <tf.Tensor 'ExpandDims_105:0' shape=(None, 1) dtype=int64>, 'brick_cheese': <tf.Tensor 'ExpandDims_106:0' shape=(None, 1) dtype=int64>, 'brie_cheese': <tf.Tensor 'ExpandDims_107:0' shape=(None, 1) dtype=int64>, 'broccoli': <tf.Tensor 'ExpandDims_108:0' shape=(None, 1) dtype=int64>, 'broccoli_raab': <tf.Tensor 'ExpandDims_109:0' shape=(None, 1) dtype=int64>, 'broth': <tf.Tensor 'ExpandDims_110:0' shape=(None, 1) dtype=int64>, 'brown_gravy_mix': <tf.Tensor 'ExpandDims_111:0' shape=(None, 1) dtype=int64>, 'brown_onion': <tf.Tensor 'ExpandDims_112:0' shape=(None, 1) dtype=int64>, 'brown_rice': <tf.Tensor 'ExpandDims_113:0' shape=(None, 1) dtype=int64>, 'brown_sugar': <tf.Tensor 'ExpandDims_114:0' shape=(None, 1) dtype=int64>, 'brussels_sprout': <tf.Tensor 'ExpandDims_115:0' shape=(None, 1) dtype=int64>, 'buckwheat': <tf.Tensor 'ExpandDims_116:0' shape=(None, 1) dtype=int64>, 'bulgur_wheat': <tf.Tensor 'ExpandDims_117:0' shape=(None, 1) dtype=int64>, 'burdock_root': <tf.Tensor 'ExpandDims_118:0' shape=(None, 1) dtype=int64>, 'butter': <tf.Tensor 'ExpandDims_119:0' shape=(None, 1) dtype=int64>, 'buttermilk': <tf.Tensor 'ExpandDims_120:0' shape=(None, 1) dtype=int64>, 'butternut_squash': <tf.Tensor 'ExpandDims_121:0' shape=(None, 1) dtype=int64>, 'cabbage': <tf.Tensor 'ExpandDims_122:0' shape=(None, 1) dtype=int64>, 'cajun_seasoning': <tf.Tensor 'ExpandDims_123:0' shape=(None, 1) dtype=int64>, 'calvados': <tf.Tensor 'ExpandDims_124:0' shape=(None, 1) dtype=int64>, 'campbell': <tf.Tensor 'ExpandDims_125:0' shape=(None, 1) dtype=int64>, 'canadian_bacon': <tf.Tensor 'ExpandDims_126:0' shape=(None, 1) dtype=int64>, 'cannellini_bean': <tf.Tensor 'ExpandDims_127:0' shape=(None, 1) dtype=int64>, 'canola_oil': <tf.Tensor 'ExpandDims_128:0' shape=(None, 1) dtype=int64>, 'cantaloupe': <tf.Tensor 'ExpandDims_129:0' shape=(None, 1) dtype=int64>, 'caper': <tf.Tensor 'ExpandDims_130:0' shape=(None, 1) dtype=int64>, 'capsicum': <tf.Tensor 'ExpandDims_131:0' shape=(None, 1) dtype=int64>, 'caraway_seed': <tf.Tensor 'ExpandDims_132:0' shape=(None, 1) dtype=int64>, 'cardamom': <tf.Tensor 'ExpandDims_133:0' shape=(None, 1) dtype=int64>, 'carrot': <tf.Tensor 'ExpandDims_134:0' shape=(None, 1) dtype=int64>, 'cashew_nut': <tf.Tensor 'ExpandDims_135:0' shape=(None, 1) dtype=int64>, 'catfish': <tf.Tensor 'ExpandDims_136:0' shape=(None, 1) dtype=int64>, 'caul': <tf.Tensor 'ExpandDims_137:0' shape=(None, 1) dtype=int64>, 'cauliflower': <tf.Tensor 'ExpandDims_138:0' shape=(None, 1) dtype=int64>, 'cayenne_pepper': <tf.Tensor 'ExpandDims_139:0' shape=(None, 1) dtype=int64>, 'celeriac': <tf.Tensor 'ExpandDims_140:0' shape=(None, 1) dtype=int64>, 'celery': <tf.Tensor 'ExpandDims_141:0' shape=(None, 1) dtype=int64>, 'celery_salt': <tf.Tensor 'ExpandDims_142:0' shape=(None, 1) dtype=int64>, 'celery_seed': <tf.Tensor 'ExpandDims_143:0' shape=(None, 1) dtype=int64>, 'celery_soup': <tf.Tensor 'ExpandDims_144:0' shape=(None, 1) dtype=int64>, 'chanterelle_mushroom': <tf.Tensor 'ExpandDims_145:0' shape=(None, 1) dtype=int64>, 'chard': <tf.Tensor 'ExpandDims_146:0' shape=(None, 1) dtype=int64>, 'chayote': <tf.Tensor 'ExpandDims_147:0' shape=(None, 1) dtype=int64>, 'cheddar_cheese': <tf.Tensor 'ExpandDims_148:0' shape=(None, 1) dtype=int64>, 'cheddar_cheese_soup': <tf.Tensor 'ExpandDims_149:0' shape=(None, 1) dtype=int64>, 'cheese': <tf.Tensor 'ExpandDims_150:0' shape=(None, 1) dtype=int64>, 'cherry': <tf.Tensor 'ExpandDims_151:0' shape=(None, 1) dtype=int64>, 'chervil': <tf.Tensor 'ExpandDims_152:0' shape=(None, 1) dtype=int64>, 'cheshire_cheese': <tf.Tensor 'ExpandDims_153:0' shape=(None, 1) dtype=int64>, 'chestnut': <tf.Tensor 'ExpandDims_154:0' shape=(None, 1) dtype=int64>, 'chestnut_mushroom': <tf.Tensor 'ExpandDims_155:0' shape=(None, 1) dtype=int64>, 'chicken': <tf.Tensor 'ExpandDims_156:0' shape=(None, 1) dtype=int64>, 'chicken_base': <tf.Tensor 'ExpandDims_157:0' shape=(None, 1) dtype=int64>, 'chicken_liver': <tf.Tensor 'ExpandDims_158:0' shape=(None, 1) dtype=int64>, 'chicken_sausage': <tf.Tensor 'ExpandDims_159:0' shape=(None, 1) dtype=int64>, 'chicken_soup': <tf.Tensor 'ExpandDims_160:0' shape=(None, 1) dtype=int64>, 'chicken_stock': <tf.Tensor 'ExpandDims_161:0' shape=(None, 1) dtype=int64>, 'chickpea': <tf.Tensor 'ExpandDims_162:0' shape=(None, 1) dtype=int64>, 'chile_pepper': <tf.Tensor 'ExpandDims_163:0' shape=(None, 1) dtype=int64>, 'chili_oil': <tf.Tensor 'ExpandDims_164:0' shape=(None, 1) dtype=int64>, 'chili_paste': <tf.Tensor 'ExpandDims_165:0' shape=(None, 1) dtype=int64>, 'chili_powder': <tf.Tensor 'ExpandDims_166:0' shape=(None, 1) dtype=int64>, 'chili_sauce': <tf.Tensor 'ExpandDims_167:0' shape=(None, 1) dtype=int64>, 'chinese_cabbage': <tf.Tensor 'ExpandDims_168:0' shape=(None, 1) dtype=int64>, 'chinese_chives': <tf.Tensor 'ExpandDims_169:0' shape=(None, 1) dtype=int64>, 'Chinese_mushroom': <tf.Tensor 'ExpandDims:0' shape=(None, 1) dtype=int64>, 'Chinese_noodles': <tf.Tensor 'ExpandDims_1:0' shape=(None, 1) dtype=int64>, 'chipotle_pepper': <tf.Tensor 'ExpandDims_170:0' shape=(None, 1) dtype=int64>, 'chives': <tf.Tensor 'ExpandDims_171:0' shape=(None, 1) dtype=int64>, 'chocolate': <tf.Tensor 'ExpandDims_172:0' shape=(None, 1) dtype=int64>, 'chutney': <tf.Tensor 'ExpandDims_173:0' shape=(None, 1) dtype=int64>, 'cider': <tf.Tensor 'ExpandDims_174:0' shape=(None, 1) dtype=int64>, 'cider_vinegar': <tf.Tensor 'ExpandDims_175:0' shape=(None, 1) dtype=int64>, 'cilantro': <tf.Tensor 'ExpandDims_176:0' shape=(None, 1) dtype=int64>, 'cinnamon': <tf.Tensor 'ExpandDims_177:0' shape=(None, 1) dtype=int64>, 'clam': <tf.Tensor 'ExpandDims_178:0' shape=(None, 1) dtype=int64>, 'clam_juice': <tf.Tensor 'ExpandDims_179:0' shape=(None, 1) dtype=int64>, 'clarified_butter': <tf.Tensor 'ExpandDims_180:0' shape=(None, 1) dtype=int64>, 'clove': <tf.Tensor 'ExpandDims_181:0' shape=(None, 1) dtype=int64>, 'cocoa_powder': <tf.Tensor 'ExpandDims_182:0' shape=(None, 1) dtype=int64>, 'coconut': <tf.Tensor 'ExpandDims_183:0' shape=(None, 1) dtype=int64>, 'coconut_milk_or_cream': <tf.Tensor 'ExpandDims_184:0' shape=(None, 1) dtype=int64>, 'coconut_oil': <tf.Tensor 'ExpandDims_185:0' shape=(None, 1) dtype=int64>, 'cod': <tf.Tensor 'ExpandDims_186:0' shape=(None, 1) dtype=int64>, 'colby': <tf.Tensor 'ExpandDims_187:0' shape=(None, 1) dtype=int64>, 'coleslaw_mix': <tf.Tensor 'ExpandDims_188:0' shape=(None, 1) dtype=int64>, 'confit': <tf.Tensor 'ExpandDims_189:0' shape=(None, 1) dtype=int64>, 'coriander': <tf.Tensor 'ExpandDims_190:0' shape=(None, 1) dtype=int64>, 'coriander_root': <tf.Tensor 'ExpandDims_191:0' shape=(None, 1) dtype=int64>, 'corn': <tf.Tensor 'ExpandDims_192:0' shape=(None, 1) dtype=int64>, 'corn_chips': <tf.Tensor 'ExpandDims_193:0' shape=(None, 1) dtype=int64>, 'corn_flour': <tf.Tensor 'ExpandDims_194:0' shape=(None, 1) dtype=int64>, 'corn_oil': <tf.Tensor 'ExpandDims_195:0' shape=(None, 1) dtype=int64>, 'cornbread': <tf.Tensor 'ExpandDims_196:0' shape=(None, 1) dtype=int64>, 'corned_beef': <tf.Tensor 'ExpandDims_197:0' shape=(None, 1) dtype=int64>, 'cornish_hen': <tf.Tensor 'ExpandDims_198:0' shape=(None, 1) dtype=int64>, 'cornmeal': <tf.Tensor 'ExpandDims_199:0' shape=(None, 1) dtype=int64>, 'cornstarch': <tf.Tensor 'ExpandDims_200:0' shape=(None, 1) dtype=int64>, 'cottage_cheese': <tf.Tensor 'ExpandDims_201:0' shape=(None, 1) dtype=int64>, 'couscous': <tf.Tensor 'ExpandDims_202:0' shape=(None, 1) dtype=int64>, 'crab': <tf.Tensor 'ExpandDims_203:0' shape=(None, 1) dtype=int64>, 'crab_claw': <tf.Tensor 'ExpandDims_204:0' shape=(None, 1) dtype=int64>, 'cracker': <tf.Tensor 'ExpandDims_205:0' shape=(None, 1) dtype=int64>, 'craisin': <tf.Tensor 'ExpandDims_206:0' shape=(None, 1) dtype=int64>, 'cranberry': <tf.Tensor 'ExpandDims_207:0' shape=(None, 1) dtype=int64>, 'cranberry_bean': <tf.Tensor 'ExpandDims_208:0' shape=(None, 1) dtype=int64>, 'crayfish': <tf.Tensor 'ExpandDims_209:0' shape=(None, 1) dtype=int64>, 'cream': <tf.Tensor 'ExpandDims_210:0' shape=(None, 1) dtype=int64>, 'cream_cheese': <tf.Tensor 'ExpandDims_211:0' shape=(None, 1) dtype=int64>, 'creamed_corn': <tf.Tensor 'ExpandDims_212:0' shape=(None, 1) dtype=int64>, 'creme_fraiche': <tf.Tensor 'ExpandDims_213:0' shape=(None, 1) dtype=int64>, 'cremini_mushroom': <tf.Tensor 'ExpandDims_214:0' shape=(None, 1) dtype=int64>, 'crouton': <tf.Tensor 'ExpandDims_215:0' shape=(None, 1) dtype=int64>, 'crusty_bread': <tf.Tensor 'ExpandDims_216:0' shape=(None, 1) dtype=int64>, 'crystallized_ginger': <tf.Tensor 'ExpandDims_217:0' shape=(None, 1) dtype=int64>, 'cucumber': <tf.Tensor 'ExpandDims_218:0' shape=(None, 1) dtype=int64>, 'cumin': <tf.Tensor 'ExpandDims_219:0' shape=(None, 1) dtype=int64>, 'currant': <tf.Tensor 'ExpandDims_220:0' shape=(None, 1) dtype=int64>, 'curry_leaf': <tf.Tensor 'ExpandDims_221:0' shape=(None, 1) dtype=int64>, 'curry_paste': <tf.Tensor 'ExpandDims_222:0' shape=(None, 1) dtype=int64>, 'curry_powder': <tf.Tensor 'ExpandDims_223:0' shape=(None, 1) dtype=int64>, 'curry_sauce': <tf.Tensor 'ExpandDims_224:0' shape=(None, 1) dtype=int64>, 'dashi': <tf.Tensor 'ExpandDims_225:0' shape=(None, 1) dtype=int64>, 'dashi_stock': <tf.Tensor 'ExpandDims_226:0' shape=(None, 1) dtype=int64>, 'date': <tf.Tensor 'ExpandDims_227:0' shape=(None, 1) dtype=int64>, 'dehydrated_onion': <tf.Tensor 'ExpandDims_228:0' shape=(None, 1) dtype=int64>, 'demi_glace': <tf.Tensor 'ExpandDims_229:0' shape=(None, 1) dtype=int64>, 'dill': <tf.Tensor 'ExpandDims_230:0' shape=(None, 1) dtype=int64>, 'dried_apricot': <tf.Tensor 'ExpandDims_231:0' shape=(None, 1) dtype=int64>, 'dried_chile_pepper': <tf.Tensor 'ExpandDims_232:0' shape=(None, 1) dtype=int64>, 'dripping': <tf.Tensor 'ExpandDims_233:0' shape=(None, 1) dtype=int64>, 'duck': <tf.Tensor 'ExpandDims_234:0' shape=(None, 1) dtype=int64>, 'edamame': <tf.Tensor 'ExpandDims_235:0' shape=(None, 1) dtype=int64>, 'eel': <tf.Tensor 'ExpandDims_236:0' shape=(None, 1) dtype=int64>, 'egg': <tf.Tensor 'ExpandDims_237:0' shape=(None, 1) dtype=int64>, 'egg_noodles': <tf.Tensor 'ExpandDims_238:0' shape=(None, 1) dtype=int64>, 'egg_roll_wraps': <tf.Tensor 'ExpandDims_239:0' shape=(None, 1) dtype=int64>, 'egg_substitute': <tf.Tensor 'ExpandDims_240:0' shape=(None, 1) dtype=int64>, 'eggplant': <tf.Tensor 'ExpandDims_241:0' shape=(None, 1) dtype=int64>, 'emeril': <tf.Tensor 'ExpandDims_242:0' shape=(None, 1) dtype=int64>, 'emmenthaler_cheese': <tf.Tensor 'ExpandDims_243:0' shape=(None, 1) dtype=int64>, 'enchilada_sauce': <tf.Tensor 'ExpandDims_244:0' shape=(None, 1) dtype=int64>, 'endive': <tf.Tensor 'ExpandDims_245:0' shape=(None, 1) dtype=int64>, 'english_muffin': <tf.Tensor 'ExpandDims_246:0' shape=(None, 1) dtype=int64>, 'english_mustard': <tf.Tensor 'ExpandDims_247:0' shape=(None, 1) dtype=int64>, 'enoki_mushroom': <tf.Tensor 'ExpandDims_248:0' shape=(None, 1) dtype=int64>, 'epazote': <tf.Tensor 'ExpandDims_249:0' shape=(None, 1) dtype=int64>, 'evaporated_milk': <tf.Tensor 'ExpandDims_250:0' shape=(None, 1) dtype=int64>, 'fajita_seasoning_mix': <tf.Tensor 'ExpandDims_251:0' shape=(None, 1) dtype=int64>, 'fat-free_refried_beans': <tf.Tensor 'ExpandDims_252:0' shape=(None, 1) dtype=int64>, 'fava_bean': <tf.Tensor 'ExpandDims_253:0' shape=(None, 1) dtype=int64>, 'fennel': <tf.Tensor 'ExpandDims_254:0' shape=(None, 1) dtype=int64>, 'fennel_seeds': <tf.Tensor 'ExpandDims_255:0' shape=(None, 1) dtype=int64>, 'fenugreek': <tf.Tensor 'ExpandDims_256:0' shape=(None, 1) dtype=int64>, 'fermented_black_bean': <tf.Tensor 'ExpandDims_257:0' shape=(None, 1) dtype=int64>, 'feta_cheese': <tf.Tensor 'ExpandDims_258:0' shape=(None, 1) dtype=int64>, 'fig': <tf.Tensor 'ExpandDims_259:0' shape=(None, 1) dtype=int64>, 'filet_mignon': <tf.Tensor 'ExpandDims_260:0' shape=(None, 1) dtype=int64>, 'fire_oil': <tf.Tensor 'ExpandDims_261:0' shape=(None, 1) dtype=int64>, 'fish': <tf.Tensor 'ExpandDims_262:0' shape=(None, 1) dtype=int64>, 'fish_sauce': <tf.Tensor 'ExpandDims_263:0' shape=(None, 1) dtype=int64>, 'fish_stock': <tf.Tensor 'ExpandDims_264:0' shape=(None, 1) dtype=int64>, 'five_spice_powder': <tf.Tensor 'ExpandDims_265:0' shape=(None, 1) dtype=int64>, 'flat_bread': <tf.Tensor 'ExpandDims_266:0' shape=(None, 1) dtype=int64>, 'flax_seed': <tf.Tensor 'ExpandDims_267:0' shape=(None, 1) dtype=int64>, 'flounder': <tf.Tensor 'ExpandDims_268:0' shape=(None, 1) dtype=int64>, 'flour': <tf.Tensor 'ExpandDims_269:0' shape=(None, 1) dtype=int64>, 'fontina_cheese': <tf.Tensor 'ExpandDims_270:0' shape=(None, 1) dtype=int64>, 'French_beans': <tf.Tensor 'ExpandDims_2:0' shape=(None, 1) dtype=int64>, 'French_bread': <tf.Tensor 'ExpandDims_3:0' shape=(None, 1) dtype=int64>, 'French_dressing': <tf.Tensor 'ExpandDims_4:0' shape=(None, 1) dtype=int64>, 'french_style_green_bean': <tf.Tensor 'ExpandDims_271:0' shape=(None, 1) dtype=int64>, 'fresh_button_mushroom': <tf.Tensor 'ExpandDims_272:0' shape=(None, 1) dtype=int64>, 'frozen_edamame': <tf.Tensor 'ExpandDims_273:0' shape=(None, 1) dtype=int64>, 'fusilli': <tf.Tensor 'ExpandDims_274:0' shape=(None, 1) dtype=int64>, 'galangal': <tf.Tensor 'ExpandDims_275:0' shape=(None, 1) dtype=int64>, 'garlic': <tf.Tensor 'ExpandDims_276:0' shape=(None, 1) dtype=int64>, 'garlic_butter': <tf.Tensor 'ExpandDims_277:0' shape=(None, 1) dtype=int64>, 'garlic_pepper_seasoning': <tf.Tensor 'ExpandDims_278:0' shape=(None, 1) dtype=int64>, 'garlic_powder': <tf.Tensor 'ExpandDims_279:0' shape=(None, 1) dtype=int64>, 'garlic_salt': <tf.Tensor 'ExpandDims_280:0' shape=(None, 1) dtype=int64>, 'gelatin': <tf.Tensor 'ExpandDims_281:0' shape=(None, 1) dtype=int64>, 'genoa_salami': <tf.Tensor 'ExpandDims_282:0' shape=(None, 1) dtype=int64>, 'ghee': <tf.Tensor 'ExpandDims_283:0' shape=(None, 1) dtype=int64>, 'ginger': <tf.Tensor 'ExpandDims_284:0' shape=(None, 1) dtype=int64>, 'ginger-garlic_paste': <tf.Tensor 'ExpandDims_285:0' shape=(None, 1) dtype=int64>, 'gingersnap_cookie': <tf.Tensor 'ExpandDims_286:0' shape=(None, 1) dtype=int64>, 'gingersnaps': <tf.Tensor 'ExpandDims_287:0' shape=(None, 1) dtype=int64>, 'gloucester': <tf.Tensor 'ExpandDims_288:0' shape=(None, 1) dtype=int64>, 'gluten': <tf.Tensor 'ExpandDims_289:0' shape=(None, 1) dtype=int64>, 'goat_cheese': <tf.Tensor 'ExpandDims_290:0' shape=(None, 1) dtype=int64>, 'golden_syrup': <tf.Tensor 'ExpandDims_291:0' shape=(None, 1) dtype=int64>, 'goose': <tf.Tensor 'ExpandDims_292:0' shape=(None, 1) dtype=int64>, 'gorgonzola': <tf.Tensor 'ExpandDims_293:0' shape=(None, 1) dtype=int64>, 'gouda': <tf.Tensor 'ExpandDims_294:0' shape=(None, 1) dtype=int64>, 'grand_marnier': <tf.Tensor 'ExpandDims_295:0' shape=(None, 1) dtype=int64>, 'granulated_garlic': <tf.Tensor 'ExpandDims_296:0' shape=(None, 1) dtype=int64>, 'grape': <tf.Tensor 'ExpandDims_297:0' shape=(None, 1) dtype=int64>, 'grapefruit': <tf.Tensor 'ExpandDims_298:0' shape=(None, 1) dtype=int64>, 'grapeseed_oil': <tf.Tensor 'ExpandDims_299:0' shape=(None, 1) dtype=int64>, 'grating_cheese': <tf.Tensor 'ExpandDims_300:0' shape=(None, 1) dtype=int64>, 'gravy': <tf.Tensor 'ExpandDims_301:0' shape=(None, 1) dtype=int64>, 'great_northern_bean': <tf.Tensor 'ExpandDims_302:0' shape=(None, 1) dtype=int64>, 'Greek_salad_dressing': <tf.Tensor 'ExpandDims_5:0' shape=(None, 1) dtype=int64>, 'green_bean': <tf.Tensor 'ExpandDims_303:0' shape=(None, 1) dtype=int64>, 'green_olive': <tf.Tensor 'ExpandDims_304:0' shape=(None, 1) dtype=int64>, 'green_onion': <tf.Tensor 'ExpandDims_305:0' shape=(None, 1) dtype=int64>, 'greens': <tf.Tensor 'ExpandDims_306:0' shape=(None, 1) dtype=int64>, 'ground_meat': <tf.Tensor 'ExpandDims_307:0' shape=(None, 1) dtype=int64>, 'ground_round': <tf.Tensor 'ExpandDims_308:0' shape=(None, 1) dtype=int64>, 'gruyere': <tf.Tensor 'ExpandDims_309:0' shape=(None, 1) dtype=int64>, 'guacamole': <tf.Tensor 'ExpandDims_310:0' shape=(None, 1) dtype=int64>, 'Guinness_stout': <tf.Tensor 'ExpandDims_6:0' shape=(None, 1) dtype=int64>, 'gumbo': <tf.Tensor 'ExpandDims_311:0' shape=(None, 1) dtype=int64>, 'habanero_chile': <tf.Tensor 'ExpandDims_312:0' shape=(None, 1) dtype=int64>, 'haddock': <tf.Tensor 'ExpandDims_313:0' shape=(None, 1) dtype=int64>, 'hake_fillets': <tf.Tensor 'ExpandDims_314:0' shape=(None, 1) dtype=int64>, 'half_and_half': <tf.Tensor 'ExpandDims_315:0' shape=(None, 1) dtype=int64>, 'halibut': <tf.Tensor 'ExpandDims_316:0' shape=(None, 1) dtype=int64>, 'ham': <tf.Tensor 'ExpandDims_317:0' shape=(None, 1) dtype=int64>, 'ham_hock': <tf.Tensor 'ExpandDims_318:0' shape=(None, 1) dtype=int64>, 'hamburger': <tf.Tensor 'ExpandDims_319:0' shape=(None, 1) dtype=int64>, 'harissa': <tf.Tensor 'ExpandDims_320:0' shape=(None, 1) dtype=int64>, 'hash_browns': <tf.Tensor 'ExpandDims_321:0' shape=(None, 1) dtype=int64>, 'havarti': <tf.Tensor 'ExpandDims_322:0' shape=(None, 1) dtype=int64>, 'hazelnut': <tf.Tensor 'ExpandDims_323:0' shape=(None, 1) dtype=int64>, 'heavy_cream': <tf.Tensor 'ExpandDims_324:0' shape=(None, 1) dtype=int64>, 'herb': <tf.Tensor 'ExpandDims_325:0' shape=(None, 1) dtype=int64>, 'herbes_de_provence': <tf.Tensor 'ExpandDims_326:0' shape=(None, 1) dtype=int64>, 'hoagie_rolls': <tf.Tensor 'ExpandDims_327:0' shape=(None, 1) dtype=int64>, 'hoisin_sauce': <tf.Tensor 'ExpandDims_328:0' shape=(None, 1) dtype=int64>, 'hokkien_noodles': <tf.Tensor 'ExpandDims_329:0' shape=(None, 1) dtype=int64>, 'hominy': <tf.Tensor 'ExpandDims_330:0' shape=(None, 1) dtype=int64>, 'honey': <tf.Tensor 'ExpandDims_331:0' shape=(None, 1) dtype=int64>, 'honey_mustard': <tf.Tensor 'ExpandDims_332:0' shape=(None, 1) dtype=int64>, 'horseradish': <tf.Tensor 'ExpandDims_333:0' shape=(None, 1) dtype=int64>, 'horseradish_sauce': <tf.Tensor 'ExpandDims_334:0' shape=(None, 1) dtype=int64>, 'hot_bean_paste': <tf.Tensor 'ExpandDims_335:0' shape=(None, 1) dtype=int64>, 'hot_pepper_sauce': <tf.Tensor 'ExpandDims_336:0' shape=(None, 1) dtype=int64>, 'hummus': <tf.Tensor 'ExpandDims_337:0' shape=(None, 1) dtype=int64>, 'ice_water': <tf.Tensor 'ExpandDims_338:0' shape=(None, 1) dtype=int64>, 'Italian_bread': <tf.Tensor 'ExpandDims_7:0' shape=(None, 1) dtype=int64>, 'Italian_breadcrumbs': <tf.Tensor 'ExpandDims_8:0' shape=(None, 1) dtype=int64>, 'Italian_rolls': <tf.Tensor 'ExpandDims_9:0' shape=(None, 1) dtype=int64>, 'Italian_salad_dressing': <tf.Tensor 'ExpandDims_10:0' shape=(None, 1) dtype=int64>, 'Italian_spices': <tf.Tensor 'ExpandDims_11:0' shape=(None, 1) dtype=int64>, 'jack_cheese': <tf.Tensor 'ExpandDims_339:0' shape=(None, 1) dtype=int64>, 'jaggery': <tf.Tensor 'ExpandDims_340:0' shape=(None, 1) dtype=int64>, 'jalapeno_pepper': <tf.Tensor 'ExpandDims_341:0' shape=(None, 1) dtype=int64>, 'Japanese_breadcrumbs': <tf.Tensor 'ExpandDims_12:0' shape=(None, 1) dtype=int64>, 'jasmine_rice': <tf.Tensor 'ExpandDims_342:0' shape=(None, 1) dtype=int64>, 'jicama': <tf.Tensor 'ExpandDims_343:0' shape=(None, 1) dtype=int64>, 'juniper_berries': <tf.Tensor 'ExpandDims_344:0' shape=(None, 1) dtype=int64>, 'kaffir_lime': <tf.Tensor 'ExpandDims_345:0' shape=(None, 1) dtype=int64>, 'kalamata_olive': <tf.Tensor 'ExpandDims_346:0' shape=(None, 1) dtype=int64>, 'kale': <tf.Tensor 'ExpandDims_347:0' shape=(None, 1) dtype=int64>, 'kamaboko': <tf.Tensor 'ExpandDims_348:0' shape=(None, 1) dtype=int64>, 'ketchup': <tf.Tensor 'ExpandDims_349:0' shape=(None, 1) dtype=int64>, 'ketjap_manis': <tf.Tensor 'ExpandDims_350:0' shape=(None, 1) dtype=int64>, 'kidney_bean': <tf.Tensor 'ExpandDims_351:0' shape=(None, 1) dtype=int64>, 'kielbasa': <tf.Tensor 'ExpandDims_352:0' shape=(None, 1) dtype=int64>, 'kohlrabi': <tf.Tensor 'ExpandDims_353:0' shape=(None, 1) dtype=int64>, 'kumquat': <tf.Tensor 'ExpandDims_354:0' shape=(None, 1) dtype=int64>, 'lamb': <tf.Tensor 'ExpandDims_355:0' shape=(None, 1) dtype=int64>, 'lambs_kidney': <tf.Tensor 'ExpandDims_356:0' shape=(None, 1) dtype=int64>, 'lancashire_cheese': <tf.Tensor 'ExpandDims_357:0' shape=(None, 1) dtype=int64>, 'lard': <tf.Tensor 'ExpandDims_358:0' shape=(None, 1) dtype=int64>, 'lardon': <tf.Tensor 'ExpandDims_359:0' shape=(None, 1) dtype=int64>, 'lasagna_noodles': <tf.Tensor 'ExpandDims_360:0' shape=(None, 1) dtype=int64>, 'leek': <tf.Tensor 'ExpandDims_361:0' shape=(None, 1) dtype=int64>, 'lemon': <tf.Tensor 'ExpandDims_362:0' shape=(None, 1) dtype=int64>, 'lemon_grass': <tf.Tensor 'ExpandDims_363:0' shape=(None, 1) dtype=int64>, 'lemon_juice': <tf.Tensor 'ExpandDims_364:0' shape=(None, 1) dtype=int64>, 'lentil': <tf.Tensor 'ExpandDims_365:0' shape=(None, 1) dtype=int64>, 'lettuce': <tf.Tensor 'ExpandDims_366:0' shape=(None, 1) dtype=int64>, 'lima_bean': <tf.Tensor 'ExpandDims_367:0' shape=(None, 1) dtype=int64>, 'lime': <tf.Tensor 'ExpandDims_368:0' shape=(None, 1) dtype=int64>, 'liver': <tf.Tensor 'ExpandDims_369:0' shape=(None, 1) dtype=int64>, 'lobster': <tf.Tensor 'ExpandDims_370:0' shape=(None, 1) dtype=int64>, 'long_grain_and_wild_rice_blend': <tf.Tensor 'ExpandDims_371:0' shape=(None, 1) dtype=int64>, 'lotus': <tf.Tensor 'ExpandDims_372:0' shape=(None, 1) dtype=int64>, 'macaroni': <tf.Tensor 'ExpandDims_373:0' shape=(None, 1) dtype=int64>, 'mace': <tf.Tensor 'ExpandDims_374:0' shape=(None, 1) dtype=int64>, 'mackerel': <tf.Tensor 'ExpandDims_375:0' shape=(None, 1) dtype=int64>, 'madeira': <tf.Tensor 'ExpandDims_376:0' shape=(None, 1) dtype=int64>, 'mahi_mahi': <tf.Tensor 'ExpandDims_377:0' shape=(None, 1) dtype=int64>, 'malt_vinegar': <tf.Tensor 'ExpandDims_378:0' shape=(None, 1) dtype=int64>, 'mandarin_orange': <tf.Tensor 'ExpandDims_379:0' shape=(None, 1) dtype=int64>, 'mange_tout': <tf.Tensor 'ExpandDims_380:0' shape=(None, 1) dtype=int64>, 'mango': <tf.Tensor 'ExpandDims_381:0' shape=(None, 1) dtype=int64>, 'manicotti': <tf.Tensor 'ExpandDims_382:0' shape=(None, 1) dtype=int64>, 'manzanilla_olive': <tf.Tensor 'ExpandDims_383:0' shape=(None, 1) dtype=int64>, 'maple_syrup': <tf.Tensor 'ExpandDims_384:0' shape=(None, 1) dtype=int64>, 'margarine': <tf.Tensor 'ExpandDims_385:0' shape=(None, 1) dtype=int64>, 'marinara_sauce': <tf.Tensor 'ExpandDims_386:0' shape=(None, 1) dtype=int64>, 'marjoram': <tf.Tensor 'ExpandDims_387:0' shape=(None, 1) dtype=int64>, 'marmalade': <tf.Tensor 'ExpandDims_388:0' shape=(None, 1) dtype=int64>, 'marmite': <tf.Tensor 'ExpandDims_389:0' shape=(None, 1) dtype=int64>, 'marsala': <tf.Tensor 'ExpandDims_390:0' shape=(None, 1) dtype=int64>, 'masa_harina': <tf.Tensor 'ExpandDims_391:0' shape=(None, 1) dtype=int64>, 'masala': <tf.Tensor 'ExpandDims_392:0' shape=(None, 1) dtype=int64>, 'mascarpone': <tf.Tensor 'ExpandDims_393:0' shape=(None, 1) dtype=int64>, 'masoor_dal': <tf.Tensor 'ExpandDims_394:0' shape=(None, 1) dtype=int64>, 'mayonnaise': <tf.Tensor 'ExpandDims_395:0' shape=(None, 1) dtype=int64>, 'meat_stock': <tf.Tensor 'ExpandDims_396:0' shape=(None, 1) dtype=int64>, 'meatloaf_mix': <tf.Tensor 'ExpandDims_397:0' shape=(None, 1) dtype=int64>, 'Mexican_blend_cheese': <tf.Tensor 'ExpandDims_13:0' shape=(None, 1) dtype=int64>, 'mexicorn': <tf.Tensor 'ExpandDims_398:0' shape=(None, 1) dtype=int64>, 'milk': <tf.Tensor 'ExpandDims_399:0' shape=(None, 1) dtype=int64>, 'mint': <tf.Tensor 'ExpandDims_400:0' shape=(None, 1) dtype=int64>, 'miracle_whip_dressing': <tf.Tensor 'ExpandDims_401:0' shape=(None, 1) dtype=int64>, 'miso': <tf.Tensor 'ExpandDims_402:0' shape=(None, 1) dtype=int64>, 'mixed_spice': <tf.Tensor 'ExpandDims_403:0' shape=(None, 1) dtype=int64>, 'mizithra_cheese': <tf.Tensor 'ExpandDims_404:0' shape=(None, 1) dtype=int64>, 'molasses': <tf.Tensor 'ExpandDims_405:0' shape=(None, 1) dtype=int64>, 'monosodium_glutamate': <tf.Tensor 'ExpandDims_406:0' shape=(None, 1) dtype=int64>, 'moroccan_seasoning': <tf.Tensor 'ExpandDims_407:0' shape=(None, 1) dtype=int64>, 'mortadella': <tf.Tensor 'ExpandDims_408:0' shape=(None, 1) dtype=int64>, 'mozzarella': <tf.Tensor 'ExpandDims_409:0' shape=(None, 1) dtype=int64>, 'mung_beans': <tf.Tensor 'ExpandDims_410:0' shape=(None, 1) dtype=int64>, 'muscovado_sugar': <tf.Tensor 'ExpandDims_411:0' shape=(None, 1) dtype=int64>, 'mushroom': <tf.Tensor 'ExpandDims_412:0' shape=(None, 1) dtype=int64>, 'mushroom_soup': <tf.Tensor 'ExpandDims_413:0' shape=(None, 1) dtype=int64>, 'mussel': <tf.Tensor 'ExpandDims_414:0' shape=(None, 1) dtype=int64>, 'mustard_oil': <tf.Tensor 'ExpandDims_415:0' shape=(None, 1) dtype=int64>, 'mustard_prepared': <tf.Tensor 'ExpandDims_416:0' shape=(None, 1) dtype=int64>, 'mustard_seed_and_powder': <tf.Tensor 'ExpandDims_417:0' shape=(None, 1) dtype=int64>, 'navy_bean': <tf.Tensor 'ExpandDims_418:0' shape=(None, 1) dtype=int64>, 'nectarine': <tf.Tensor 'ExpandDims_419:0' shape=(None, 1) dtype=int64>, 'nigella_seeds': <tf.Tensor 'ExpandDims_420:0' shape=(None, 1) dtype=int64>, 'nori': <tf.Tensor 'ExpandDims_421:0' shape=(None, 1) dtype=int64>, 'nutmeg': <tf.Tensor 'ExpandDims_422:0' shape=(None, 1) dtype=int64>, 'oatmeal': <tf.Tensor 'ExpandDims_423:0' shape=(None, 1) dtype=int64>, 'okra': <tf.Tensor 'ExpandDims_424:0' shape=(None, 1) dtype=int64>, 'Old_Bay_Seasoning': <tf.Tensor 'ExpandDims_14:0' shape=(None, 1) dtype=int64>, 'olive': <tf.Tensor 'ExpandDims_425:0' shape=(None, 1) dtype=int64>, 'olive_oil': <tf.Tensor 'ExpandDims_426:0' shape=(None, 1) dtype=int64>, 'onion': <tf.Tensor 'ExpandDims_427:0' shape=(None, 1) dtype=int64>, 'onion_powder': <tf.Tensor 'ExpandDims_428:0' shape=(None, 1) dtype=int64>, 'onion_soup': <tf.Tensor 'ExpandDims_429:0' shape=(None, 1) dtype=int64>, 'orange': <tf.Tensor 'ExpandDims_430:0' shape=(None, 1) dtype=int64>, 'orange_juice': <tf.Tensor 'ExpandDims_431:0' shape=(None, 1) dtype=int64>, 'orange_rind': <tf.Tensor 'ExpandDims_432:0' shape=(None, 1) dtype=int64>, 'oregano': <tf.Tensor 'ExpandDims_433:0' shape=(None, 1) dtype=int64>, 'oxtails': <tf.Tensor 'ExpandDims_434:0' shape=(None, 1) dtype=int64>, 'oyster_sauce': <tf.Tensor 'ExpandDims_435:0' shape=(None, 1) dtype=int64>, 'oysters': <tf.Tensor 'ExpandDims_436:0' shape=(None, 1) dtype=int64>, 'Pace_Picante_Sauce': <tf.Tensor 'ExpandDims_15:0' shape=(None, 1) dtype=int64>, 'palm_sugar': <tf.Tensor 'ExpandDims_437:0' shape=(None, 1) dtype=int64>, 'pancetta': <tf.Tensor 'ExpandDims_438:0' shape=(None, 1) dtype=int64>, 'panir': <tf.Tensor 'ExpandDims_439:0' shape=(None, 1) dtype=int64>, 'panko': <tf.Tensor 'ExpandDims_440:0' shape=(None, 1) dtype=int64>, 'papaya': <tf.Tensor 'ExpandDims_441:0' shape=(None, 1) dtype=int64>, 'paprika': <tf.Tensor 'ExpandDims_442:0' shape=(None, 1) dtype=int64>, 'parchment_paper': <tf.Tensor 'ExpandDims_443:0' shape=(None, 1) dtype=int64>, 'parmesan_cheese': <tf.Tensor 'ExpandDims_444:0' shape=(None, 1) dtype=int64>, 'parsley': <tf.Tensor 'ExpandDims_445:0' shape=(None, 1) dtype=int64>, 'parsnip': <tf.Tensor 'ExpandDims_446:0' shape=(None, 1) dtype=int64>, 'passata': <tf.Tensor 'ExpandDims_447:0' shape=(None, 1) dtype=int64>, 'pasta': <tf.Tensor 'ExpandDims_448:0' shape=(None, 1) dtype=int64>, 'pasta_sauce': <tf.Tensor 'ExpandDims_449:0' shape=(None, 1) dtype=int64>, 'pastis': <tf.Tensor 'ExpandDims_450:0' shape=(None, 1) dtype=int64>, 'pastry_dough': <tf.Tensor 'ExpandDims_451:0' shape=(None, 1) dtype=int64>, 'pea': <tf.Tensor 'ExpandDims_452:0' shape=(None, 1) dtype=int64>, 'peanut': <tf.Tensor 'ExpandDims_453:0' shape=(None, 1) dtype=int64>, 'peanut_butter': <tf.Tensor 'ExpandDims_454:0' shape=(None, 1) dtype=int64>, 'peanut_oil': <tf.Tensor 'ExpandDims_455:0' shape=(None, 1) dtype=int64>, 'peanut_sauce': <tf.Tensor 'ExpandDims_456:0' shape=(None, 1) dtype=int64>, 'pear': <tf.Tensor 'ExpandDims_457:0' shape=(None, 1) dtype=int64>, 'pearl_onion': <tf.Tensor 'ExpandDims_458:0' shape=(None, 1) dtype=int64>, 'pecan': <tf.Tensor 'ExpandDims_459:0' shape=(None, 1) dtype=int64>, 'pecorino_cheese': <tf.Tensor 'ExpandDims_460:0' shape=(None, 1) dtype=int64>, 'penne_rigate': <tf.Tensor 'ExpandDims_461:0' shape=(None, 1) dtype=int64>, 'pepper': <tf.Tensor 'ExpandDims_462:0' shape=(None, 1) dtype=int64>, 'pepperoncini_pepper': <tf.Tensor 'ExpandDims_463:0' shape=(None, 1) dtype=int64>, 'pepperoni': <tf.Tensor 'ExpandDims_464:0' shape=(None, 1) dtype=int64>, 'pesto': <tf.Tensor 'ExpandDims_465:0' shape=(None, 1) dtype=int64>, 'phyllo_dough': <tf.Tensor 'ExpandDims_466:0' shape=(None, 1) dtype=int64>, 'picante_sauce': <tf.Tensor 'ExpandDims_467:0' shape=(None, 1) dtype=int64>, 'pickle': <tf.Tensor 'ExpandDims_468:0' shape=(None, 1) dtype=int64>, 'pickling_spices': <tf.Tensor 'ExpandDims_469:0' shape=(None, 1) dtype=int64>, 'pie_crust': <tf.Tensor 'ExpandDims_470:0' shape=(None, 1) dtype=int64>, 'pierogi': <tf.Tensor 'ExpandDims_471:0' shape=(None, 1) dtype=int64>, 'pimiento': <tf.Tensor 'ExpandDims_472:0' shape=(None, 1) dtype=int64>, 'pine_nuts': <tf.Tensor 'ExpandDims_473:0' shape=(None, 1) dtype=int64>, 'pineapple': <tf.Tensor 'ExpandDims_474:0' shape=(None, 1) dtype=int64>, 'pink_bean': <tf.Tensor 'ExpandDims_475:0' shape=(None, 1) dtype=int64>, 'pinto_bean': <tf.Tensor 'ExpandDims_476:0' shape=(None, 1) dtype=int64>, 'piquillo_pepper': <tf.Tensor 'ExpandDims_477:0' shape=(None, 1) dtype=int64>, 'pistachio': <tf.Tensor 'ExpandDims_478:0' shape=(None, 1) dtype=int64>, 'pita_bread': <tf.Tensor 'ExpandDims_479:0' shape=(None, 1) dtype=int64>, 'pitted_prunes': <tf.Tensor 'ExpandDims_480:0' shape=(None, 1) dtype=int64>, 'pizza_crust': <tf.Tensor 'ExpandDims_481:0' shape=(None, 1) dtype=int64>, 'pizza_dough': <tf.Tensor 'ExpandDims_482:0' shape=(None, 1) dtype=int64>, 'pizza_sauce': <tf.Tensor 'ExpandDims_483:0' shape=(None, 1) dtype=int64>, 'plum': <tf.Tensor 'ExpandDims_484:0' shape=(None, 1) dtype=int64>, 'plum_tomato': <tf.Tensor 'ExpandDims_485:0' shape=(None, 1) dtype=int64>, 'poblano_pepper': <tf.Tensor 'ExpandDims_486:0' shape=(None, 1) dtype=int64>, 'polenta': <tf.Tensor 'ExpandDims_487:0' shape=(None, 1) dtype=int64>, 'pomegranate': <tf.Tensor 'ExpandDims_488:0' shape=(None, 1) dtype=int64>, 'ponzu': <tf.Tensor 'ExpandDims_489:0' shape=(None, 1) dtype=int64>, 'poppy_seed': <tf.Tensor 'ExpandDims_490:0' shape=(None, 1) dtype=int64>, 'porcino': <tf.Tensor 'ExpandDims_491:0' shape=(None, 1) dtype=int64>, 'pork': <tf.Tensor 'ExpandDims_492:0' shape=(None, 1) dtype=int64>, 'portabella': <tf.Tensor 'ExpandDims_493:0' shape=(None, 1) dtype=int64>, 'potato': <tf.Tensor 'ExpandDims_494:0' shape=(None, 1) dtype=int64>, 'potato_starch': <tf.Tensor 'ExpandDims_495:0' shape=(None, 1) dtype=int64>, 'poultry_seasoning': <tf.Tensor 'ExpandDims_496:0' shape=(None, 1) dtype=int64>, 'powdered_sugar': <tf.Tensor 'ExpandDims_497:0' shape=(None, 1) dtype=int64>, 'prawn': <tf.Tensor 'ExpandDims_498:0' shape=(None, 1) dtype=int64>, 'prepared_horseradish': <tf.Tensor 'ExpandDims_499:0' shape=(None, 1) dtype=int64>, 'preserved_lemon': <tf.Tensor 'ExpandDims_500:0' shape=(None, 1) dtype=int64>, 'processed_cheese': <tf.Tensor 'ExpandDims_501:0' shape=(None, 1) dtype=int64>, 'prosciutto': <tf.Tensor 'ExpandDims_502:0' shape=(None, 1) dtype=int64>, 'provolone': <tf.Tensor 'ExpandDims_503:0' shape=(None, 1) dtype=int64>, 'prune': <tf.Tensor 'ExpandDims_504:0' shape=(None, 1) dtype=int64>, 'puff_pastry': <tf.Tensor 'ExpandDims_505:0' shape=(None, 1) dtype=int64>, 'pumpkin': <tf.Tensor 'ExpandDims_506:0' shape=(None, 1) dtype=int64>, 'pumpkin_pie_spice': <tf.Tensor 'ExpandDims_507:0' shape=(None, 1) dtype=int64>, 'quark': <tf.Tensor 'ExpandDims_508:0' shape=(None, 1) dtype=int64>, 'queso': <tf.Tensor 'ExpandDims_509:0' shape=(None, 1) dtype=int64>, 'quinoa': <tf.Tensor 'ExpandDims_510:0' shape=(None, 1) dtype=int64>, 'rabbit': <tf.Tensor 'ExpandDims_511:0' shape=(None, 1) dtype=int64>, 'radicchio': <tf.Tensor 'ExpandDims_512:0' shape=(None, 1) dtype=int64>, 'radish': <tf.Tensor 'ExpandDims_513:0' shape=(None, 1) dtype=int64>, 'raisin': <tf.Tensor 'ExpandDims_514:0' shape=(None, 1) dtype=int64>, 'ramen_noodles': <tf.Tensor 'ExpandDims_515:0' shape=(None, 1) dtype=int64>, 'Ranch_salad_dressing': <tf.Tensor 'ExpandDims_16:0' shape=(None, 1) dtype=int64>, 'rapeseed_oil': <tf.Tensor 'ExpandDims_516:0' shape=(None, 1) dtype=int64>, 'ras_el_hanout': <tf.Tensor 'ExpandDims_517:0' shape=(None, 1) dtype=int64>, 'raw_sugar': <tf.Tensor 'ExpandDims_518:0' shape=(None, 1) dtype=int64>, 'red_beans': <tf.Tensor 'ExpandDims_519:0' shape=(None, 1) dtype=int64>, 'red_chile_powder': <tf.Tensor 'ExpandDims_520:0' shape=(None, 1) dtype=int64>, 'red_food_coloring': <tf.Tensor 'ExpandDims_521:0' shape=(None, 1) dtype=int64>, 'red_pepper_flakes': <tf.Tensor 'ExpandDims_522:0' shape=(None, 1) dtype=int64>, 'red_pepper_sauce': <tf.Tensor 'ExpandDims_523:0' shape=(None, 1) dtype=int64>, 'red_snapper': <tf.Tensor 'ExpandDims_524:0' shape=(None, 1) dtype=int64>, 'refried_beans': <tf.Tensor 'ExpandDims_525:0' shape=(None, 1) dtype=int64>, 'rhubarb': <tf.Tensor 'ExpandDims_526:0' shape=(None, 1) dtype=int64>, 'rice': <tf.Tensor 'ExpandDims_527:0' shape=(None, 1) dtype=int64>, 'rice_flour': <tf.Tensor 'ExpandDims_528:0' shape=(None, 1) dtype=int64>, 'rice_noodles': <tf.Tensor 'ExpandDims_529:0' shape=(None, 1) dtype=int64>, 'rice_vermicelli': <tf.Tensor 'ExpandDims_530:0' shape=(None, 1) dtype=int64>, 'rice_vinegar': <tf.Tensor 'ExpandDims_531:0' shape=(None, 1) dtype=int64>, 'rice_wine': <tf.Tensor 'ExpandDims_532:0' shape=(None, 1) dtype=int64>, 'ricotta_cheese': <tf.Tensor 'ExpandDims_533:0' shape=(None, 1) dtype=int64>, 'roll': <tf.Tensor 'ExpandDims_534:0' shape=(None, 1) dtype=int64>, 'romaine_lettuce': <tf.Tensor 'ExpandDims_535:0' shape=(None, 1) dtype=int64>, 'romano_cheese': <tf.Tensor 'ExpandDims_536:0' shape=(None, 1) dtype=int64>, 'rose_water': <tf.Tensor 'ExpandDims_537:0' shape=(None, 1) dtype=int64>, 'rosemary': <tf.Tensor 'ExpandDims_538:0' shape=(None, 1) dtype=int64>, 'rotel_tomatoes': <tf.Tensor 'ExpandDims_539:0' shape=(None, 1) dtype=int64>, 'rum': <tf.Tensor 'ExpandDims_540:0' shape=(None, 1) dtype=int64>, 'rutabaga': <tf.Tensor 'ExpandDims_541:0' shape=(None, 1) dtype=int64>, 'rye_bread': <tf.Tensor 'ExpandDims_542:0' shape=(None, 1) dtype=int64>, 'saffron': <tf.Tensor 'ExpandDims_543:0' shape=(None, 1) dtype=int64>, 'sage': <tf.Tensor 'ExpandDims_544:0' shape=(None, 1) dtype=int64>, 'salmon': <tf.Tensor 'ExpandDims_545:0' shape=(None, 1) dtype=int64>, 'salsa': <tf.Tensor 'ExpandDims_546:0' shape=(None, 1) dtype=int64>, 'salsify': <tf.Tensor 'ExpandDims_547:0' shape=(None, 1) dtype=int64>, 'salt': <tf.Tensor 'ExpandDims_548:0' shape=(None, 1) dtype=int64>, 'salt_pork': <tf.Tensor 'ExpandDims_549:0' shape=(None, 1) dtype=int64>, 'sambal': <tf.Tensor 'ExpandDims_550:0' shape=(None, 1) dtype=int64>, 'sardine': <tf.Tensor 'ExpandDims_551:0' shape=(None, 1) dtype=int64>, 'satay': <tf.Tensor 'ExpandDims_552:0' shape=(None, 1) dtype=int64>, 'sauerkraut': <tf.Tensor 'ExpandDims_553:0' shape=(None, 1) dtype=int64>, 'sausage': <tf.Tensor 'ExpandDims_554:0' shape=(None, 1) dtype=int64>, 'savory': <tf.Tensor 'ExpandDims_555:0' shape=(None, 1) dtype=int64>, 'sazon': <tf.Tensor 'ExpandDims_556:0' shape=(None, 1) dtype=int64>, 'sazon_goya': <tf.Tensor 'ExpandDims_557:0' shape=(None, 1) dtype=int64>, 'scallop': <tf.Tensor 'ExpandDims_558:0' shape=(None, 1) dtype=int64>, 'screwpine_leaves': <tf.Tensor 'ExpandDims_559:0' shape=(None, 1) dtype=int64>, 'sea_cucumber': <tf.Tensor 'ExpandDims_560:0' shape=(None, 1) dtype=int64>, 'seafood': <tf.Tensor 'ExpandDims_561:0' shape=(None, 1) dtype=int64>, 'seitan': <tf.Tensor 'ExpandDims_562:0' shape=(None, 1) dtype=int64>, 'semolina': <tf.Tensor 'ExpandDims_563:0' shape=(None, 1) dtype=int64>, 'serrano_pepper': <tf.Tensor 'ExpandDims_564:0' shape=(None, 1) dtype=int64>, 'sesame_oil': <tf.Tensor 'ExpandDims_565:0' shape=(None, 1) dtype=int64>, 'sesame_seed': <tf.Tensor 'ExpandDims_566:0' shape=(None, 1) dtype=int64>, 'shallot': <tf.Tensor 'ExpandDims_567:0' shape=(None, 1) dtype=int64>, 'sherry': <tf.Tensor 'ExpandDims_568:0' shape=(None, 1) dtype=int64>, 'shimeji_mushroom': <tf.Tensor 'ExpandDims_569:0' shape=(None, 1) dtype=int64>, 'shirataki_noodles': <tf.Tensor 'ExpandDims_570:0' shape=(None, 1) dtype=int64>, 'shitake': <tf.Tensor 'ExpandDims_571:0' shape=(None, 1) dtype=int64>, 'shortcrust_pastry': <tf.Tensor 'ExpandDims_572:0' shape=(None, 1) dtype=int64>, 'shortening': <tf.Tensor 'ExpandDims_573:0' shape=(None, 1) dtype=int64>, 'shoyu': <tf.Tensor 'ExpandDims_574:0' shape=(None, 1) dtype=int64>, 'shrimp': <tf.Tensor 'ExpandDims_575:0' shape=(None, 1) dtype=int64>, 'shrimp_paste': <tf.Tensor 'ExpandDims_576:0' shape=(None, 1) dtype=int64>, 'sliced_pepperoni': <tf.Tensor 'ExpandDims_577:0' shape=(None, 1) dtype=int64>, 'smen': <tf.Tensor 'ExpandDims_578:0' shape=(None, 1) dtype=int64>, 'smoke_seasoning': <tf.Tensor 'ExpandDims_579:0' shape=(None, 1) dtype=int64>, 'smoked_salmon': <tf.Tensor 'ExpandDims_580:0' shape=(None, 1) dtype=int64>, 'snap_pea': <tf.Tensor 'ExpandDims_581:0' shape=(None, 1) dtype=int64>, 'snow_pea': <tf.Tensor 'ExpandDims_582:0' shape=(None, 1) dtype=int64>, 'soba_noodles': <tf.Tensor 'ExpandDims_583:0' shape=(None, 1) dtype=int64>, 'sofrito_sauce': <tf.Tensor 'ExpandDims_584:0' shape=(None, 1) dtype=int64>, 'sole': <tf.Tensor 'ExpandDims_585:0' shape=(None, 1) dtype=int64>, 'sorrel': <tf.Tensor 'ExpandDims_586:0' shape=(None, 1) dtype=int64>, 'sour_cream': <tf.Tensor 'ExpandDims_587:0' shape=(None, 1) dtype=int64>, 'sourdough_bread': <tf.Tensor 'ExpandDims_588:0' shape=(None, 1) dtype=int64>, 'soy_sauce': <tf.Tensor 'ExpandDims_589:0' shape=(None, 1) dtype=int64>, 'soybean': <tf.Tensor 'ExpandDims_590:0' shape=(None, 1) dtype=int64>, 'soymilk': <tf.Tensor 'ExpandDims_591:0' shape=(None, 1) dtype=int64>, 'spaetzle': <tf.Tensor 'ExpandDims_592:0' shape=(None, 1) dtype=int64>, 'spaghetti_squash': <tf.Tensor 'ExpandDims_593:0' shape=(None, 1) dtype=int64>, 'spanish_cheese': <tf.Tensor 'ExpandDims_594:0' shape=(None, 1) dtype=int64>, 'spanish_chorizo': <tf.Tensor 'ExpandDims_595:0' shape=(None, 1) dtype=int64>, 'spareribs': <tf.Tensor 'ExpandDims_596:0' shape=(None, 1) dtype=int64>, 'spearmint': <tf.Tensor 'ExpandDims_597:0' shape=(None, 1) dtype=int64>, 'specialty_cheese': <tf.Tensor 'ExpandDims_598:0' shape=(None, 1) dtype=int64>, 'spinach': <tf.Tensor 'ExpandDims_599:0' shape=(None, 1) dtype=int64>, 'squash': <tf.Tensor 'ExpandDims_600:0' shape=(None, 1) dtype=int64>, 'squid': <tf.Tensor 'ExpandDims_601:0' shape=(None, 1) dtype=int64>, 'sriracha_hot_sauce': <tf.Tensor 'ExpandDims_602:0' shape=(None, 1) dtype=int64>, 'star_anise': <tf.Tensor 'ExpandDims_603:0' shape=(None, 1) dtype=int64>, 'steaks': <tf.Tensor 'ExpandDims_604:0' shape=(None, 1) dtype=int64>, 'stilton': <tf.Tensor 'ExpandDims_605:0' shape=(None, 1) dtype=int64>, 'stir_fry_vegetables': <tf.Tensor 'ExpandDims_606:0' shape=(None, 1) dtype=int64>, 'stock': <tf.Tensor 'ExpandDims_607:0' shape=(None, 1) dtype=int64>, 'straw_mushroom': <tf.Tensor 'ExpandDims_608:0' shape=(None, 1) dtype=int64>, 'strawberry': <tf.Tensor 'ExpandDims_609:0' shape=(None, 1) dtype=int64>, 'stuffing': <tf.Tensor 'ExpandDims_610:0' shape=(None, 1) dtype=int64>, 'suet': <tf.Tensor 'ExpandDims_611:0' shape=(None, 1) dtype=int64>, 'sugar': <tf.Tensor 'ExpandDims_612:0' shape=(None, 1) dtype=int64>, 'sultana': <tf.Tensor 'ExpandDims_613:0' shape=(None, 1) dtype=int64>, 'sumac': <tf.Tensor 'ExpandDims_614:0' shape=(None, 1) dtype=int64>, 'sunflower_oil': <tf.Tensor 'ExpandDims_615:0' shape=(None, 1) dtype=int64>, 'sunflower_seed': <tf.Tensor 'ExpandDims_616:0' shape=(None, 1) dtype=int64>, 'sushi_rice': <tf.Tensor 'ExpandDims_617:0' shape=(None, 1) dtype=int64>, 'sweet_pepper': <tf.Tensor 'ExpandDims_618:0' shape=(None, 1) dtype=int64>, 'sweet_potato': <tf.Tensor 'ExpandDims_619:0' shape=(None, 1) dtype=int64>, 'swiss_cheese': <tf.Tensor 'ExpandDims_620:0' shape=(None, 1) dtype=int64>, 'swordfish': <tf.Tensor 'ExpandDims_621:0' shape=(None, 1) dtype=int64>, 'szechuan_peppercorn': <tf.Tensor 'ExpandDims_622:0' shape=(None, 1) dtype=int64>, 'tabasco_sauce': <tf.Tensor 'ExpandDims_623:0' shape=(None, 1) dtype=int64>, 'taco_seasoning': <tf.Tensor 'ExpandDims_624:0' shape=(None, 1) dtype=int64>, 'tahini': <tf.Tensor 'ExpandDims_625:0' shape=(None, 1) dtype=int64>, 'tamarind': <tf.Tensor 'ExpandDims_626:0' shape=(None, 1) dtype=int64>, 'tapioca': <tf.Tensor 'ExpandDims_627:0' shape=(None, 1) dtype=int64>, 'tarragon': <tf.Tensor 'ExpandDims_628:0' shape=(None, 1) dtype=int64>, 'tatsoi': <tf.Tensor 'ExpandDims_629:0' shape=(None, 1) dtype=int64>, 'tea': <tf.Tensor 'ExpandDims_630:0' shape=(None, 1) dtype=int64>, 'tempeh': <tf.Tensor 'ExpandDims_631:0' shape=(None, 1) dtype=int64>, 'tequila': <tf.Tensor 'ExpandDims_632:0' shape=(None, 1) dtype=int64>, 'teriyaki_marinade': <tf.Tensor 'ExpandDims_633:0' shape=(None, 1) dtype=int64>, 'teriyaki_sauce': <tf.Tensor 'ExpandDims_634:0' shape=(None, 1) dtype=int64>, 'Thai_chile_pepper': <tf.Tensor 'ExpandDims_17:0' shape=(None, 1) dtype=int64>, 'Thai_curry_paste': <tf.Tensor 'ExpandDims_18:0' shape=(None, 1) dtype=int64>, 'thousand_island_dressing': <tf.Tensor 'ExpandDims_635:0' shape=(None, 1) dtype=int64>, 'thyme': <tf.Tensor 'ExpandDims_636:0' shape=(None, 1) dtype=int64>, 'tilapia_fillets': <tf.Tensor 'ExpandDims_637:0' shape=(None, 1) dtype=int64>, 'tofu': <tf.Tensor 'ExpandDims_638:0' shape=(None, 1) dtype=int64>, 'togarashi': <tf.Tensor 'ExpandDims_639:0' shape=(None, 1) dtype=int64>, 'tomatillos': <tf.Tensor 'ExpandDims_640:0' shape=(None, 1) dtype=int64>, 'tomato': <tf.Tensor 'ExpandDims_641:0' shape=(None, 1) dtype=int64>, 'tomato_juice': <tf.Tensor 'ExpandDims_642:0' shape=(None, 1) dtype=int64>, 'tomato_paste': <tf.Tensor 'ExpandDims_643:0' shape=(None, 1) dtype=int64>, 'tomato_puree': <tf.Tensor 'ExpandDims_644:0' shape=(None, 1) dtype=int64>, 'tomato_sauce': <tf.Tensor 'ExpandDims_645:0' shape=(None, 1) dtype=int64>, 'tomato_soup': <tf.Tensor 'ExpandDims_646:0' shape=(None, 1) dtype=int64>, 'toor_dal': <tf.Tensor 'ExpandDims_647:0' shape=(None, 1) dtype=int64>, 'tortilla': <tf.Tensor 'ExpandDims_648:0' shape=(None, 1) dtype=int64>, 'trout': <tf.Tensor 'ExpandDims_649:0' shape=(None, 1) dtype=int64>, 'truffle': <tf.Tensor 'ExpandDims_650:0' shape=(None, 1) dtype=int64>, 'truffle_oil': <tf.Tensor 'ExpandDims_651:0' shape=(None, 1) dtype=int64>, 'tuna': <tf.Tensor 'ExpandDims_652:0' shape=(None, 1) dtype=int64>, 'turbinado_sugar': <tf.Tensor 'ExpandDims_653:0' shape=(None, 1) dtype=int64>, 'turkey': <tf.Tensor 'ExpandDims_654:0' shape=(None, 1) dtype=int64>, 'turmeric': <tf.Tensor 'ExpandDims_655:0' shape=(None, 1) dtype=int64>, 'turnip': <tf.Tensor 'ExpandDims_656:0' shape=(None, 1) dtype=int64>, 'udon_noodles': <tf.Tensor 'ExpandDims_657:0' shape=(None, 1) dtype=int64>, 'umeboshi': <tf.Tensor 'ExpandDims_658:0' shape=(None, 1) dtype=int64>, 'unsweetened_chocolate': <tf.Tensor 'ExpandDims_659:0' shape=(None, 1) dtype=int64>, 'urad_dal': <tf.Tensor 'ExpandDims_660:0' shape=(None, 1) dtype=int64>, 'vanilla': <tf.Tensor 'ExpandDims_661:0' shape=(None, 1) dtype=int64>, 'veal': <tf.Tensor 'ExpandDims_662:0' shape=(None, 1) dtype=int64>, 'vegan_brand_names': <tf.Tensor 'ExpandDims_663:0' shape=(None, 1) dtype=int64>, 'vegetable_juice': <tf.Tensor 'ExpandDims_664:0' shape=(None, 1) dtype=int64>, 'vegetable_oil': <tf.Tensor 'ExpandDims_665:0' shape=(None, 1) dtype=int64>, 'vegetable_stock': <tf.Tensor 'ExpandDims_666:0' shape=(None, 1) dtype=int64>, 'vegetarian_brand_names': <tf.Tensor 'ExpandDims_667:0' shape=(None, 1) dtype=int64>, 'velveeta': <tf.Tensor 'ExpandDims_668:0' shape=(None, 1) dtype=int64>, 'venison': <tf.Tensor 'ExpandDims_669:0' shape=(None, 1) dtype=int64>, 'vermicelli': <tf.Tensor 'ExpandDims_670:0' shape=(None, 1) dtype=int64>, 'vermouth': <tf.Tensor 'ExpandDims_671:0' shape=(None, 1) dtype=int64>, 'Vidalia_onion': <tf.Tensor 'ExpandDims_19:0' shape=(None, 1) dtype=int64>, 'vinegar': <tf.Tensor 'ExpandDims_672:0' shape=(None, 1) dtype=int64>, 'vital_wheat_gluten': <tf.Tensor 'ExpandDims_673:0' shape=(None, 1) dtype=int64>, 'vodka': <tf.Tensor 'ExpandDims_674:0' shape=(None, 1) dtype=int64>, 'walnut': <tf.Tensor 'ExpandDims_675:0' shape=(None, 1) dtype=int64>, 'wasabi': <tf.Tensor 'ExpandDims_676:0' shape=(None, 1) dtype=int64>, 'washed_rind_cheese': <tf.Tensor 'ExpandDims_677:0' shape=(None, 1) dtype=int64>, 'water': <tf.Tensor 'ExpandDims_678:0' shape=(None, 1) dtype=int64>, 'water_chestnut': <tf.Tensor 'ExpandDims_679:0' shape=(None, 1) dtype=int64>, 'watercress': <tf.Tensor 'ExpandDims_680:0' shape=(None, 1) dtype=int64>, 'watermelon': <tf.Tensor 'ExpandDims_681:0' shape=(None, 1) dtype=int64>, 'wheat_bread': <tf.Tensor 'ExpandDims_682:0' shape=(None, 1) dtype=int64>, 'whipping_cream': <tf.Tensor 'ExpandDims_683:0' shape=(None, 1) dtype=int64>, 'white_bean': <tf.Tensor 'ExpandDims_684:0' shape=(None, 1) dtype=int64>, 'white_bread': <tf.Tensor 'ExpandDims_685:0' shape=(None, 1) dtype=int64>, 'white_cheese': <tf.Tensor 'ExpandDims_686:0' shape=(None, 1) dtype=int64>, 'white_fish': <tf.Tensor 'ExpandDims_687:0' shape=(None, 1) dtype=int64>, 'white_onion': <tf.Tensor 'ExpandDims_688:0' shape=(None, 1) dtype=int64>, 'white_peppercorns': <tf.Tensor 'ExpandDims_689:0' shape=(None, 1) dtype=int64>, 'white_sesame_seeds': <tf.Tensor 'ExpandDims_690:0' shape=(None, 1) dtype=int64>, 'whole_grain_mustard': <tf.Tensor 'ExpandDims_691:0' shape=(None, 1) dtype=int64>, 'whole_milk': <tf.Tensor 'ExpandDims_692:0' shape=(None, 1) dtype=int64>, 'wild_rice': <tf.Tensor 'ExpandDims_693:0' shape=(None, 1) dtype=int64>, 'wine': <tf.Tensor 'ExpandDims_694:0' shape=(None, 1) dtype=int64>, 'wine_vinegar': <tf.Tensor 'ExpandDims_695:0' shape=(None, 1) dtype=int64>, 'winter_squash': <tf.Tensor 'ExpandDims_696:0' shape=(None, 1) dtype=int64>, 'worcestershire_sauce': <tf.Tensor 'ExpandDims_697:0' shape=(None, 1) dtype=int64>, 'xanthan_gum': <tf.Tensor 'ExpandDims_698:0' shape=(None, 1) dtype=int64>, 'yams': <tf.Tensor 'ExpandDims_699:0' shape=(None, 1) dtype=int64>, 'yeast': <tf.Tensor 'ExpandDims_700:0' shape=(None, 1) dtype=int64>, 'yellow_curry_paste': <tf.Tensor 'ExpandDims_701:0' shape=(None, 1) dtype=int64>, 'yellow_food_coloring': <tf.Tensor 'ExpandDims_702:0' shape=(None, 1) dtype=int64>, 'yellow_split_pea': <tf.Tensor 'ExpandDims_703:0' shape=(None, 1) dtype=int64>, 'yellow_squash': <tf.Tensor 'ExpandDims_704:0' shape=(None, 1) dtype=int64>, 'yogurt': <tf.Tensor 'ExpandDims_705:0' shape=(None, 1) dtype=int64>, 'zaatar': <tf.Tensor 'ExpandDims_706:0' shape=(None, 1) dtype=int64>, 'zest': <tf.Tensor 'ExpandDims_707:0' shape=(None, 1) dtype=int64>, 'zucchini': <tf.Tensor 'ExpandDims_708:0' shape=(None, 1) dtype=int64>}\n",
            "Consider rewriting this model with the Functional API.\n",
            "47/47 [==============================] - 28s 246ms/step - loss: 2.4183 - sparse_categorical_accuracy: 0.1892 - val_loss: 2.0062 - val_sparse_categorical_accuracy: 0.6411\n",
            "Epoch 2/1000\n",
            "47/47 [==============================] - 4s 73ms/step - loss: 1.8562 - sparse_categorical_accuracy: 0.6147 - val_loss: 1.3536 - val_sparse_categorical_accuracy: 0.7509\n",
            "Epoch 3/1000\n",
            "47/47 [==============================] - 4s 74ms/step - loss: 1.2940 - sparse_categorical_accuracy: 0.6877 - val_loss: 1.0090 - val_sparse_categorical_accuracy: 0.7686\n",
            "Epoch 4/1000\n",
            "47/47 [==============================] - 4s 73ms/step - loss: 1.0017 - sparse_categorical_accuracy: 0.7415 - val_loss: 0.8618 - val_sparse_categorical_accuracy: 0.7698\n",
            "Epoch 5/1000\n",
            "47/47 [==============================] - 4s 73ms/step - loss: 0.8550 - sparse_categorical_accuracy: 0.7694 - val_loss: 0.7880 - val_sparse_categorical_accuracy: 0.7721\n",
            "Epoch 6/1000\n",
            "47/47 [==============================] - 4s 74ms/step - loss: 0.7756 - sparse_categorical_accuracy: 0.7873 - val_loss: 0.7494 - val_sparse_categorical_accuracy: 0.7804\n",
            "Epoch 7/1000\n",
            "47/47 [==============================] - 4s 74ms/step - loss: 0.6843 - sparse_categorical_accuracy: 0.8058 - val_loss: 0.7247 - val_sparse_categorical_accuracy: 0.7922\n",
            "Epoch 8/1000\n",
            "47/47 [==============================] - 4s 74ms/step - loss: 0.6548 - sparse_categorical_accuracy: 0.8135 - val_loss: 0.7110 - val_sparse_categorical_accuracy: 0.7910\n",
            "Epoch 9/1000\n",
            "47/47 [==============================] - 4s 72ms/step - loss: 0.5938 - sparse_categorical_accuracy: 0.8263 - val_loss: 0.7021 - val_sparse_categorical_accuracy: 0.7922\n",
            "Epoch 10/1000\n",
            "47/47 [==============================] - 4s 73ms/step - loss: 0.6069 - sparse_categorical_accuracy: 0.8297 - val_loss: 0.6964 - val_sparse_categorical_accuracy: 0.7875\n",
            "Epoch 11/1000\n",
            "47/47 [==============================] - 4s 76ms/step - loss: 0.5533 - sparse_categorical_accuracy: 0.8470 - val_loss: 0.6945 - val_sparse_categorical_accuracy: 0.7898\n",
            "Epoch 12/1000\n",
            "47/47 [==============================] - 4s 75ms/step - loss: 0.5390 - sparse_categorical_accuracy: 0.8440 - val_loss: 0.6939 - val_sparse_categorical_accuracy: 0.7887\n",
            "Epoch 13/1000\n",
            "47/47 [==============================] - 4s 73ms/step - loss: 0.5211 - sparse_categorical_accuracy: 0.8593 - val_loss: 0.6948 - val_sparse_categorical_accuracy: 0.7934\n",
            "Epoch 14/1000\n",
            "47/47 [==============================] - 4s 72ms/step - loss: 0.5201 - sparse_categorical_accuracy: 0.8519 - val_loss: 0.6968 - val_sparse_categorical_accuracy: 0.7851\n",
            "Epoch 15/1000\n",
            "47/47 [==============================] - 4s 73ms/step - loss: 0.4661 - sparse_categorical_accuracy: 0.8877 - val_loss: 0.6965 - val_sparse_categorical_accuracy: 0.7839\n",
            "Epoch 16/1000\n",
            "47/47 [==============================] - 4s 71ms/step - loss: 0.4332 - sparse_categorical_accuracy: 0.8897 - val_loss: 0.6995 - val_sparse_categorical_accuracy: 0.7851\n",
            "Epoch 17/1000\n",
            "47/47 [==============================] - 4s 74ms/step - loss: 0.4556 - sparse_categorical_accuracy: 0.8818 - val_loss: 0.7030 - val_sparse_categorical_accuracy: 0.7816\n",
            "Epoch 18/1000\n",
            "47/47 [==============================] - 4s 73ms/step - loss: 0.4277 - sparse_categorical_accuracy: 0.8961 - val_loss: 0.7067 - val_sparse_categorical_accuracy: 0.7851\n",
            "Epoch 19/1000\n",
            "47/47 [==============================] - 4s 77ms/step - loss: 0.4103 - sparse_categorical_accuracy: 0.8896 - val_loss: 0.7123 - val_sparse_categorical_accuracy: 0.7792\n",
            "Epoch 20/1000\n",
            "47/47 [==============================] - 4s 77ms/step - loss: 0.4026 - sparse_categorical_accuracy: 0.8941 - val_loss: 0.7166 - val_sparse_categorical_accuracy: 0.7757\n",
            "Epoch 21/1000\n",
            "47/47 [==============================] - 4s 77ms/step - loss: 0.4079 - sparse_categorical_accuracy: 0.8987 - val_loss: 0.7187 - val_sparse_categorical_accuracy: 0.7839\n",
            "Epoch 22/1000\n",
            "47/47 [==============================] - 4s 72ms/step - loss: 0.3812 - sparse_categorical_accuracy: 0.9065 - val_loss: 0.7231 - val_sparse_categorical_accuracy: 0.7828\n",
            "Restoring model weights from the end of the best epoch.\n",
            "Epoch 00022: early stopping\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f9f4c3317d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iZQWmo0AEEZi",
        "outputId": "b5915e06-f33d-4dda-df13-c20777549e12"
      },
      "source": [
        "test_losss, test_accuracy = model.evaluate(test_ds)\n",
        "print(test_accuracy)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "7/7 [==============================] - 1s 45ms/step - loss: 0.7165 - sparse_categorical_accuracy: 0.7925\n",
            "0.7924528121948242\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IsPQV5QD2hqm"
      },
      "source": [
        "#### Confusion matrix for original feature representation data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "30NWqpe14uvm"
      },
      "source": [
        "test_results = model.predict(test_ds)\n",
        "test_results = [np.argmax(x) for x in test_results]"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ogaZBB3yUZTJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "outputId": "17cfbfd0-7634-4f2c-a6bb-ccd26a9a3005"
      },
      "source": [
        "# Import cuisine labels\n",
        "cuisines_df = pd.read_csv('Cuisines.csv', header=None)\n",
        "cuisine_labels = cuisines_df[1].values\n",
        "print(cuisine_labels)\n",
        "cuisines_df.head()"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['Chinese' 'English' 'French' 'German' 'Greek' 'Indian' 'Italian'\n",
            " 'Japanese' 'Mexican' 'Moroccan' 'Spanish' 'Thai']\n"
          ],
          "name": "stdout"
        },
        {
          "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>0</th>\n",
              "      <th>1</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>Chinese</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>English</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>French</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>German</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>Greek</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   0        1\n",
              "0  1  Chinese\n",
              "1  2  English\n",
              "2  3   French\n",
              "3  4   German\n",
              "4  5    Greek"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9cG-x_LoTflk",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 535
        },
        "outputId": "3e65d3b8-0e5b-4768-ba68-ca72173c63c3"
      },
      "source": [
        "# Plot confusion matrix\n",
        "confusion_mtx = tf.math.confusion_matrix(test.cuisine.values, test_results) \n",
        "plt.figure(figsize=(10, 8))\n",
        "sns.heatmap(confusion_mtx, xticklabels=cuisine_labels, yticklabels=cuisine_labels, \n",
        "            annot=True, fmt='g')\n",
        "plt.xlabel('Prediction')\n",
        "plt.ylabel('Label')\n",
        "plt.savefig('confusion_matrix.pdf', pad_inches=0)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x576 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jaWSjWAW2YV0"
      },
      "source": [
        "### Training model on PCA/UMAP data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2GZEhmKDPrU-",
        "outputId": "e3fed80a-0f66-4432-9958-c09a0ccfe57c"
      },
      "source": [
        "model = tf.keras.Sequential([\n",
        "  layers.Dense(128, activation='relu', kernel_regularizer=tf.keras.regularizers.l2(l2=1e-4)),\n",
        "  layers.Dropout(0.5),\n",
        "  layers.Dense(12, kernel_regularizer=tf.keras.regularizers.l2(l2=1e-4))\n",
        "])\n",
        "\n",
        "model.compile(optimizer='adam',\n",
        "              loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True),\n",
        "              metrics=tf.metrics.CategoricalAccuracy())\n",
        "\n",
        "model.fit(train_ds,\n",
        "          validation_data=val_ds,\n",
        "          callbacks=[early_stop_callback],\n",
        "          batch_size=64,\n",
        "          epochs=1000)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/1000\n",
            "47/47 [==============================] - 1s 5ms/step - loss: 3.0407 - categorical_accuracy: 0.1048 - val_loss: 2.1540 - val_categorical_accuracy: 0.2078\n",
            "Epoch 2/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 2.3806 - categorical_accuracy: 0.1647 - val_loss: 1.9962 - val_categorical_accuracy: 0.2361\n",
            "Epoch 3/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 2.1653 - categorical_accuracy: 0.2147 - val_loss: 1.9272 - val_categorical_accuracy: 0.2538\n",
            "Epoch 4/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 2.0432 - categorical_accuracy: 0.2298 - val_loss: 1.8831 - val_categorical_accuracy: 0.2810\n",
            "Epoch 5/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.9488 - categorical_accuracy: 0.2729 - val_loss: 1.8536 - val_categorical_accuracy: 0.2940\n",
            "Epoch 6/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.9101 - categorical_accuracy: 0.2718 - val_loss: 1.8309 - val_categorical_accuracy: 0.3058\n",
            "Epoch 7/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.8574 - categorical_accuracy: 0.2988 - val_loss: 1.8152 - val_categorical_accuracy: 0.3223\n",
            "Epoch 8/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.8463 - categorical_accuracy: 0.3149 - val_loss: 1.7982 - val_categorical_accuracy: 0.3424\n",
            "Epoch 9/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.8170 - categorical_accuracy: 0.3221 - val_loss: 1.7898 - val_categorical_accuracy: 0.3259\n",
            "Epoch 10/1000\n",
            "47/47 [==============================] - 0s 3ms/step - loss: 1.7901 - categorical_accuracy: 0.3262 - val_loss: 1.7788 - val_categorical_accuracy: 0.3388\n",
            "Epoch 11/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.7773 - categorical_accuracy: 0.3307 - val_loss: 1.7655 - val_categorical_accuracy: 0.4050\n",
            "Epoch 12/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.7803 - categorical_accuracy: 0.3395 - val_loss: 1.7557 - val_categorical_accuracy: 0.3731\n",
            "Epoch 13/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.7439 - categorical_accuracy: 0.3447 - val_loss: 1.7500 - val_categorical_accuracy: 0.3672\n",
            "Epoch 14/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.7313 - categorical_accuracy: 0.3636 - val_loss: 1.7475 - val_categorical_accuracy: 0.4333\n",
            "Epoch 15/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.7079 - categorical_accuracy: 0.3783 - val_loss: 1.7383 - val_categorical_accuracy: 0.4498\n",
            "Epoch 16/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.6899 - categorical_accuracy: 0.3957 - val_loss: 1.7296 - val_categorical_accuracy: 0.4357\n",
            "Epoch 17/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.6821 - categorical_accuracy: 0.3956 - val_loss: 1.7232 - val_categorical_accuracy: 0.4581\n",
            "Epoch 18/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.6661 - categorical_accuracy: 0.4020 - val_loss: 1.7126 - val_categorical_accuracy: 0.4699\n",
            "Epoch 19/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.6618 - categorical_accuracy: 0.4184 - val_loss: 1.7059 - val_categorical_accuracy: 0.4923\n",
            "Epoch 20/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.6527 - categorical_accuracy: 0.4337 - val_loss: 1.6967 - val_categorical_accuracy: 0.5030\n",
            "Epoch 21/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.6248 - categorical_accuracy: 0.4357 - val_loss: 1.6891 - val_categorical_accuracy: 0.4770\n",
            "Epoch 22/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.6161 - categorical_accuracy: 0.4555 - val_loss: 1.6761 - val_categorical_accuracy: 0.4805\n",
            "Epoch 23/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5969 - categorical_accuracy: 0.4718 - val_loss: 1.6758 - val_categorical_accuracy: 0.4900\n",
            "Epoch 24/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5771 - categorical_accuracy: 0.4665 - val_loss: 1.6672 - val_categorical_accuracy: 0.5136\n",
            "Epoch 25/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5734 - categorical_accuracy: 0.4784 - val_loss: 1.6567 - val_categorical_accuracy: 0.5301\n",
            "Epoch 26/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5542 - categorical_accuracy: 0.4720 - val_loss: 1.6481 - val_categorical_accuracy: 0.4970\n",
            "Epoch 27/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5543 - categorical_accuracy: 0.4756 - val_loss: 1.6412 - val_categorical_accuracy: 0.5254\n",
            "Epoch 28/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5342 - categorical_accuracy: 0.4834 - val_loss: 1.6411 - val_categorical_accuracy: 0.5360\n",
            "Epoch 29/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5088 - categorical_accuracy: 0.5039 - val_loss: 1.6329 - val_categorical_accuracy: 0.5230\n",
            "Epoch 30/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5055 - categorical_accuracy: 0.5007 - val_loss: 1.6314 - val_categorical_accuracy: 0.5289\n",
            "Epoch 31/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4848 - categorical_accuracy: 0.5182 - val_loss: 1.6210 - val_categorical_accuracy: 0.5455\n",
            "Epoch 32/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.5033 - categorical_accuracy: 0.5090 - val_loss: 1.6146 - val_categorical_accuracy: 0.5443\n",
            "Epoch 33/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4742 - categorical_accuracy: 0.4953 - val_loss: 1.6106 - val_categorical_accuracy: 0.5443\n",
            "Epoch 34/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4751 - categorical_accuracy: 0.5003 - val_loss: 1.6045 - val_categorical_accuracy: 0.5514\n",
            "Epoch 35/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4740 - categorical_accuracy: 0.5355 - val_loss: 1.5986 - val_categorical_accuracy: 0.5773\n",
            "Epoch 36/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4472 - categorical_accuracy: 0.5276 - val_loss: 1.5994 - val_categorical_accuracy: 0.5702\n",
            "Epoch 37/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4526 - categorical_accuracy: 0.5274 - val_loss: 1.5980 - val_categorical_accuracy: 0.5561\n",
            "Epoch 38/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4344 - categorical_accuracy: 0.5350 - val_loss: 1.5884 - val_categorical_accuracy: 0.5762\n",
            "Epoch 39/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4350 - categorical_accuracy: 0.5343 - val_loss: 1.5809 - val_categorical_accuracy: 0.5797\n",
            "Epoch 40/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4272 - categorical_accuracy: 0.5499 - val_loss: 1.5835 - val_categorical_accuracy: 0.5797\n",
            "Epoch 41/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4220 - categorical_accuracy: 0.5325 - val_loss: 1.5879 - val_categorical_accuracy: 0.5702\n",
            "Epoch 42/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.4027 - categorical_accuracy: 0.5431 - val_loss: 1.5804 - val_categorical_accuracy: 0.5596\n",
            "Epoch 43/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3956 - categorical_accuracy: 0.5393 - val_loss: 1.5808 - val_categorical_accuracy: 0.5773\n",
            "Epoch 44/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3892 - categorical_accuracy: 0.5590 - val_loss: 1.5781 - val_categorical_accuracy: 0.5785\n",
            "Epoch 45/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3876 - categorical_accuracy: 0.5453 - val_loss: 1.5722 - val_categorical_accuracy: 0.5868\n",
            "Epoch 46/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3864 - categorical_accuracy: 0.5592 - val_loss: 1.5716 - val_categorical_accuracy: 0.5868\n",
            "Epoch 47/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3855 - categorical_accuracy: 0.5510 - val_loss: 1.5698 - val_categorical_accuracy: 0.5762\n",
            "Epoch 48/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3768 - categorical_accuracy: 0.5595 - val_loss: 1.5694 - val_categorical_accuracy: 0.5809\n",
            "Epoch 49/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3942 - categorical_accuracy: 0.5546 - val_loss: 1.5688 - val_categorical_accuracy: 0.5773\n",
            "Epoch 50/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3680 - categorical_accuracy: 0.5553 - val_loss: 1.5637 - val_categorical_accuracy: 0.5797\n",
            "Epoch 51/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3807 - categorical_accuracy: 0.5496 - val_loss: 1.5646 - val_categorical_accuracy: 0.5762\n",
            "Epoch 52/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3563 - categorical_accuracy: 0.5613 - val_loss: 1.5670 - val_categorical_accuracy: 0.5868\n",
            "Epoch 53/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3520 - categorical_accuracy: 0.5819 - val_loss: 1.5590 - val_categorical_accuracy: 0.5915\n",
            "Epoch 54/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3486 - categorical_accuracy: 0.5766 - val_loss: 1.5718 - val_categorical_accuracy: 0.5868\n",
            "Epoch 55/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3475 - categorical_accuracy: 0.5834 - val_loss: 1.5699 - val_categorical_accuracy: 0.5832\n",
            "Epoch 56/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3502 - categorical_accuracy: 0.5815 - val_loss: 1.5713 - val_categorical_accuracy: 0.5880\n",
            "Epoch 57/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3387 - categorical_accuracy: 0.5699 - val_loss: 1.5688 - val_categorical_accuracy: 0.5797\n",
            "Epoch 58/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3435 - categorical_accuracy: 0.5677 - val_loss: 1.5677 - val_categorical_accuracy: 0.5868\n",
            "Epoch 59/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3452 - categorical_accuracy: 0.5692 - val_loss: 1.5710 - val_categorical_accuracy: 0.5832\n",
            "Epoch 60/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3315 - categorical_accuracy: 0.5620 - val_loss: 1.5654 - val_categorical_accuracy: 0.5915\n",
            "Epoch 61/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3341 - categorical_accuracy: 0.5752 - val_loss: 1.5670 - val_categorical_accuracy: 0.5856\n",
            "Epoch 62/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3292 - categorical_accuracy: 0.5751 - val_loss: 1.5631 - val_categorical_accuracy: 0.5927\n",
            "Epoch 63/1000\n",
            "47/47 [==============================] - 0s 2ms/step - loss: 1.3272 - categorical_accuracy: 0.5813 - val_loss: 1.5652 - val_categorical_accuracy: 0.5903\n",
            "Restoring model weights from the end of the best epoch.\n",
            "Epoch 00063: early stopping\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f26fb90e490>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 222
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6jJrzpUaTXdn",
        "outputId": "de9c0fa6-7ff8-4ac4-f6f9-4b160a3b29fb"
      },
      "source": [
        "# PCA\n",
        "test_losss, test_accuracy = model.evaluate(test_ds)\n",
        "print(test_accuracy)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "7/7 [==============================] - 0s 2ms/step - loss: 0.8669 - categorical_accuracy: 0.7146\n",
            "0.7146226167678833\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0S-yPJWuU2xm",
        "outputId": "e3d3d89d-fa63-4734-b087-e183f692e748"
      },
      "source": [
        "# UMAP\n",
        "test_losss, test_accuracy = model.evaluate(test_ds)\n",
        "print(test_accuracy)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "7/7 [==============================] - 0s 1ms/step - loss: 1.5682 - categorical_accuracy: 0.5896\n",
            "0.5896226167678833\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "E3hWwcpUnyOY"
      },
      "source": [
        "## Model 2 - Logistic Regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dy6-87WRM37h",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "cd4527c6-2ed7-4b2b-826c-98a172dd85a6"
      },
      "source": [
        "clf = LogisticRegression(random_state=0, max_iter=300, C=0.8).fit(X_train, y_train)\n",
        "clf.predict(X_val)\n",
        "clf.score(X_val, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7756788665879575"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 301
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Sd6ijdKGmxoy",
        "outputId": "9d192fc5-1f19-421e-8e36-cfac3f06851e"
      },
      "source": [
        "clf.score(X_test, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7665094339622641"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 302
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oYd_PVZcaRbW",
        "outputId": "530d82d7-02b7-4274-b8fd-c08a2ae94d05"
      },
      "source": [
        "# With dimensionally reduced data\n",
        "clf = LogisticRegression(random_state=0, max_iter=300, C=0.4).fit(X_train_PCA, y_train)\n",
        "clf.predict(X_val_PCA)\n",
        "clf.score(X_val_PCA, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7367178276269185"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 104
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8HQaEG4Xm2C8",
        "outputId": "272b51b5-8269-47b0-d99b-56bfad05a26e"
      },
      "source": [
        "clf.score(X_test_PCA, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7169811320754716"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DNubV2h7siPF",
        "outputId": "46c16974-2e6f-465b-c0a5-badbe73e6a6e"
      },
      "source": [
        "# with UMAP data\n",
        "clf = LogisticRegression(random_state=0, max_iter=500, C=0.5)\n",
        "clf.fit(X_train_umap, y_train)\n",
        "clf.score(X_val_umap, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5501770956316411"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 113
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D_X2sSmIsxZW",
        "outputId": "89acd81d-bc91-41cd-dea3-dfae5102bfa7"
      },
      "source": [
        "clf.score(X_test_umap, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5754716981132075"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 114
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gn2oEc4un1PC"
      },
      "source": [
        "## Model 3 - Random Forest"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5ctXvnJfU5cg",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7db51f13-9802-4d91-e509-ab23aad73aaa"
      },
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "clf = RandomForestClassifier(n_estimators=300, max_depth=50, random_state=0, min_samples_split=3)\n",
        "clf.fit(X_train, y_train)\n",
        "clf.predict(X_val)\n",
        "clf.score(X_val, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.781582054309327"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5GNySQ08cI3J",
        "outputId": "460c8d7a-5ced-418c-86c2-a871edb25556"
      },
      "source": [
        "clf.score(X_test, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7311320754716981"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 307
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oRs5kn7pZygB",
        "outputId": "ec0cb3e6-667d-4afc-a03a-1758edd8142f"
      },
      "source": [
        "# With dimensionally reduced data\n",
        "clf = RandomForestClassifier(n_estimators=300, max_depth=30, random_state=0, min_samples_split=5)\n",
        "clf.fit(X_train_PCA, y_train)\n",
        "clf.score(X_val_PCA, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7284533648170012"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 119
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bm5U17UlnB5b",
        "outputId": "489f3fed-75bd-4b97-a2db-36e0d56ea753"
      },
      "source": [
        "clf.score(X_test_PCA, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6698113207547169"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KNsyGKlus9VG",
        "outputId": "bcb8823f-5a94-411e-86ec-94a46a8783e1"
      },
      "source": [
        "# with UMAP data\n",
        "clf = RandomForestClassifier(n_estimators=300, max_depth=10, random_state=0, min_samples_split=10)\n",
        "clf.fit(X_train_umap, y_train)\n",
        "clf.score(X_val_umap, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6210153482880756"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FKuwl7RUtWlf",
        "outputId": "fdb7147b-c53d-4eba-c111-c27ae35d1167"
      },
      "source": [
        "clf.score(X_test_umap, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5966981132075472"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aZUurUxrn28m"
      },
      "source": [
        "## Model 4 - K Nearest Neighbors"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7bb2RNBIVV8a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ff4b1a46-cc97-46c0-85c8-0c5f29689f3f"
      },
      "source": [
        "\n",
        "neigh = KNeighborsClassifier(n_neighbors=12, n_jobs=-1, algorithm='ball_tree', leaf_size=30)\n",
        "neigh.fit(X_train, y_train)\n",
        "neigh.score(X_val, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.667060212514758"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 156
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pQkGbGTmmq5n",
        "outputId": "5c2466f9-5117-4ca4-aea4-43776f702728"
      },
      "source": [
        "neigh.score(X_test, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6155660377358491"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 157
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0SzPqUYDgfaC",
        "outputId": "8b6dd01e-3f91-471c-dc00-e55b1e74d30e"
      },
      "source": [
        "# With dimensionally reduced data\n",
        "neigh = KNeighborsClassifier(n_neighbors=15, n_jobs=-1, algorithm='ball_tree', leaf_size=30)\n",
        "neigh.fit(X_train_PCA, y_train)\n",
        "neigh.score(X_val_PCA, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6611570247933884"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 158
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pimwxi6OnIO-",
        "outputId": "f1408f12-2a69-446e-d34b-148dde3096c6"
      },
      "source": [
        "neigh.score(X_test_PCA, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6132075471698113"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 159
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d9BrIrE2rHwn",
        "outputId": "ada51bfe-e1aa-493d-c8d6-2a11cee5c266"
      },
      "source": [
        "# with UMAP data\n",
        "neigh = KNeighborsClassifier(n_neighbors=20, n_jobs=-1, algorithm='ball_tree', leaf_size=30)\n",
        "neigh.fit(X_train_umap, y_train)\n",
        "neigh.score(X_val_umap, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6245572609208973"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 152
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "00i3gAaOLxxT",
        "outputId": "12255723-e435-410b-b011-218987281886"
      },
      "source": [
        "neigh.score(X_test_umap, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6037735849056604"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 153
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8vM2YFBClccx"
      },
      "source": [
        "## Model 5 - Support Vector Machine"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7mlCtUH6iB_z",
        "outputId": "d83d9b24-3a94-404b-ade7-596bc5131a8a"
      },
      "source": [
        "svc_clf = SVC() #C=1, degree=5 same as default\n",
        "svc_clf.fit(X_train, y_train)\n",
        "svc_clf.score(X_val, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7768595041322314"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 298
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rj1PcYDtmnSC",
        "outputId": "d825dc42-877b-424c-ea74-fa0915e7e680"
      },
      "source": [
        "svc_clf.score(X_test, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7452830188679245"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 315
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ebf0eaQonTXd",
        "outputId": "993b71c4-3ea8-4034-ee96-9ac1166d24e7"
      },
      "source": [
        "svc_clf = SVC() #C=1, degree=5 same as default\n",
        "svc_clf.fit(X_train_PCA, y_train)\n",
        "svc_clf.score(X_val_PCA, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7485242030696576"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qL9AUXDbnWcm",
        "outputId": "d4b9df50-2490-4cf8-bf6a-963bcb8512ad"
      },
      "source": [
        "svc_clf.score(X_test_PCA, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7099056603773585"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 317
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f6zvByjeuAJL",
        "outputId": "0e560ebc-74e5-47a8-8a5c-dc4a4187e564"
      },
      "source": [
        "svc_clf.fit(X_train_umap, y_train)\n",
        "svc_clf.score(X_val_umap, y_val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6174734356552538"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S5ueqUdTuJIx",
        "outputId": "a8b8a155-46a9-4cb5-e095-9c8633bb159c"
      },
      "source": [
        "svc_clf.score(X_test_umap, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5990566037735849"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 61
        }
      ]
    }
  ]
}