ecommerce_reverse_image_search/ecommerce-reverse-image-search.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nehaCVLCpO8e"
      },
      "source": [
        "# Ecommerce: reverse image search\n",
        "\n",
        "All e-commerce platforms need a search mechanism. The built-in methods usually rely on some variation of full-text search which finds the relevant documents based on the presence of the words used in a query. In some cases, it might be enough, but there are ways to improve that mechanism and increase sales. If your customer can easily find a product they need, they are more likely to buy it.\n",
        "\n",
        "Semantic search is one of the possibilities. It relies not only on keywords but considers the meaning and intention of the query. However, reverse image search might be a way to go if you want to enable non-textual search capabilities. Your customers may struggle to express themselves, so why don't you ease that and start accepting images as your search queries?\n",
        "\n",
        "# Amazon product dataset 2020\n",
        "\n",
        "We will use the [Amazon product dataset 2020](https://www.kaggle.com/datasets/promptcloud/amazon-product-dataset-2020/) and see how to enable visual queries for it. The following lines will download it from the cloud and create a directory structure so you can reproduce the results independently."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:08:43.494319Z",
          "start_time": "2023-07-20T09:08:42.607368Z"
        },
        "is_executing": true,
        "id": "fNppSO_PpO8f",
        "outputId": "74d1632a-c918-4adf-f906-baad844e6f80",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "mkdir: cannot create directory ‘data’: File exists\n",
            "mkdir: cannot create directory ‘data/images’: File exists\n",
            "mkdir: cannot create directory ‘queries’: File exists\n",
            "File ‘data/amazon-product-dataset-2020.zip’ already there; not retrieving.\n",
            "\n",
            "File ‘queries/ecommerce-reverse-image-search-queries.zip’ already there; not retrieving.\n",
            "\n",
            "Archive:  queries/ecommerce-reverse-image-search-queries.zip\n"
          ]
        }
      ],
      "source": [
        "!mkdir data\n",
        "!mkdir data/images\n",
        "!mkdir queries\n",
        "!wget -nc --directory-prefix=data/ \"https://storage.googleapis.com/qdrant-examples/amazon-product-dataset-2020.zip\"\n",
        "!wget -nc --directory-prefix=queries/ \"https://storage.googleapis.com/qdrant-examples/ecommerce-reverse-image-search-queries.zip\"\n",
        "!unzip -u -d queries queries/ecommerce-reverse-image-search-queries.zip"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:08:47.779869Z",
          "start_time": "2023-07-20T09:08:43.503915Z"
        },
        "is_executing": true,
        "scrolled": true,
        "id": "DeJbGicApO8g",
        "outputId": "684a4b5c-f018-48ee-dbdb-32a5e14c9544",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: jupyter in /usr/local/lib/python3.10/dist-packages (1.0.0)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (2.0.3)\n",
            "Requirement already satisfied: sentence_transformers in /usr/local/lib/python3.10/dist-packages (3.0.1)\n",
            "Collecting qdrant_client~=1.1.1\n",
            "  Downloading qdrant_client-1.1.7-py3-none-any.whl (127 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m127.8/127.8 kB\u001b[0m \u001b[31m1.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pyarrow in /usr/local/lib/python3.10/dist-packages (14.0.2)\n",
            "Requirement already satisfied: fastembed in /usr/local/lib/python3.10/dist-packages (0.3.3)\n",
            "Requirement already satisfied: notebook in /usr/local/lib/python3.10/dist-packages (from jupyter) (6.5.5)\n",
            "Requirement already satisfied: qtconsole in /usr/local/lib/python3.10/dist-packages (from jupyter) (5.5.2)\n",
            "Requirement already satisfied: jupyter-console in /usr/local/lib/python3.10/dist-packages (from jupyter) (6.1.0)\n",
            "Requirement already satisfied: nbconvert in /usr/local/lib/python3.10/dist-packages (from jupyter) (6.5.4)\n",
            "Requirement already satisfied: ipykernel in /usr/local/lib/python3.10/dist-packages (from jupyter) (5.5.6)\n",
            "Requirement already satisfied: ipywidgets in /usr/local/lib/python3.10/dist-packages (from jupyter) (7.7.1)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from pandas) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas) (2023.4)\n",
            "Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas) (2024.1)\n",
            "Requirement already satisfied: numpy>=1.21.0 in /usr/local/lib/python3.10/dist-packages (from pandas) (1.25.2)\n",
            "Requirement already satisfied: transformers<5.0.0,>=4.34.0 in /usr/local/lib/python3.10/dist-packages (from sentence_transformers) (4.41.2)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from sentence_transformers) (4.66.4)\n",
            "Requirement already satisfied: torch>=1.11.0 in /usr/local/lib/python3.10/dist-packages (from sentence_transformers) (2.3.0+cu121)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (from sentence_transformers) (1.2.2)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.10/dist-packages (from sentence_transformers) (1.11.4)\n",
            "Requirement already satisfied: huggingface-hub>=0.15.1 in /usr/local/lib/python3.10/dist-packages (from sentence_transformers) (0.23.4)\n",
            "Requirement already satisfied: Pillow in /usr/local/lib/python3.10/dist-packages (from sentence_transformers) (10.4.0)\n",
            "Requirement already satisfied: grpcio>=1.41.0 in /usr/local/lib/python3.10/dist-packages (from qdrant_client~=1.1.1) (1.64.1)\n",
            "Requirement already satisfied: grpcio-tools>=1.41.0 in /usr/local/lib/python3.10/dist-packages (from qdrant_client~=1.1.1) (1.64.1)\n",
            "Requirement already satisfied: httpx[http2]>=0.14.0 in /usr/local/lib/python3.10/dist-packages (from qdrant_client~=1.1.1) (0.27.0)\n",
            "Requirement already satisfied: portalocker<3.0.0,>=2.7.0 in /usr/local/lib/python3.10/dist-packages (from qdrant_client~=1.1.1) (2.10.0)\n",
            "Collecting pydantic<2.0,>=1.8 (from qdrant_client~=1.1.1)\n",
            "  Downloading pydantic-1.10.17-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.1 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m3.1/3.1 MB\u001b[0m \u001b[31m18.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: typing-extensions<5.0.0,>=4.0.0 in /usr/local/lib/python3.10/dist-packages (from qdrant_client~=1.1.1) (4.12.2)\n",
            "Requirement already satisfied: urllib3<2.0.0,>=1.26.14 in /usr/local/lib/python3.10/dist-packages (from qdrant_client~=1.1.1) (1.26.19)\n",
            "Requirement already satisfied: PyStemmer<3.0.0,>=2.2.0 in /usr/local/lib/python3.10/dist-packages (from fastembed) (2.2.0.1)\n",
            "Requirement already satisfied: loguru<0.8.0,>=0.7.2 in /usr/local/lib/python3.10/dist-packages (from fastembed) (0.7.2)\n",
            "Requirement already satisfied: mmh3<5.0,>=4.0 in /usr/local/lib/python3.10/dist-packages (from fastembed) (4.1.0)\n",
            "Requirement already satisfied: onnx<2.0.0,>=1.15.0 in /usr/local/lib/python3.10/dist-packages (from fastembed) (1.16.1)\n",
            "Requirement already satisfied: onnxruntime<2.0.0,>=1.17.0 in /usr/local/lib/python3.10/dist-packages (from fastembed) (1.18.1)\n",
            "Requirement already satisfied: requests<3.0,>=2.31 in /usr/local/lib/python3.10/dist-packages (from fastembed) (2.31.0)\n",
            "Requirement already satisfied: snowballstemmer<3.0.0,>=2.2.0 in /usr/local/lib/python3.10/dist-packages (from fastembed) (2.2.0)\n",
            "Requirement already satisfied: tokenizers<1.0,>=0.15 in /usr/local/lib/python3.10/dist-packages (from fastembed) (0.19.1)\n",
            "Requirement already satisfied: protobuf<6.0dev,>=5.26.1 in /usr/local/lib/python3.10/dist-packages (from grpcio-tools>=1.41.0->qdrant_client~=1.1.1) (5.27.2)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from grpcio-tools>=1.41.0->qdrant_client~=1.1.1) (67.7.2)\n",
            "Requirement already satisfied: anyio in /usr/local/lib/python3.10/dist-packages (from httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (3.7.1)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.10/dist-packages (from httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (2024.6.2)\n",
            "Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.10/dist-packages (from httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (1.0.5)\n",
            "Requirement already satisfied: idna in /usr/local/lib/python3.10/dist-packages (from httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (3.7)\n",
            "Requirement already satisfied: sniffio in /usr/local/lib/python3.10/dist-packages (from httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (1.3.1)\n",
            "Requirement already satisfied: h2<5,>=3 in /usr/local/lib/python3.10/dist-packages (from httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (4.1.0)\n",
            "Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.10/dist-packages (from httpcore==1.*->httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (0.14.0)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from huggingface-hub>=0.15.1->sentence_transformers) (3.15.4)\n",
            "Requirement already satisfied: fsspec>=2023.5.0 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub>=0.15.1->sentence_transformers) (2023.6.0)\n",
            "Requirement already satisfied: packaging>=20.9 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub>=0.15.1->sentence_transformers) (24.1)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub>=0.15.1->sentence_transformers) (6.0.1)\n",
            "Requirement already satisfied: coloredlogs in /usr/local/lib/python3.10/dist-packages (from onnxruntime<2.0.0,>=1.17.0->fastembed) (15.0.1)\n",
            "Requirement already satisfied: flatbuffers in /usr/local/lib/python3.10/dist-packages (from onnxruntime<2.0.0,>=1.17.0->fastembed) (24.3.25)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from onnxruntime<2.0.0,>=1.17.0->fastembed) (1.12.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.31->fastembed) (3.3.2)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (3.3)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (3.1.4)\n",
            "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (12.1.105)\n",
            "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (12.1.105)\n",
            "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (12.1.105)\n",
            "Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (8.9.2.26)\n",
            "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (12.1.3.1)\n",
            "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (11.0.2.54)\n",
            "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (10.3.2.106)\n",
            "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (11.4.5.107)\n",
            "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (12.1.0.106)\n",
            "Requirement already satisfied: nvidia-nccl-cu12==2.20.5 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (2.20.5)\n",
            "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (12.1.105)\n",
            "Requirement already satisfied: triton==2.3.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.11.0->sentence_transformers) (2.3.0)\n",
            "Requirement already satisfied: nvidia-nvjitlink-cu12 in /usr/local/lib/python3.10/dist-packages (from nvidia-cusolver-cu12==11.4.5.107->torch>=1.11.0->sentence_transformers) (12.5.82)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.10/dist-packages (from transformers<5.0.0,>=4.34.0->sentence_transformers) (2024.5.15)\n",
            "Requirement already satisfied: safetensors>=0.4.1 in /usr/local/lib/python3.10/dist-packages (from transformers<5.0.0,>=4.34.0->sentence_transformers) (0.4.3)\n",
            "Requirement already satisfied: ipython-genutils in /usr/local/lib/python3.10/dist-packages (from ipykernel->jupyter) (0.2.0)\n",
            "Requirement already satisfied: ipython>=5.0.0 in /usr/local/lib/python3.10/dist-packages (from ipykernel->jupyter) (7.34.0)\n",
            "Requirement already satisfied: traitlets>=4.1.0 in /usr/local/lib/python3.10/dist-packages (from ipykernel->jupyter) (5.7.1)\n",
            "Requirement already satisfied: jupyter-client in /usr/local/lib/python3.10/dist-packages (from ipykernel->jupyter) (6.1.12)\n",
            "Requirement already satisfied: tornado>=4.2 in /usr/local/lib/python3.10/dist-packages (from ipykernel->jupyter) (6.3.3)\n",
            "Requirement already satisfied: widgetsnbextension~=3.6.0 in /usr/local/lib/python3.10/dist-packages (from ipywidgets->jupyter) (3.6.6)\n",
            "Requirement already satisfied: jupyterlab-widgets>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from ipywidgets->jupyter) (3.0.11)\n",
            "Requirement already satisfied: prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from jupyter-console->jupyter) (3.0.47)\n",
            "Requirement already satisfied: pygments in /usr/local/lib/python3.10/dist-packages (from jupyter-console->jupyter) (2.16.1)\n",
            "Requirement already satisfied: lxml in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (4.9.4)\n",
            "Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (4.12.3)\n",
            "Requirement already satisfied: bleach in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (6.1.0)\n",
            "Requirement already satisfied: defusedxml in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (0.7.1)\n",
            "Requirement already satisfied: entrypoints>=0.2.2 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (0.4)\n",
            "Requirement already satisfied: jupyter-core>=4.7 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (5.7.2)\n",
            "Requirement already satisfied: jupyterlab-pygments in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (0.3.0)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (2.1.5)\n",
            "Requirement already satisfied: mistune<2,>=0.8.1 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (0.8.4)\n",
            "Requirement already satisfied: nbclient>=0.5.0 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (0.10.0)\n",
            "Requirement already satisfied: nbformat>=5.1 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (5.10.4)\n",
            "Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (1.5.1)\n",
            "Requirement already satisfied: tinycss2 in /usr/local/lib/python3.10/dist-packages (from nbconvert->jupyter) (1.3.0)\n",
            "Requirement already satisfied: pyzmq<25,>=17 in /usr/local/lib/python3.10/dist-packages (from notebook->jupyter) (24.0.1)\n",
            "Requirement already satisfied: argon2-cffi in /usr/local/lib/python3.10/dist-packages (from notebook->jupyter) (23.1.0)\n",
            "Requirement already satisfied: nest-asyncio>=1.5 in /usr/local/lib/python3.10/dist-packages (from notebook->jupyter) (1.6.0)\n",
            "Requirement already satisfied: Send2Trash>=1.8.0 in /usr/local/lib/python3.10/dist-packages (from notebook->jupyter) (1.8.3)\n",
            "Requirement already satisfied: terminado>=0.8.3 in /usr/local/lib/python3.10/dist-packages (from notebook->jupyter) (0.18.1)\n",
            "Requirement already satisfied: prometheus-client in /usr/local/lib/python3.10/dist-packages (from notebook->jupyter) (0.20.0)\n",
            "Requirement already satisfied: nbclassic>=0.4.7 in /usr/local/lib/python3.10/dist-packages (from notebook->jupyter) (1.1.0)\n",
            "Requirement already satisfied: qtpy>=2.4.0 in /usr/local/lib/python3.10/dist-packages (from qtconsole->jupyter) (2.4.1)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->sentence_transformers) (1.4.2)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->sentence_transformers) (3.5.0)\n",
            "Requirement already satisfied: hyperframe<7,>=6.0 in /usr/local/lib/python3.10/dist-packages (from h2<5,>=3->httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (6.0.1)\n",
            "Requirement already satisfied: hpack<5,>=4.0 in /usr/local/lib/python3.10/dist-packages (from h2<5,>=3->httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (4.0.0)\n",
            "Requirement already satisfied: jedi>=0.16 in /usr/local/lib/python3.10/dist-packages (from ipython>=5.0.0->ipykernel->jupyter) (0.19.1)\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.10/dist-packages (from ipython>=5.0.0->ipykernel->jupyter) (4.4.2)\n",
            "Requirement already satisfied: pickleshare in /usr/local/lib/python3.10/dist-packages (from ipython>=5.0.0->ipykernel->jupyter) (0.7.5)\n",
            "Requirement already satisfied: backcall in /usr/local/lib/python3.10/dist-packages (from ipython>=5.0.0->ipykernel->jupyter) (0.2.0)\n",
            "Requirement already satisfied: matplotlib-inline in /usr/local/lib/python3.10/dist-packages (from ipython>=5.0.0->ipykernel->jupyter) (0.1.7)\n",
            "Requirement already satisfied: pexpect>4.3 in /usr/local/lib/python3.10/dist-packages (from ipython>=5.0.0->ipykernel->jupyter) (4.9.0)\n",
            "Requirement already satisfied: platformdirs>=2.5 in /usr/local/lib/python3.10/dist-packages (from jupyter-core>=4.7->nbconvert->jupyter) (4.2.2)\n",
            "Requirement already satisfied: notebook-shim>=0.2.3 in /usr/local/lib/python3.10/dist-packages (from nbclassic>=0.4.7->notebook->jupyter) (0.2.4)\n",
            "Requirement already satisfied: fastjsonschema>=2.15 in /usr/local/lib/python3.10/dist-packages (from nbformat>=5.1->nbconvert->jupyter) (2.20.0)\n",
            "Requirement already satisfied: jsonschema>=2.6 in /usr/local/lib/python3.10/dist-packages (from nbformat>=5.1->nbconvert->jupyter) (4.19.2)\n",
            "Requirement already satisfied: wcwidth in /usr/local/lib/python3.10/dist-packages (from prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0->jupyter-console->jupyter) (0.2.13)\n",
            "Requirement already satisfied: ptyprocess in /usr/local/lib/python3.10/dist-packages (from terminado>=0.8.3->notebook->jupyter) (0.7.0)\n",
            "Requirement already satisfied: exceptiongroup in /usr/local/lib/python3.10/dist-packages (from anyio->httpx[http2]>=0.14.0->qdrant_client~=1.1.1) (1.2.1)\n",
            "Requirement already satisfied: argon2-cffi-bindings in /usr/local/lib/python3.10/dist-packages (from argon2-cffi->notebook->jupyter) (21.2.0)\n",
            "Requirement already satisfied: soupsieve>1.2 in /usr/local/lib/python3.10/dist-packages (from beautifulsoup4->nbconvert->jupyter) (2.5)\n",
            "Requirement already satisfied: webencodings in /usr/local/lib/python3.10/dist-packages (from bleach->nbconvert->jupyter) (0.5.1)\n",
            "Requirement already satisfied: humanfriendly>=9.1 in /usr/local/lib/python3.10/dist-packages (from coloredlogs->onnxruntime<2.0.0,>=1.17.0->fastembed) (10.0)\n",
            "Requirement already satisfied: mpmath<1.4.0,>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from sympy->onnxruntime<2.0.0,>=1.17.0->fastembed) (1.3.0)\n",
            "Requirement already satisfied: parso<0.9.0,>=0.8.3 in /usr/local/lib/python3.10/dist-packages (from jedi>=0.16->ipython>=5.0.0->ipykernel->jupyter) (0.8.4)\n",
            "Requirement already satisfied: attrs>=22.2.0 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=2.6->nbformat>=5.1->nbconvert->jupyter) (23.2.0)\n",
            "Requirement already satisfied: jsonschema-specifications>=2023.03.6 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=2.6->nbformat>=5.1->nbconvert->jupyter) (2023.12.1)\n",
            "Requirement already satisfied: referencing>=0.28.4 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=2.6->nbformat>=5.1->nbconvert->jupyter) (0.35.1)\n",
            "Requirement already satisfied: rpds-py>=0.7.1 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=2.6->nbformat>=5.1->nbconvert->jupyter) (0.18.1)\n",
            "Requirement already satisfied: jupyter-server<3,>=1.8 in /usr/local/lib/python3.10/dist-packages (from notebook-shim>=0.2.3->nbclassic>=0.4.7->notebook->jupyter) (1.24.0)\n",
            "Requirement already satisfied: cffi>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from argon2-cffi-bindings->argon2-cffi->notebook->jupyter) (1.16.0)\n",
            "Requirement already satisfied: pycparser in /usr/local/lib/python3.10/dist-packages (from cffi>=1.0.1->argon2-cffi-bindings->argon2-cffi->notebook->jupyter) (2.22)\n",
            "Requirement already satisfied: websocket-client in /usr/local/lib/python3.10/dist-packages (from jupyter-server<3,>=1.8->notebook-shim>=0.2.3->nbclassic>=0.4.7->notebook->jupyter) (1.8.0)\n",
            "Installing collected packages: pydantic, qdrant_client\n",
            "  Attempting uninstall: pydantic\n",
            "    Found existing installation: pydantic 2.8.0\n",
            "    Uninstalling pydantic-2.8.0:\n",
            "      Successfully uninstalled pydantic-2.8.0\n",
            "  Attempting uninstall: qdrant_client\n",
            "    Found existing installation: qdrant-client 1.3.2\n",
            "    Uninstalling qdrant-client-1.3.2:\n",
            "      Successfully uninstalled qdrant-client-1.3.2\n",
            "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
            "google-cloud-aiplatform 1.57.0 requires protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5, but you have protobuf 5.27.2 which is incompatible.\u001b[0m\u001b[31m\n",
            "\u001b[0mSuccessfully installed pydantic-1.10.17 qdrant_client-1.1.7\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "pydantic"
                ]
              },
              "id": "4838ad44a061409fb2cf9d76e56d8faf"
            }
          },
          "metadata": {}
        }
      ],
      "source": [
        "!pip install jupyter pandas sentence_transformers \"qdrant_client~=1.1.1\" pyarrow fastembed"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vC-SYcvupO8h"
      },
      "source": [
        "The dataset is provided as a CSV file and contains multiple attributes of the products, including URLs of the product images. That gives us a real case to work on. Let's check the dataset structure and prepare it for further processing.\n",
        "\n",
        "**For the testing purposes, we can use only a small subset of the dataset. It's enough to show the concept, but you can easily scale it to the whole dataset. The variable below is a fraction of the dataset that will be used for the rest of the notebook. Feel free to change it to `1.0` if you want to use the whole dataset.**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:54.881549Z",
          "start_time": "2023-07-20T09:11:54.875438Z"
        },
        "id": "n5qI1Mi4pO8h"
      },
      "outputs": [],
      "source": [
        "DATASET_FRACTION = 0.1"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "29eJiHVnpO8h"
      },
      "source": [
        "Now, we can load the dataset and see what it contains."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:55.716143Z",
          "start_time": "2023-07-20T09:11:54.885807Z"
        },
        "id": "e2w4jq6gpO8h",
        "outputId": "c27c62be-1160-4f97-cd31-4a782e87008b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                                    4248  \\\n",
              "Uniq Id                                 8fef5093eb479b4d6e57e9da43a7f2c6   \n",
              "Product Name           Eureka Classroom Supplies Learn to Count Count...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category                                                             NaN   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $10.36   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                      867470   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ASIN:B000T2YKIM|ShippingWeight:7.4ounces(Views...   \n",
              "Technical Details      Color:Animal Counters show up to 2 reviews by ...   \n",
              "Shipping Weight                                               7.4 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants               https://www.amazon.com/Eureka-Classroom-Suppli...   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Eureka-Classroom-Suppli...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "\n",
              "                                                                    5784  \\\n",
              "Uniq Id                                 95eae46a6d1531897b36b4533571e401   \n",
              "Product Name           Rockland Luggage 17 Inch Rolling Backpack, Ban...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Clothing, Shoes & Jewelry | Luggage & Travel G...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $28.00   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                 R01-BANDANA   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:13x10x17inches|ItemWeight:10...   \n",
              "Technical Details      Go to your orders and start the return Select ...   \n",
              "Shipping Weight                                               3.8 pounds   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants               https://www.amazon.com/Rockland-Luggage-Rollin...   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Rockland-Luggage-Rollin...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "\n",
              "                                                                    8839  \\\n",
              "Uniq Id                                 2bb3761b6fe3bbba43eca93ebe735678   \n",
              "Product Name           Losi High Endurance Clutch Bell, 14T: 2.0, LOS...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Toys & Games | Hobbies | Remote & App Controll...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $21.99   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                    LOSA9127   \n",
              "About Product          Losi High Endurance Clutch Bell, 14T: 2.0, LOS...   \n",
              "Product Specification  ProductDimensions:4x2.1x1.2inches|ItemWeight:0...   \n",
              "Technical Details      0.8 ounces (View shipping rates and policies) ...   \n",
              "Shipping Weight                                               0.8 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Team-Losi-High-Enduranc...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "\n",
              "                                                                    9128  \\\n",
              "Uniq Id                                 aa1ee02920a0175c5f924b95d58b83fe   \n",
              "Product Name           Hygloss Products Emoji Emoticon Stickers - 240...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category                         Toys & Games | Arts & Crafts | Stickers   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                      $8.17   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                        1896   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:7.8x4.6x0.1inches|ItemWeight...   \n",
              "Technical Details      Go to your orders and start the return Select ...   \n",
              "Shipping Weight                                               0.8 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants               https://www.amazon.com/Hygloss-Products-Emoji-...   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Hygloss-Products-Emoji-...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "\n",
              "                                                                    3981  \n",
              "Uniq Id                                 88165b349420bf28bdad83d297142e86  \n",
              "Product Name                   In the Breeze Thin Patriot Tails, 24-Inch  \n",
              "Brand Name                                                           NaN  \n",
              "Asin                                                                 NaN  \n",
              "Category               Toys & Games | Sports & Outdoor Play | Kites &...  \n",
              "Upc Ean Code                                                         NaN  \n",
              "List Price                                                           NaN  \n",
              "Selling Price                                                      $8.60  \n",
              "Quantity                                                             NaN  \n",
              "Model Number                                                         NaN  \n",
              "About Product          Make sure this fits by entering your model num...  \n",
              "Product Specification                                                NaN  \n",
              "Technical Details      Go to your orders and start the return Select ...  \n",
              "Shipping Weight                                                      NaN  \n",
              "Product Dimensions                                                   NaN  \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...  \n",
              "Variants               https://www.amazon.com/Breeze-Thin-Patriot-Tai...  \n",
              "Sku                                                                  NaN  \n",
              "Product Url            https://www.amazon.com/Breeze-Thin-Patriot-Tai...  \n",
              "Stock                                                                NaN  \n",
              "Product Details                                                      NaN  \n",
              "Dimensions                                                           NaN  \n",
              "Color                                                                NaN  \n",
              "Ingredients                                                          NaN  \n",
              "Direction To Use                                                     NaN  \n",
              "Is Amazon Seller                                                       Y  \n",
              "Size Quantity Variant                                                NaN  \n",
              "Product Description                                                  NaN  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-874e77c7-35bf-4c69-9c13-63a948f5607f\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>4248</th>\n",
              "      <th>5784</th>\n",
              "      <th>8839</th>\n",
              "      <th>9128</th>\n",
              "      <th>3981</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Uniq Id</th>\n",
              "      <td>8fef5093eb479b4d6e57e9da43a7f2c6</td>\n",
              "      <td>95eae46a6d1531897b36b4533571e401</td>\n",
              "      <td>2bb3761b6fe3bbba43eca93ebe735678</td>\n",
              "      <td>aa1ee02920a0175c5f924b95d58b83fe</td>\n",
              "      <td>88165b349420bf28bdad83d297142e86</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Name</th>\n",
              "      <td>Eureka Classroom Supplies Learn to Count Count...</td>\n",
              "      <td>Rockland Luggage 17 Inch Rolling Backpack, Ban...</td>\n",
              "      <td>Losi High Endurance Clutch Bell, 14T: 2.0, LOS...</td>\n",
              "      <td>Hygloss Products Emoji Emoticon Stickers - 240...</td>\n",
              "      <td>In the Breeze Thin Patriot Tails, 24-Inch</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Brand Name</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Asin</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Category</th>\n",
              "      <td>NaN</td>\n",
              "      <td>Clothing, Shoes &amp; Jewelry | Luggage &amp; Travel G...</td>\n",
              "      <td>Toys &amp; Games | Hobbies | Remote &amp; App Controll...</td>\n",
              "      <td>Toys &amp; Games | Arts &amp; Crafts | Stickers</td>\n",
              "      <td>Toys &amp; Games | Sports &amp; Outdoor Play | Kites &amp;...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Upc Ean Code</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>List Price</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Selling Price</th>\n",
              "      <td>$10.36</td>\n",
              "      <td>$28.00</td>\n",
              "      <td>$21.99</td>\n",
              "      <td>$8.17</td>\n",
              "      <td>$8.60</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Quantity</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Model Number</th>\n",
              "      <td>867470</td>\n",
              "      <td>R01-BANDANA</td>\n",
              "      <td>LOSA9127</td>\n",
              "      <td>1896</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>About Product</th>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Losi High Endurance Clutch Bell, 14T: 2.0, LOS...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Specification</th>\n",
              "      <td>ASIN:B000T2YKIM|ShippingWeight:7.4ounces(Views...</td>\n",
              "      <td>ProductDimensions:13x10x17inches|ItemWeight:10...</td>\n",
              "      <td>ProductDimensions:4x2.1x1.2inches|ItemWeight:0...</td>\n",
              "      <td>ProductDimensions:7.8x4.6x0.1inches|ItemWeight...</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Technical Details</th>\n",
              "      <td>Color:Animal Counters show up to 2 reviews by ...</td>\n",
              "      <td>Go to your orders and start the return Select ...</td>\n",
              "      <td>0.8 ounces (View shipping rates and policies) ...</td>\n",
              "      <td>Go to your orders and start the return Select ...</td>\n",
              "      <td>Go to your orders and start the return Select ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Shipping Weight</th>\n",
              "      <td>7.4 ounces</td>\n",
              "      <td>3.8 pounds</td>\n",
              "      <td>0.8 ounces</td>\n",
              "      <td>0.8 ounces</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Dimensions</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Image</th>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Variants</th>\n",
              "      <td>https://www.amazon.com/Eureka-Classroom-Suppli...</td>\n",
              "      <td>https://www.amazon.com/Rockland-Luggage-Rollin...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>https://www.amazon.com/Hygloss-Products-Emoji-...</td>\n",
              "      <td>https://www.amazon.com/Breeze-Thin-Patriot-Tai...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Sku</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Url</th>\n",
              "      <td>https://www.amazon.com/Eureka-Classroom-Suppli...</td>\n",
              "      <td>https://www.amazon.com/Rockland-Luggage-Rollin...</td>\n",
              "      <td>https://www.amazon.com/Team-Losi-High-Enduranc...</td>\n",
              "      <td>https://www.amazon.com/Hygloss-Products-Emoji-...</td>\n",
              "      <td>https://www.amazon.com/Breeze-Thin-Patriot-Tai...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Stock</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Details</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Dimensions</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Color</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Ingredients</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Direction To Use</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Is Amazon Seller</th>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Size Quantity Variant</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Description</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-874e77c7-35bf-4c69-9c13-63a948f5607f')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-874e77c7-35bf-4c69-9c13-63a948f5607f button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-874e77c7-35bf-4c69-9c13-63a948f5607f');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-1278c137-8089-4f9a-be7d-08b21acde8f4\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-1278c137-8089-4f9a-be7d-08b21acde8f4')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-1278c137-8089-4f9a-be7d-08b21acde8f4 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"dataset_df\",\n  \"rows\": 28,\n  \"fields\": [\n    {\n      \"column\": 4248,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 12,\n        \"samples\": [\n          \"https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B000T2YKIM\",\n          \"https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B000T2YKIM|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B002QVUIOO|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B0007P94Z0|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B000FA6DXS|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B00374J6H4|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B0007P94ZK|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B0007P94ZA|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B002PUN2T4|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B003D8G1AO|https://www.amazon.com/Eureka-Classroom-Supplies-Counting-Animals/dp/B000FA1TBO\",\n          \"8fef5093eb479b4d6e57e9da43a7f2c6\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 5784,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 13,\n        \"samples\": [\n          \"https://www.amazon.com/Rockland-Luggage-Rolling-Backpack-Bandana/dp/B005N2AWO4\",\n          \"https://images-na.ssl-images-amazon.com/images/I/51CbqK4WMjL.jpg|https://images-na.ssl-images-amazon.com/images/G/01/x-locale/common/transparent-pixel.jpg\",\n          \"95eae46a6d1531897b36b4533571e401\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 8839,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 11,\n        \"samples\": [\n          \"ProductDimensions:4x2.1x1.2inches|ItemWeight:0.8ounces|ShippingWeight:0.8ounces(Viewshippingratesandpolicies)|ASIN:B001P7DP40|Itemmodelnumber:LOSA9127|Manufacturerrecommendedage:14yearsandup\",\n          \"2bb3761b6fe3bbba43eca93ebe735678\",\n          \"https://www.amazon.com/Team-Losi-High-Endurance-Clutch/dp/B001P7DP40\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 9128,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 13,\n        \"samples\": [\n          \"https://www.amazon.com/Hygloss-Products-Emoji-Emoticon-Stickers/dp/B07864WMVJ\",\n          \"https://images-na.ssl-images-amazon.com/images/I/61MVbfDfgZL.jpg|https://images-na.ssl-images-amazon.com/images/G/01/x-locale/common/transparent-pixel.jpg\",\n          \"aa1ee02920a0175c5f924b95d58b83fe\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 3981,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 10,\n        \"samples\": [\n          \"https://www.amazon.com/Breeze-Thin-Patriot-Tails-24-Inch/dp/B00BEDWHEO\",\n          \"In the Breeze Thin Patriot Tails, 24-Inch\",\n          \"Go to your orders and start the return Select the ship method Ship it! | Go to your orders and start the return Select the ship method Ship it!\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 3
        }
      ],
      "source": [
        "import pandas as pd\n",
        "import zipfile\n",
        "\n",
        "with zipfile.ZipFile(\"./data/amazon-product-dataset-2020.zip\", \"r\") as z:\n",
        "    with z.open(\"home/sdf/marketing_sample_for_amazon_com-ecommerce__20200101_20200131__10k_data.csv\") as f:\n",
        "        dataset_df = pd.read_csv(f).sample(frac=DATASET_FRACTION)\n",
        "\n",
        "dataset_df.sample(n=5).T"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:55.721463Z",
          "start_time": "2023-07-20T09:11:55.717905Z"
        },
        "id": "u5cP-onfpO8h",
        "outputId": "7a1b314b-0906-43fc-c60d-36c1f6cd1084",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1000, 28)"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ],
      "source": [
        "dataset_df.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:55.734587Z",
          "start_time": "2023-07-20T09:11:55.723111Z"
        },
        "id": "lw0euUkCpO8h",
        "outputId": "363d35ab-0f05-4af1-b76d-044dc8a46a8e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 107
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'https://images-na.ssl-images-amazon.com/images/I/41SiG59Kt%2BL.jpg|https://images-na.ssl-images-amazon.com/images/I/51gvvBRjG5L.jpg|https://images-na.ssl-images-amazon.com/images/I/51WofS2WwHL.jpg|https://images-na.ssl-images-amazon.com/images/I/51Dk-9h2XEL.jpg|https://images-na.ssl-images-amazon.com/images/I/31tjBvZmVEL.jpg|https://images-na.ssl-images-amazon.com/images/I/31mH8F6HYyL.jpg|https://images-na.ssl-images-amazon.com/images/I/31DVHiEty%2BL.jpg|https://images-na.ssl-images-amazon.com/images/G/01/x-locale/common/transparent-pixel.jpg'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "dataset_df.iloc[0][\"Image\"]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i432zTv0pO8h"
      },
      "source": [
        "It turns out a single product may have several images. They are stored in a pipe-separated string."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:55.740435Z",
          "start_time": "2023-07-20T09:11:55.736888Z"
        },
        "id": "ZvbJ8vWtpO8i",
        "outputId": "bdacf60b-ce75-4847-c8ed-177d20d716b5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['https://images-na.ssl-images-amazon.com/images/I/41SiG59Kt%2BL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/51gvvBRjG5L.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/51WofS2WwHL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/51Dk-9h2XEL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/31tjBvZmVEL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/31mH8F6HYyL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/31DVHiEty%2BL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/G/01/x-locale/common/transparent-pixel.jpg']"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ],
      "source": [
        "dataset_df.iloc[0][\"Image\"].split(\"|\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BRNMnpSbpO8i"
      },
      "source": [
        "The last entry is common for multiple products, so we can simply remove it."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:55.746874Z",
          "start_time": "2023-07-20T09:11:55.742074Z"
        },
        "id": "0LLxtAw_pO8i"
      },
      "outputs": [],
      "source": [
        "dataset_df[\"Image\"] = dataset_df[\"Image\"].map(lambda x: x.split(\"|\")[:-1])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:55.755008Z",
          "start_time": "2023-07-20T09:11:55.749269Z"
        },
        "id": "6ARcbX6BpO8i",
        "outputId": "132b5537-f140-4e5f-f79f-e1a9d0ce1594",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['https://images-na.ssl-images-amazon.com/images/I/41SiG59Kt%2BL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/51gvvBRjG5L.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/51WofS2WwHL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/51Dk-9h2XEL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/31tjBvZmVEL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/31mH8F6HYyL.jpg',\n",
              " 'https://images-na.ssl-images-amazon.com/images/I/31DVHiEty%2BL.jpg']"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ],
      "source": [
        "dataset_df.iloc[0][\"Image\"]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:55.776871Z",
          "start_time": "2023-07-20T09:11:55.756290Z"
        },
        "id": "uinFw5wkpO8i",
        "outputId": "f1499903-c82a-4a69-96ae-f6acb9ae96c3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                                    7646  \\\n",
              "Uniq Id                                 c87f138612fe395f96266d0a1597faf7   \n",
              "Product Name           Amscan Dream Big Disposable Paper Dessert Plat...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Home & Kitchen | Event & Party Supplies | Tabl...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $12.89   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                      741377   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:7x7x2.3inches|ItemWeight:0.3...   \n",
              "Technical Details      Dream big graduate! Share them to your guests ...   \n",
              "Shipping Weight                                               9.6 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Disposable-Dessert-Grad...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "\n",
              "                                                                    1705  \\\n",
              "Uniq Id                                 a3daad923af839e1e0c4adf447f1b0d8   \n",
              "Product Name                                  12 Inch Trolls Girls' Bike   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Sports & Outdoors | Outdoor Recreation | Cycli...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $99.99   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                  8008-50ZTJ   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  Product Dimensions:         28.2 x 6.8 x 15 in...   \n",
              "Technical Details      Go to your orders and start the return Select ...   \n",
              "Shipping Weight                                              23.1 pounds   \n",
              "Product Dimensions                   28.2 x 6.8 x 15 inches  20.3 pounds   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Trolls-12-Inch-Girls-Bi...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "\n",
              "                                                                    1656  \\\n",
              "Uniq Id                                 925f24d67d03a19f33f83015c759c23a   \n",
              "Product Name           Home Dynamix Harper Ash Accent Rug, 48\" Round,...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Home & Kitchen | Home Décor | Kids' Room Décor...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $52.39   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                         NaN   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification                                                NaN   \n",
              "Technical Details      Go to your orders and start the return Select ...   \n",
              "Shipping Weight                                                      NaN   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants               https://www.amazon.com/Home-Dynamix-Harper-Acc...   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Home-Dynamix-Harper-Acc...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "\n",
              "                                                                    6934  \\\n",
              "Uniq Id                                 83177fad1c5979dc0a5b46bbb9580260   \n",
              "Product Name           FanWraps Star Wars Resistance Droids Passenger...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category                                                             NaN   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $19.99   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                     FW 1221   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:19.8x13.8x0.1inches|ItemWeig...   \n",
              "Technical Details      show up to 2 reviews by default Made from perf...   \n",
              "Shipping Weight                                                 5 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/FanWraps-Resistance-Dro...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "\n",
              "                                                                    2912  \n",
              "Uniq Id                                 78d3087cc7035ed78eeb102366a76e02  \n",
              "Product Name           MightySkins Skin Compatible with Hover-1 H1 Ho...  \n",
              "Brand Name                                                           NaN  \n",
              "Asin                                                                 NaN  \n",
              "Category               Sports & Outdoors | Outdoor Recreation | Skate...  \n",
              "Upc Ean Code                                                         NaN  \n",
              "List Price                                                           NaN  \n",
              "Selling Price                                             $9.86 - $11.86  \n",
              "Quantity                                                             NaN  \n",
              "Model Number                                       HOVH1-Laughing Skulls  \n",
              "About Product          SET THE TREND: Show off your own unique style ...  \n",
              "Product Specification  ProductDimensions:13x0.1x12inches|ItemWeight:0...  \n",
              "Technical Details      Do You Want Your Hover-1 Ultra Hover board Sco...  \n",
              "Shipping Weight                                              0.96 ounces  \n",
              "Product Dimensions                                                   NaN  \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...  \n",
              "Variants                                                             NaN  \n",
              "Sku                                                                  NaN  \n",
              "Product Url            https://www.amazon.com/MightySkins-Hover-1-Ult...  \n",
              "Stock                                                                NaN  \n",
              "Product Details                                                      NaN  \n",
              "Dimensions                                                           NaN  \n",
              "Color                                                                NaN  \n",
              "Ingredients                                                          NaN  \n",
              "Direction To Use                                                     NaN  \n",
              "Is Amazon Seller                                                       N  \n",
              "Size Quantity Variant                                                NaN  \n",
              "Product Description                                                  NaN  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-344fa27d-f9d7-418c-a03b-be445b7affe2\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>7646</th>\n",
              "      <th>1705</th>\n",
              "      <th>1656</th>\n",
              "      <th>6934</th>\n",
              "      <th>2912</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Uniq Id</th>\n",
              "      <td>c87f138612fe395f96266d0a1597faf7</td>\n",
              "      <td>a3daad923af839e1e0c4adf447f1b0d8</td>\n",
              "      <td>925f24d67d03a19f33f83015c759c23a</td>\n",
              "      <td>83177fad1c5979dc0a5b46bbb9580260</td>\n",
              "      <td>78d3087cc7035ed78eeb102366a76e02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Name</th>\n",
              "      <td>Amscan Dream Big Disposable Paper Dessert Plat...</td>\n",
              "      <td>12 Inch Trolls Girls' Bike</td>\n",
              "      <td>Home Dynamix Harper Ash Accent Rug, 48\" Round,...</td>\n",
              "      <td>FanWraps Star Wars Resistance Droids Passenger...</td>\n",
              "      <td>MightySkins Skin Compatible with Hover-1 H1 Ho...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Brand Name</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Asin</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Category</th>\n",
              "      <td>Home &amp; Kitchen | Event &amp; Party Supplies | Tabl...</td>\n",
              "      <td>Sports &amp; Outdoors | Outdoor Recreation | Cycli...</td>\n",
              "      <td>Home &amp; Kitchen | Home Décor | Kids' Room Décor...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Sports &amp; Outdoors | Outdoor Recreation | Skate...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Upc Ean Code</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>List Price</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Selling Price</th>\n",
              "      <td>$12.89</td>\n",
              "      <td>$99.99</td>\n",
              "      <td>$52.39</td>\n",
              "      <td>$19.99</td>\n",
              "      <td>$9.86 - $11.86</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Quantity</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Model Number</th>\n",
              "      <td>741377</td>\n",
              "      <td>8008-50ZTJ</td>\n",
              "      <td>NaN</td>\n",
              "      <td>FW 1221</td>\n",
              "      <td>HOVH1-Laughing Skulls</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>About Product</th>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>SET THE TREND: Show off your own unique style ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Specification</th>\n",
              "      <td>ProductDimensions:7x7x2.3inches|ItemWeight:0.3...</td>\n",
              "      <td>Product Dimensions:         28.2 x 6.8 x 15 in...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>ProductDimensions:19.8x13.8x0.1inches|ItemWeig...</td>\n",
              "      <td>ProductDimensions:13x0.1x12inches|ItemWeight:0...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Technical Details</th>\n",
              "      <td>Dream big graduate! Share them to your guests ...</td>\n",
              "      <td>Go to your orders and start the return Select ...</td>\n",
              "      <td>Go to your orders and start the return Select ...</td>\n",
              "      <td>show up to 2 reviews by default Made from perf...</td>\n",
              "      <td>Do You Want Your Hover-1 Ultra Hover board Sco...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Shipping Weight</th>\n",
              "      <td>9.6 ounces</td>\n",
              "      <td>23.1 pounds</td>\n",
              "      <td>NaN</td>\n",
              "      <td>5 ounces</td>\n",
              "      <td>0.96 ounces</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Dimensions</th>\n",
              "      <td>NaN</td>\n",
              "      <td>28.2 x 6.8 x 15 inches  20.3 pounds</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Image</th>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Variants</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>https://www.amazon.com/Home-Dynamix-Harper-Acc...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Sku</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Url</th>\n",
              "      <td>https://www.amazon.com/Disposable-Dessert-Grad...</td>\n",
              "      <td>https://www.amazon.com/Trolls-12-Inch-Girls-Bi...</td>\n",
              "      <td>https://www.amazon.com/Home-Dynamix-Harper-Acc...</td>\n",
              "      <td>https://www.amazon.com/FanWraps-Resistance-Dro...</td>\n",
              "      <td>https://www.amazon.com/MightySkins-Hover-1-Ult...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Stock</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Details</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Dimensions</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Color</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Ingredients</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Direction To Use</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Is Amazon Seller</th>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>N</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Size Quantity Variant</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Description</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-344fa27d-f9d7-418c-a03b-be445b7affe2')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-344fa27d-f9d7-418c-a03b-be445b7affe2 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-344fa27d-f9d7-418c-a03b-be445b7affe2');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-5dc0a224-9cda-4bd1-a4b4-54ab308a464f\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-5dc0a224-9cda-4bd1-a4b4-54ab308a464f')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-5dc0a224-9cda-4bd1-a4b4-54ab308a464f button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"dataset_df\",\n  \"rows\": 28,\n  \"fields\": [\n    {\n      \"column\": 7646,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 12,\n        \"samples\": [\n          \"https://www.amazon.com/Disposable-Dessert-Graduation-Tableware-Supplies/dp/B00I6LBJWS\",\n          \"https://images-na.ssl-images-amazon.com/images/I/51f1Bq5Mc3L.jpg\",\n          \"c87f138612fe395f96266d0a1597faf7\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 1705,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 13,\n        \"samples\": [\n          \"https://www.amazon.com/Trolls-12-Inch-Girls-Bike/dp/B01LVTSW4G\",\n          \"28.2 x 6.8 x 15 inches  20.3 pounds\",\n          \"a3daad923af839e1e0c4adf447f1b0d8\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 1656,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 10,\n        \"samples\": [\n          \"https://www.amazon.com/Home-Dynamix-Harper-Accent-Round/dp/B07WHDN4GH\",\n          \"Home Dynamix Harper Ash Accent Rug, 48\\\" Round, Gray\",\n          \"Go to your orders and start the return Select the ship method Ship it! | Go to your orders and start the return Select the ship method Ship it!\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 6934,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 11,\n        \"samples\": [\n          \"ProductDimensions:19.8x13.8x0.1inches|ItemWeight:5ounces|ShippingWeight:5ounces(Viewshippingratesandpolicies)|ASIN:B07NJMQ1W1|Itemmodelnumber:FW1221|Manufacturerrecommendedage:20-20years\",\n          \"83177fad1c5979dc0a5b46bbb9580260\",\n          \"https://www.amazon.com/FanWraps-Resistance-Droids-Passenger-Window/dp/B07NJMQ1W1\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 2912,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 12,\n        \"samples\": [\n          \"https://www.amazon.com/MightySkins-Hover-1-Ultra-Hoverboard-Scooter/dp/B07JQ7L86B\",\n          \"https://images-na.ssl-images-amazon.com/images/I/51ZW7IsvP1L.jpg\",\n          \"78d3087cc7035ed78eeb102366a76e02\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 9
        }
      ],
      "source": [
        "dataset_df = dataset_df.explode(\"Image\").dropna(subset=[\"Image\"])\n",
        "dataset_df.sample(n=5).T"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GE7Bte2VpO8j"
      },
      "source": [
        "## Downloading the images\n",
        "\n",
        "We want to create the embeddings out of the images, but we need to have them downloaded in the first place."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:11:55.784129Z",
          "start_time": "2023-07-20T09:11:55.778907Z"
        },
        "id": "5dZ3RShxpO8j"
      },
      "outputs": [],
      "source": [
        "from typing import Optional\n",
        "\n",
        "import urllib\n",
        "import os\n",
        "\n",
        "def download_file(url: str) -> Optional[str]:\n",
        "    basename = os.path.basename(url)\n",
        "    target_path = f\"./data/images/{basename}\"\n",
        "    if not os.path.exists(target_path):\n",
        "        try:\n",
        "            urllib.request.urlretrieve(url, target_path)\n",
        "        except urllib.error.HTTPError:\n",
        "            return None\n",
        "    return target_path"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:12:01.589455Z",
          "start_time": "2023-07-20T09:11:55.786290Z"
        },
        "id": "jGB-rlYRpO8j",
        "outputId": "978232af-b909-4536-9740-04907c05787a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                                    64    \\\n",
              "Uniq Id                                 b4358a38037a7e7fbcd7fe16970e7bff   \n",
              "Product Name           DC Collectibles DCTV: Black Lightning Resin St...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Toys & Games | Collectible Toys | Statues, Bob...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $17.75   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                   DEC170420   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:3x3x12.4inches|ItemWeight:2p...   \n",
              "Technical Details      show up to 2 reviews by default Jefferson Pier...   \n",
              "Shipping Weight                                               3.7 pounds   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/DC-Collectibles-DCTV-Li...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "LocalImage                                 ./data/images/41vJ5amvKeL.jpg   \n",
              "\n",
              "                                                                    2591  \\\n",
              "Uniq Id                                 853ae53d4c3dd833d07787f452e3c13a   \n",
              "Product Name           Click N' Play Set of 8 Kids Pretend Play Beaut...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Toys & Games | Dress Up & Pretend Play | Beaut...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $19.99   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                    CNP01008   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:12x12x2inches|ItemWeight:11....   \n",
              "Technical Details      Go to your orders and start the return Select ...   \n",
              "Shipping Weight                                              11.2 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Click-Play-Pretend-Hair...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "LocalImage                                 ./data/images/51sNsHAae4L.jpg   \n",
              "\n",
              "                                                                    3483  \\\n",
              "Uniq Id                                 cb7e81a2574c42628feb2ea88cc465c8   \n",
              "Product Name           Application DC Comics Originals Justice League...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category                                    Toys & Games | Arts & Crafts   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                      $6.99   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                 P-DC-0136-S   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:4.8x0x6inches|ItemWeight:0.8...   \n",
              "Technical Details      show up to 2 reviews by default C&D Visionary ...   \n",
              "Shipping Weight                                               0.8 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Application-Comics-Orig...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "LocalImage                                 ./data/images/61g66OGcaaL.jpg   \n",
              "\n",
              "                                                                    3045  \\\n",
              "Uniq Id                                 76cd3a9b8a05103b70fefe3766459d60   \n",
              "Product Name           Suck UK TV Lunch Box I Food Storage Containers...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Home & Kitchen | Kitchen & Dining | Storage & ...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $18.48   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                 SK LUNCHTV3   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:3.9x8.1x7.7inches|ItemWeight...   \n",
              "Technical Details      Go to your orders and start the return Select ...   \n",
              "Shipping Weight                                              15.5 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants               https://www.amazon.com/Suck-UK-Storage-Contain...   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Suck-UK-Storage-Contain...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "LocalImage                                 ./data/images/41EULQJ9LxL.jpg   \n",
              "\n",
              "                                                                    7589  \n",
              "Uniq Id                                 10801e1b6ef7e39eb8875234a49b7d70  \n",
              "Product Name           LEGO Disney Ariel’s Seaside Castle 41160 4+ Bu...  \n",
              "Brand Name                                                           NaN  \n",
              "Asin                                                                 NaN  \n",
              "Category                    Toys & Games | Building Toys | Building Sets  \n",
              "Upc Ean Code                                                         NaN  \n",
              "List Price                                                           NaN  \n",
              "Selling Price                                           $19.78 $ 19 . 78  \n",
              "Quantity                                                             NaN  \n",
              "Model Number                                                     6250992  \n",
              "About Product          Make sure this fits by entering your model num...  \n",
              "Product Specification  ProductDimensions:10.3x7.5x2.4inches|ItemWeigh...  \n",
              "Technical Details      show up to 2 reviews by default Let your child...  \n",
              "Shipping Weight                                              11.2 ounces  \n",
              "Product Dimensions                                                   NaN  \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...  \n",
              "Variants                                                             NaN  \n",
              "Sku                                                                  NaN  \n",
              "Product Url            https://www.amazon.com/LEGO-Disney-Ariels-Seas...  \n",
              "Stock                                                                NaN  \n",
              "Product Details                                                      NaN  \n",
              "Dimensions                                                           NaN  \n",
              "Color                                                                NaN  \n",
              "Ingredients                                                          NaN  \n",
              "Direction To Use                                                     NaN  \n",
              "Is Amazon Seller                                                       Y  \n",
              "Size Quantity Variant                                                NaN  \n",
              "Product Description                                                  NaN  \n",
              "LocalImage                                 ./data/images/41Np1paMBvL.jpg  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5fa5faa3-ba0b-4312-b0a1-a8154a42ddca\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>64</th>\n",
              "      <th>2591</th>\n",
              "      <th>3483</th>\n",
              "      <th>3045</th>\n",
              "      <th>7589</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Uniq Id</th>\n",
              "      <td>b4358a38037a7e7fbcd7fe16970e7bff</td>\n",
              "      <td>853ae53d4c3dd833d07787f452e3c13a</td>\n",
              "      <td>cb7e81a2574c42628feb2ea88cc465c8</td>\n",
              "      <td>76cd3a9b8a05103b70fefe3766459d60</td>\n",
              "      <td>10801e1b6ef7e39eb8875234a49b7d70</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Name</th>\n",
              "      <td>DC Collectibles DCTV: Black Lightning Resin St...</td>\n",
              "      <td>Click N' Play Set of 8 Kids Pretend Play Beaut...</td>\n",
              "      <td>Application DC Comics Originals Justice League...</td>\n",
              "      <td>Suck UK TV Lunch Box I Food Storage Containers...</td>\n",
              "      <td>LEGO Disney Ariel’s Seaside Castle 41160 4+ Bu...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Brand Name</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Asin</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Category</th>\n",
              "      <td>Toys &amp; Games | Collectible Toys | Statues, Bob...</td>\n",
              "      <td>Toys &amp; Games | Dress Up &amp; Pretend Play | Beaut...</td>\n",
              "      <td>Toys &amp; Games | Arts &amp; Crafts</td>\n",
              "      <td>Home &amp; Kitchen | Kitchen &amp; Dining | Storage &amp; ...</td>\n",
              "      <td>Toys &amp; Games | Building Toys | Building Sets</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Upc Ean Code</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>List Price</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Selling Price</th>\n",
              "      <td>$17.75</td>\n",
              "      <td>$19.99</td>\n",
              "      <td>$6.99</td>\n",
              "      <td>$18.48</td>\n",
              "      <td>$19.78 $ 19 . 78</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Quantity</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Model Number</th>\n",
              "      <td>DEC170420</td>\n",
              "      <td>CNP01008</td>\n",
              "      <td>P-DC-0136-S</td>\n",
              "      <td>SK LUNCHTV3</td>\n",
              "      <td>6250992</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>About Product</th>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Specification</th>\n",
              "      <td>ProductDimensions:3x3x12.4inches|ItemWeight:2p...</td>\n",
              "      <td>ProductDimensions:12x12x2inches|ItemWeight:11....</td>\n",
              "      <td>ProductDimensions:4.8x0x6inches|ItemWeight:0.8...</td>\n",
              "      <td>ProductDimensions:3.9x8.1x7.7inches|ItemWeight...</td>\n",
              "      <td>ProductDimensions:10.3x7.5x2.4inches|ItemWeigh...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Technical Details</th>\n",
              "      <td>show up to 2 reviews by default Jefferson Pier...</td>\n",
              "      <td>Go to your orders and start the return Select ...</td>\n",
              "      <td>show up to 2 reviews by default C&amp;D Visionary ...</td>\n",
              "      <td>Go to your orders and start the return Select ...</td>\n",
              "      <td>show up to 2 reviews by default Let your child...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Shipping Weight</th>\n",
              "      <td>3.7 pounds</td>\n",
              "      <td>11.2 ounces</td>\n",
              "      <td>0.8 ounces</td>\n",
              "      <td>15.5 ounces</td>\n",
              "      <td>11.2 ounces</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Dimensions</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Image</th>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Variants</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>https://www.amazon.com/Suck-UK-Storage-Contain...</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Sku</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Url</th>\n",
              "      <td>https://www.amazon.com/DC-Collectibles-DCTV-Li...</td>\n",
              "      <td>https://www.amazon.com/Click-Play-Pretend-Hair...</td>\n",
              "      <td>https://www.amazon.com/Application-Comics-Orig...</td>\n",
              "      <td>https://www.amazon.com/Suck-UK-Storage-Contain...</td>\n",
              "      <td>https://www.amazon.com/LEGO-Disney-Ariels-Seas...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Stock</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Details</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Dimensions</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Color</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Ingredients</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Direction To Use</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Is Amazon Seller</th>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Size Quantity Variant</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Description</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LocalImage</th>\n",
              "      <td>./data/images/41vJ5amvKeL.jpg</td>\n",
              "      <td>./data/images/51sNsHAae4L.jpg</td>\n",
              "      <td>./data/images/61g66OGcaaL.jpg</td>\n",
              "      <td>./data/images/41EULQJ9LxL.jpg</td>\n",
              "      <td>./data/images/41Np1paMBvL.jpg</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5fa5faa3-ba0b-4312-b0a1-a8154a42ddca')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-5fa5faa3-ba0b-4312-b0a1-a8154a42ddca button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-5fa5faa3-ba0b-4312-b0a1-a8154a42ddca');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-609642a7-3ada-4b9b-8e7d-7abd459ba343\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-609642a7-3ada-4b9b-8e7d-7abd459ba343')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-609642a7-3ada-4b9b-8e7d-7abd459ba343 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"dataset_df\",\n  \"rows\": 29,\n  \"fields\": [\n    {\n      \"column\": 64,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 13,\n        \"samples\": [\n          \"Y\",\n          \"https://images-na.ssl-images-amazon.com/images/I/41vJ5amvKeL.jpg\",\n          \"b4358a38037a7e7fbcd7fe16970e7bff\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 2591,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 13,\n        \"samples\": [\n          \"Y\",\n          \"https://images-na.ssl-images-amazon.com/images/I/51sNsHAae4L.jpg\",\n          \"853ae53d4c3dd833d07787f452e3c13a\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 3483,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 13,\n        \"samples\": [\n          \"Y\",\n          \"https://images-na.ssl-images-amazon.com/images/I/61g66OGcaaL.jpg\",\n          \"cb7e81a2574c42628feb2ea88cc465c8\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 3045,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 14,\n        \"samples\": [\n          \"https://images-na.ssl-images-amazon.com/images/I/41EULQJ9LxL.jpg\",\n          \"https://www.amazon.com/Suck-UK-Storage-Containers-Organization/dp/B07P8QPXQX\",\n          \"76cd3a9b8a05103b70fefe3766459d60\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 7589,\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 13,\n        \"samples\": [\n          \"Y\",\n          \"https://images-na.ssl-images-amazon.com/images/I/41Np1paMBvL.jpg\",\n          \"10801e1b6ef7e39eb8875234a49b7d70\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 11
        }
      ],
      "source": [
        "import numpy as np\n",
        "\n",
        "# Our download_file function returns None in case of any HTTP issues.\n",
        "# We can use that property to filter out the problematic images.\n",
        "dataset_df[\"LocalImage\"] = dataset_df[\"Image\"].map(download_file).replace({None: np.nan})\n",
        "dataset_df = dataset_df.dropna(subset=[\"LocalImage\"])\n",
        "dataset_df.sample(n=5).T"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-06-30T07:02:48.470302Z",
          "start_time": "2023-06-30T07:02:48.470222Z"
        },
        "id": "jVRtbJqPpO8k"
      },
      "source": [
        "## Creating the embeddings\n",
        "\n",
        "There are various options for creating the embeddings out of our images. But do not even think about training your neural encoder from scratch! Plenty of pre-trained models are available, and some may already give you some decent results in your domain. And if not, you can use them as a base for the fine-tuning that might be done way faster than the full training.\n",
        "\n",
        "### Available options\n",
        "\n",
        "Using the pretrained models is easy if you choose a library that exposes them with a convenient interface. Some of the possibilities are:\n",
        "\n",
        "- [torchvision](https://pytorch.org/vision/stable/index.html) - part of PyTorch\n",
        "- [embetter](https://koaning.github.io/embetter/) - if you prefer using pandas-like API, that's a great choice\n",
        "- [Sentence-Transformers](https://www.sbert.net/examples/applications/image-search/README.html) - one of the standard libraries for NLP exposes OpenAI CLIP model as well\n",
        "- [FastEmbed](https://github.com/qdrant/fastembed) - Light-weight, CPU-first library to generate vector embeddings using the [ONNX runtime](https://onnxruntime.ai/) by Qdrant.\n",
        "\n",
        "### Choosing the right model\n",
        "\n",
        "If you run an e-commerce business, you probably already have a standard full-text search mechanism. Reverse image search is one option to enrich the user experience, but if you also want to experiment with [hybrid search](https://qdrant.tech/articles/hybrid-search/), you should keep that in mind from the beginning. If that's your scenario, it's better to consider multimodality from day one. Such a model can encode texts and images in the same vector space.\n",
        "\n",
        "For that reason, we are going to use the OpenAI CLIP model, so in the future, we can extend our search mechanism with a semantic search using the same component."
      ]
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "vqNahytqrp_l"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:12:06.155891Z",
          "start_time": "2023-07-20T09:12:01.593671Z"
        },
        "id": "MEVCrXHUpO8k",
        "outputId": "7460a6e3-24bf-4b9b-bf2c-83e228042327",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "19cb8b64a289461e87e8bb51306d2d78",
            "0589ea29e2fc430dbf16bf29b7ae0cef",
            "fb1f79e2b2ba4bbd84e0643edd248022",
            "f35866309d3d47c6aac8cbb26b661309",
            "e576a7d25f814fd3888dd75737fc34e5",
            "0bcb9d85295f4954897c1486ad34ab79",
            "698a070e877543099578b9f236702fcd",
            "57246b79791c4735a98ef343e4ee74e1",
            "5d9c771f7a9949958583e2777b4e9689",
            "5a6e08f495514648a85fdf5a5c3baab1",
            "cd374bc44e8c45d9b4c2b62dbe77af93"
          ]
        }
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Fetching 3 files:   0%|          | 0/3 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "19cb8b64a289461e87e8bb51306d2d78"
            }
          },
          "metadata": {}
        }
      ],
      "source": [
        "from fastembed import ImageEmbedding\n",
        "\n",
        "model = ImageEmbedding(model_name=\"Qdrant/clip-ViT-B-32-vision\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:12:06.199167Z",
          "start_time": "2023-07-20T09:12:06.157390Z"
        },
        "id": "xi5N5fWipO8k",
        "outputId": "2db7a201-f305-41f1-e4c6-ec0e0529fb00",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 517
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=333x500>"
            ],
            "image/png": "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\n",
            "image/jpeg": "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\n"
          },
          "metadata": {},
          "execution_count": 15
        }
      ],
      "source": [
        "from PIL import Image\n",
        "\n",
        "image = Image.open(dataset_df.iloc[0][\"LocalImage\"])\n",
        "image"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:12:08.382103Z",
          "start_time": "2023-07-20T09:12:06.202918Z"
        },
        "id": "tGKQCG6YpO8k",
        "outputId": "e7c2249f-9c10-48b2-bd88-aa645865987a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(512,)"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ],
      "source": [
        "image_embedding = next(model.embed([dataset_df.iloc[0][\"LocalImage\"]]))\n",
        "image_embedding.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:12:08.391751Z",
          "start_time": "2023-07-20T09:12:08.383699Z"
        },
        "id": "Hv8ve8PWpO8k"
      },
      "outputs": [],
      "source": [
        "image_embedding"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1pp2qkZFpO8k"
      },
      "source": [
        "That's what a single embedding look like. But we need to calculate them for each image in our dataset."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:12:08.439328Z",
          "start_time": "2023-07-20T09:12:08.393561Z"
        },
        "id": "nx4dW9TppO8k"
      },
      "outputs": [],
      "source": [
        "from typing import List\n",
        "\n",
        "def calculate_embedding(image_path: str) -> Optional[List[float]]:\n",
        "    try:\n",
        "        return next(model.embed([image_path])).tolist()\n",
        "    except:\n",
        "        return None"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:14:43.715687Z",
          "start_time": "2023-07-20T09:12:08.442647Z"
        },
        "id": "J58BHKmJpO8k",
        "outputId": "d60c1a79-0925-4668-882c-860986c864fa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                                    8688  \\\n",
              "Uniq Id                                 4f870803642cd124be4b6f0c89e9d4bf   \n",
              "Product Name           Safescan ScanSafe Concierge Card Case with RFI...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category                               Toys & Games | Novelty & Gag Toys   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                      $8.32   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                    SCON-TEA   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  ProductDimensions:4.1x3x0.2inches|ItemWeight:0...   \n",
              "Technical Details      Color:Teal  |  Size:One Size show up to 2 revi...   \n",
              "Shipping Weight                                              0.64 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants               https://www.amazon.com/Scansafe-Concierge-Card...   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Scansafe-Concierge-Card...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "LocalImage                                 ./data/images/51Sws0v1GkL.jpg   \n",
              "Embedding              [-0.01141652837395668, 0.014521843753755093, 0...   \n",
              "\n",
              "                                                                    2453  \\\n",
              "Uniq Id                                 1b038ee56304c43ba0feb5e405c05c47   \n",
              "Product Name           Playgro 0184557 Bendy Ball for Baby Infant Tod...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Baby Products | Baby Care | Pacifiers, Teether...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $12.03   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                     0184557   \n",
              "About Product          Rattle sounds for baby's auditory stimulation ...   \n",
              "Product Specification                                                NaN   \n",
              "Technical Details      Style:Bendy Ball show up to 2 reviews by defau...   \n",
              "Shipping Weight                                               9.1 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Playgro-0184557-Childre...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "LocalImage                                 ./data/images/416vTIPTJDL.jpg   \n",
              "Embedding              [0.004719291348010302, 0.0180919598788023, -0....   \n",
              "\n",
              "                                                                    7846  \\\n",
              "Uniq Id                                 0b65e29108345f6d68e9eb18fd1ee020   \n",
              "Product Name           EPIKGO Sport Balance Board Self Balance Scoote...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Sports & Outdoors | Outdoor Recreation | Skate...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                    $199.00   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                EPIKGO Sport   \n",
              "About Product          Make sure this fits by entering your model num...   \n",
              "Product Specification  Product Dimensions:         31 x 12 x 12 inche...   \n",
              "Technical Details      Introducing the EPIKGO SPORTS Self-Balancing B...   \n",
              "Shipping Weight                                              41.2 pounds   \n",
              "Product Dimensions                        31 x 12 x 12 inches  33 pounds   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants                                                             NaN   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/EPIKGO-Sport-All-Terrai...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "LocalImage                                 ./data/images/51pCEz-LL0L.jpg   \n",
              "Embedding              [-0.01707524061203003, -0.008010529913008213, ...   \n",
              "\n",
              "                                                                    9192  \\\n",
              "Uniq Id                                 c4ebd4572a2a0710b3b074fd4302bd10   \n",
              "Product Name           Rubie's Costume Men's Batman Arkham City Adult...   \n",
              "Brand Name                                                           NaN   \n",
              "Asin                                                                 NaN   \n",
              "Category               Clothing, Shoes & Jewelry | Costumes & Accesso...   \n",
              "Upc Ean Code                                                         NaN   \n",
              "List Price                                                           NaN   \n",
              "Selling Price                                                     $38.00   \n",
              "Quantity                                                             NaN   \n",
              "Model Number                                                         NaN   \n",
              "About Product          100% Synthetic | Imported | Hand Wash | Printe...   \n",
              "Product Specification                                                NaN   \n",
              "Technical Details      With costumes designed for toddlers through pl...   \n",
              "Shipping Weight                                              10.7 ounces   \n",
              "Product Dimensions                                                   NaN   \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...   \n",
              "Variants               https://www.amazon.com/Rubies-Batman-Arkham-Mu...   \n",
              "Sku                                                                  NaN   \n",
              "Product Url            https://www.amazon.com/Rubies-Batman-Arkham-Mu...   \n",
              "Stock                                                                NaN   \n",
              "Product Details                                                      NaN   \n",
              "Dimensions                                                           NaN   \n",
              "Color                                                                NaN   \n",
              "Ingredients                                                          NaN   \n",
              "Direction To Use                                                     NaN   \n",
              "Is Amazon Seller                                                       Y   \n",
              "Size Quantity Variant                                                NaN   \n",
              "Product Description                                                  NaN   \n",
              "LocalImage                                 ./data/images/51p0gDolMyL.jpg   \n",
              "Embedding              [0.009645373560488224, 0.0459967665374279, 0.0...   \n",
              "\n",
              "                                                                    642   \n",
              "Uniq Id                                 198651a0f2df856435f4e7401d2329ca  \n",
              "Product Name                                        DJI Smart Controller  \n",
              "Brand Name                                                           NaN  \n",
              "Asin                                                                 NaN  \n",
              "Category               Toys & Games | Hobbies | Remote & App Controll...  \n",
              "Upc Ean Code                                                         NaN  \n",
              "List Price                                                           NaN  \n",
              "Selling Price                                                    $718.98  \n",
              "Quantity                                                             NaN  \n",
              "Model Number                                                         NaN  \n",
              "About Product          Make sure this fits by entering your model num...  \n",
              "Product Specification                                                NaN  \n",
              "Technical Details                                                    NaN  \n",
              "Shipping Weight                                                      NaN  \n",
              "Product Dimensions                                                   NaN  \n",
              "Image                  https://images-na.ssl-images-amazon.com/images...  \n",
              "Variants                                                             NaN  \n",
              "Sku                                                                  NaN  \n",
              "Product Url            https://www.amazon.com/DJI-CP-MA-00000080-01-S...  \n",
              "Stock                                                                NaN  \n",
              "Product Details                                                      NaN  \n",
              "Dimensions                                                           NaN  \n",
              "Color                                                                NaN  \n",
              "Ingredients                                                          NaN  \n",
              "Direction To Use                                                     NaN  \n",
              "Is Amazon Seller                                                       Y  \n",
              "Size Quantity Variant                                                NaN  \n",
              "Product Description                                                  NaN  \n",
              "LocalImage                                 ./data/images/41Y6SkFMnTL.jpg  \n",
              "Embedding              [0.017103899270296097, 0.009327375330030918, 0...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-fc636be5-2c7e-4800-9283-db3ff5359fb8\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>8688</th>\n",
              "      <th>2453</th>\n",
              "      <th>7846</th>\n",
              "      <th>9192</th>\n",
              "      <th>642</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Uniq Id</th>\n",
              "      <td>4f870803642cd124be4b6f0c89e9d4bf</td>\n",
              "      <td>1b038ee56304c43ba0feb5e405c05c47</td>\n",
              "      <td>0b65e29108345f6d68e9eb18fd1ee020</td>\n",
              "      <td>c4ebd4572a2a0710b3b074fd4302bd10</td>\n",
              "      <td>198651a0f2df856435f4e7401d2329ca</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Name</th>\n",
              "      <td>Safescan ScanSafe Concierge Card Case with RFI...</td>\n",
              "      <td>Playgro 0184557 Bendy Ball for Baby Infant Tod...</td>\n",
              "      <td>EPIKGO Sport Balance Board Self Balance Scoote...</td>\n",
              "      <td>Rubie's Costume Men's Batman Arkham City Adult...</td>\n",
              "      <td>DJI Smart Controller</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Brand Name</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Asin</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Category</th>\n",
              "      <td>Toys &amp; Games | Novelty &amp; Gag Toys</td>\n",
              "      <td>Baby Products | Baby Care | Pacifiers, Teether...</td>\n",
              "      <td>Sports &amp; Outdoors | Outdoor Recreation | Skate...</td>\n",
              "      <td>Clothing, Shoes &amp; Jewelry | Costumes &amp; Accesso...</td>\n",
              "      <td>Toys &amp; Games | Hobbies | Remote &amp; App Controll...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Upc Ean Code</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>List Price</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Selling Price</th>\n",
              "      <td>$8.32</td>\n",
              "      <td>$12.03</td>\n",
              "      <td>$199.00</td>\n",
              "      <td>$38.00</td>\n",
              "      <td>$718.98</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Quantity</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Model Number</th>\n",
              "      <td>SCON-TEA</td>\n",
              "      <td>0184557</td>\n",
              "      <td>EPIKGO Sport</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>About Product</th>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>Rattle sounds for baby's auditory stimulation ...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "      <td>100% Synthetic | Imported | Hand Wash | Printe...</td>\n",
              "      <td>Make sure this fits by entering your model num...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Specification</th>\n",
              "      <td>ProductDimensions:4.1x3x0.2inches|ItemWeight:0...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Product Dimensions:         31 x 12 x 12 inche...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Technical Details</th>\n",
              "      <td>Color:Teal  |  Size:One Size show up to 2 revi...</td>\n",
              "      <td>Style:Bendy Ball show up to 2 reviews by defau...</td>\n",
              "      <td>Introducing the EPIKGO SPORTS Self-Balancing B...</td>\n",
              "      <td>With costumes designed for toddlers through pl...</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Shipping Weight</th>\n",
              "      <td>0.64 ounces</td>\n",
              "      <td>9.1 ounces</td>\n",
              "      <td>41.2 pounds</td>\n",
              "      <td>10.7 ounces</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Dimensions</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>31 x 12 x 12 inches  33 pounds</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Image</th>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "      <td>https://images-na.ssl-images-amazon.com/images...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Variants</th>\n",
              "      <td>https://www.amazon.com/Scansafe-Concierge-Card...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>https://www.amazon.com/Rubies-Batman-Arkham-Mu...</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Sku</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Url</th>\n",
              "      <td>https://www.amazon.com/Scansafe-Concierge-Card...</td>\n",
              "      <td>https://www.amazon.com/Playgro-0184557-Childre...</td>\n",
              "      <td>https://www.amazon.com/EPIKGO-Sport-All-Terrai...</td>\n",
              "      <td>https://www.amazon.com/Rubies-Batman-Arkham-Mu...</td>\n",
              "      <td>https://www.amazon.com/DJI-CP-MA-00000080-01-S...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Stock</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Details</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Dimensions</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Color</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Ingredients</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Direction To Use</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Is Amazon Seller</th>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "      <td>Y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Size Quantity Variant</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Product Description</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LocalImage</th>\n",
              "      <td>./data/images/51Sws0v1GkL.jpg</td>\n",
              "      <td>./data/images/416vTIPTJDL.jpg</td>\n",
              "      <td>./data/images/51pCEz-LL0L.jpg</td>\n",
              "      <td>./data/images/51p0gDolMyL.jpg</td>\n",
              "      <td>./data/images/41Y6SkFMnTL.jpg</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Embedding</th>\n",
              "      <td>[-0.01141652837395668, 0.014521843753755093, 0...</td>\n",
              "      <td>[0.004719291348010302, 0.0180919598788023, -0....</td>\n",
              "      <td>[-0.01707524061203003, -0.008010529913008213, ...</td>\n",
              "      <td>[0.009645373560488224, 0.0459967665374279, 0.0...</td>\n",
              "      <td>[0.017103899270296097, 0.009327375330030918, 0...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fc636be5-2c7e-4800-9283-db3ff5359fb8')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-fc636be5-2c7e-4800-9283-db3ff5359fb8 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-fc636be5-2c7e-4800-9283-db3ff5359fb8');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-527214cd-e66c-4240-bac5-5a9629fc4f01\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-527214cd-e66c-4240-bac5-5a9629fc4f01')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-527214cd-e66c-4240-bac5-5a9629fc4f01 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"dataset_df\",\n  \"rows\": 30,\n  \"fields\": [\n    {\n      \"column\": 8688,\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 2453,\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 7846,\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 9192,\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": 642,\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 21
        }
      ],
      "source": [
        "# Again, our helper function returns None in case of any error, such as\n",
        "# unsupported image format. We need to remove those entries.\n",
        "dataset_df[\"Embedding\"] = dataset_df[\"LocalImage\"].map(calculate_embedding)\n",
        "dataset_df[\"Embedding\"] = dataset_df[\"Embedding\"].replace({None: np.nan})\n",
        "dataset_df = dataset_df.dropna(subset=[\"Embedding\"])\n",
        "dataset_df.sample(n=5).T"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:14:44.118177Z",
          "start_time": "2023-07-20T09:14:43.717833Z"
        },
        "id": "iP3SshfhpO8l"
      },
      "outputs": [],
      "source": [
        "dataset_df.to_parquet(\"./data/amazon-with-embeddings.parquet\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OMzcjyMcpO8l"
      },
      "source": [
        "## Indexing embeddings in Qdrant\n",
        "\n",
        "Reverse image search compares the embeddings of the image used as a query and the embeddings of the indexed pictures. That can be theoretically done in a naive way by comparing the query to every single item from our store, but that won't scale if we even go beyond a few hundred. That's what the vector search engines are designed for. Qdrant acts as a fast retrieval layer that performs an efficient search for the closest vectors in the space.\n",
        "\n",
        "There are various ways to start using Qdrant, and even though the local mode in Python SDK is possible, it should be running as a service in production. The easiest way is to use a Docker container, which we'll do."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:14:45.419964Z",
          "start_time": "2023-07-20T09:14:44.121336Z"
        },
        "id": "O8snTETvpO8l",
        "outputId": "1bf0ee05-0d92-4a95-ba27-92a05028a541",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/bin/bash: line 1: docker: command not found\n"
          ]
        }
      ],
      "source": [
        "!docker run -d -p \"6333:6333\" -p \"6334:6334\" --name \"reverse_image_search\" qdrant/qdrant:v1.10.1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:14:45.975516Z",
          "start_time": "2023-07-20T09:14:45.425849Z"
        },
        "id": "LMZPW6u_pO8l",
        "outputId": "f2f2185f-7b5b-4c16-a984-800176a2ce80",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CollectionsResponse(collections=[])"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ],
      "source": [
        "from qdrant_client import QdrantClient\n",
        "from qdrant_client.http import models as rest\n",
        "\n",
        "try:\n",
        "    client = QdrantClient(\"localhost\")\n",
        "    collections = client.get_collections()\n",
        "except Exception:\n",
        "    # Docker is unavailable in Google Colab so we switch to local\n",
        "    # mode available in Python SDK\n",
        "    client = QdrantClient(\":memory:\")\n",
        "    collections = client.get_collections()\n",
        "\n",
        "collections"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:14:46.478406Z",
          "start_time": "2023-07-20T09:14:45.978171Z"
        },
        "id": "5se9nCc-pO8l",
        "outputId": "031a5394-167a-4a7d-f0b4-1d4bd756a75a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ],
      "source": [
        "client.recreate_collection(\n",
        "    collection_name=\"amazon\",\n",
        "    vectors_config=rest.VectorParams(\n",
        "        size=512,\n",
        "        distance=rest.Distance.COSINE,\n",
        "    )\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "11c7ZfPApO8l"
      },
      "source": [
        "It's a good practice to use batching while inserting the vectors into the collection. Python SDK has a utility method that performs it automatically. For the purposes of our demo, we're going to store vectors with the product id, name, and description as a payload."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 37,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:14:46.513789Z",
          "start_time": "2023-07-20T09:14:46.479733Z"
        },
        "id": "FkC1B8ZCpO8l",
        "outputId": "3ac401e8-33e5-4552-bce1-94e38d9affe8",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'ID': 'ec13f3287da541a31f72f5a047ec2e36',\n",
              " 'Name': 'Btswim NFL Pool Noodles (Pack of 3)',\n",
              " 'Description': 'Make sure this fits by entering your model number. | Includes: 3 high quality pool noodles measuring 57\" x 3\" | Removable and re-washable spandex-like cover | 3 unique styles per team',\n",
              " 'Image': 'https://images-na.ssl-images-amazon.com/images/I/41SiG59Kt%2BL.jpg',\n",
              " 'Path': './data/images/41SiG59Kt%2BL.jpg'}"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ],
      "source": [
        "payloads = dataset_df[[\"Uniq Id\", \"Product Name\", \"About Product\", \"Image\", \"LocalImage\"]] \\\n",
        "    .fillna(\"Unknown\") \\\n",
        "    .rename(columns={\"Uniq Id\": \"ID\",\n",
        "                     \"Product Name\": \"Name\",\n",
        "                     \"About Product\": \"Description\",\n",
        "                     \"LocalImage\": \"Path\"}) \\\n",
        "    .to_dict(\"records\")\n",
        "payloads[0]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:15:11.503216Z",
          "start_time": "2023-07-20T09:14:46.515774Z"
        },
        "scrolled": true,
        "id": "oHpTQsiUpO8l"
      },
      "outputs": [],
      "source": [
        "import uuid\n",
        "\n",
        "client.upload_collection(\n",
        "    collection_name=\"amazon\",\n",
        "    vectors=list(map(list, dataset_df[\"Embedding\"].tolist())),\n",
        "    payload=payloads,\n",
        "    ids=[uuid.uuid4().hex for _ in payloads],\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:15:11.511892Z",
          "start_time": "2023-07-20T09:15:11.504665Z"
        },
        "id": "HsIeYe0MpO8m",
        "outputId": "15cdeca8-b29b-4ef3-ffdd-8ca88081ab0a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CountResult(count=3359)"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ],
      "source": [
        "client.count(\"amazon\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "62spedwcpO8m"
      },
      "source": [
        "## Running the reverse image search\n",
        "\n",
        "As soon as we have the data indexed in Qdrant, it may already start acting as our reverse image search mechanism. Our queries no longer can be just textual, but we can freely use images to find similar items. For that, we surely need a query, and that will rarely be an image from the dataset. Let's find some different examples - for example from [Unsplash](https://unsplash.com), which is a source of freely usable images."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:15:11.584322Z",
          "start_time": "2023-07-20T09:15:11.513583Z"
        },
        "id": "1Oqxt3XcpO8m"
      },
      "outputs": [],
      "source": [
        "from io import BytesIO\n",
        "\n",
        "import base64\n",
        "\n",
        "def pillow_image_to_base64(image: Image) -> str:\n",
        "    \"\"\"\n",
        "    Convert a Pillow image to a base64 encoded string that can be used as an image\n",
        "    source in HTML.\n",
        "    :param image:\n",
        "    :return:\n",
        "    \"\"\"\n",
        "    buffered = BytesIO()\n",
        "    image.save(buffered, format=\"JPEG\")\n",
        "    img_str = base64.b64encode(buffered.getvalue()).decode(\"utf-8\")\n",
        "    return f\"data:image/jpeg;base64,{img_str}\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:15:11.614662Z",
          "start_time": "2023-07-20T09:15:11.589188Z"
        },
        "scrolled": false,
        "id": "h-hbjDBwpO8m",
        "outputId": "46130885-5922-443e-c6ae-170af00cb9b8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        }
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<table><tr><td><img src='data:image/jpeg;base64,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' width=150></td><td><img src='data:image/jpeg;base64,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' width=150></td><td><img src='data:image/jpeg;base64,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' width=150></td><td><img src='data:image/jpeg;base64,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' width=150></td></tr></table>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "from IPython.display import display, HTML\n",
        "\n",
        "import glob\n",
        "\n",
        "query_image_paths = list(glob.glob(\"./queries/*.jpg\"))\n",
        "images_html = \"\".join(\n",
        "    f\"<td><img src='{pillow_image_to_base64(Image.open(path))}' width=150></td>\"\n",
        "    for path in query_image_paths\n",
        ")\n",
        "display(HTML(f\"<table><tr>{images_html}</tr></table>\"))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:15:11.934081Z",
          "start_time": "2023-07-20T09:15:11.616193Z"
        },
        "scrolled": false,
        "id": "D6gRHuZ1pO8m"
      },
      "outputs": [],
      "source": [
        "for query_image_path in query_image_paths:\n",
        "    query_embedding = next(model.embed(query_image_path)).tolist()\n",
        "\n",
        "    results = client.query_points(\n",
        "        collection_name=\"amazon\",\n",
        "        query=query_embedding,\n",
        "        with_payload=True,\n",
        "        limit=5,\n",
        "    ).points\n",
        "\n",
        "    output_images = [\n",
        "        pillow_image_to_base64(Image.open(query_image_path)),\n",
        "    ]\n",
        "    for result in results:\n",
        "        output_images.append(result.payload[\"Image\"])\n",
        "\n",
        "    images_html = \"\".join(\n",
        "        f\"<td><img src='{path}' width=150></td>\"\n",
        "        for path in output_images\n",
        "    )\n",
        "    display(HTML(f\"<table><tr>{images_html}</tr></table>\"))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dyqima0GpO8q"
      },
      "source": [
        "We've implemented a reverse image search mechanism for e-commerce within a single notebook. We can kill the running Docker container for now, so nothing is left dangling in our environment."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "ExecuteTime": {
          "end_time": "2023-07-20T09:15:13.850345Z",
          "start_time": "2023-07-20T09:15:11.938136Z"
        },
        "id": "R23qXlxtpO8q",
        "outputId": "08f85034-47d5-485c-f183-1388989a9412"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "reverse_image_search\n",
            "reverse_image_search\n"
          ]
        }
      ],
      "source": [
        "!docker kill reverse_image_search\n",
        "!docker rm reverse_image_search"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wYn2XFtJpO8q"
      },
      "source": [
        "## Futher steps\n",
        "\n",
        "The notebook shows the general pipeline of encoding the inventory and using Qdrant to perform the reverse image search. There are, however, some challenges you may encounter while trying to implement it in the real world:\n",
        "\n",
        "1. Pretrained models are great to start with but may struggle for some specific kinds of inventory if not trained on similar examples. You can always fine-tune them with small amounts of data to avoid a full training cycle.\n",
        "2. Models should not be hosted within Jupyter notebooks, but there are some ways to serve them efficiently. We're going to describe the possibilities in a separate tutorial.\n",
        "3. If you don't want to worry about maintaining another system in your stack, please consider using [Qdrant Cloud](https://cloud.qdrant.io/), our managed solution. Our tier is free forever and available to everyone - no credit card is required."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "uNs4ySESpO8q"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3 (ipykernel)",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.10.6"
    },
    "colab": {
      "provenance": []
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "19cb8b64a289461e87e8bb51306d2d78": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_0589ea29e2fc430dbf16bf29b7ae0cef",
              "IPY_MODEL_fb1f79e2b2ba4bbd84e0643edd248022",
              "IPY_MODEL_f35866309d3d47c6aac8cbb26b661309"
            ],
            "layout": "IPY_MODEL_e576a7d25f814fd3888dd75737fc34e5"
          }
        },
        "0589ea29e2fc430dbf16bf29b7ae0cef": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0bcb9d85295f4954897c1486ad34ab79",
            "placeholder": "​",
            "style": "IPY_MODEL_698a070e877543099578b9f236702fcd",
            "value": "Fetching 3 files: 100%"
          }
        },
        "fb1f79e2b2ba4bbd84e0643edd248022": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_57246b79791c4735a98ef343e4ee74e1",
            "max": 3,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_5d9c771f7a9949958583e2777b4e9689",
            "value": 3
          }
        },
        "f35866309d3d47c6aac8cbb26b661309": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_5a6e08f495514648a85fdf5a5c3baab1",
            "placeholder": "​",
            "style": "IPY_MODEL_cd374bc44e8c45d9b4c2b62dbe77af93",
            "value": " 3/3 [00:00&lt;00:00, 95.92it/s]"
          }
        },
        "e576a7d25f814fd3888dd75737fc34e5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0bcb9d85295f4954897c1486ad34ab79": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "698a070e877543099578b9f236702fcd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "57246b79791c4735a98ef343e4ee74e1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "5d9c771f7a9949958583e2777b4e9689": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "5a6e08f495514648a85fdf5a5c3baab1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cd374bc44e8c45d9b4c2b62dbe77af93": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}