{
  "id": 306720,
  "title": "Should we use the image data? if So how should we use it?",
  "url": "/competitions/h-and-m-personalized-fashion-recommendations/discussion/306720",
  "author_name": "",
  "post_date": "2022-02-10T14:50:57.803309700Z",
  "votes": 3,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Should we store images as pixels and should we extract any features from it and create a new dataframe ? How to use this for prediction?</p>",
  "messages": [
    {
      "id": "1684516",
      "postDate": "02/10/2022 14:50:57",
      "content": "<p>Should we store images as pixels and should we extract any features from it and create a new dataframe ? How to use this for prediction?</p>",
      "rawMarkdown": "Should we store images as pixels and should we extract any features from it and create a new dataframe ? How to use this for prediction?",
      "votes": null
    },
    {
      "id": "1684643",
      "postDate": "02/10/2022 16:06:45",
      "content": "<p>One potential way to use them is to to find products with similar images to the products a customer bought, and assume it's more likely a customer would buy one of those.</p>\n<p>In shoppee competition notebooks you can find code to find similar images.</p>",
      "rawMarkdown": "One potential way to use them is to to find products with similar images to the products a customer bought, and assume it's more likely a customer would buy one of those.\n\nIn shoppee competition notebooks you can find code to find similar images.",
      "votes": null
    },
    {
      "id": "1685047",
      "postDate": "02/11/2022 01:30:21",
      "content": "<p>That helps, thank you😊</p>",
      "rawMarkdown": "That helps, thank you😊",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1684643,
      "author_name": "jacob34",
      "author_url": "",
      "post_date": "02/10/2022 16:06:45",
      "content": "<p>One potential way to use them is to to find products with similar images to the products a customer bought, and assume it's more likely a customer would buy one of those.</p>\n<p>In shoppee competition notebooks you can find code to find similar images.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1685047,
          "author_name": "chandanagopal2712",
          "author_url": "",
          "post_date": "02/11/2022 01:30:21",
          "content": "<p>That helps, thank you😊</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1684516": "Should we store images as pixels and should we extract any features from it and create a new dataframe ? How to use this for prediction?",
    "1684643": "One potential way to use them is to to find products with similar images to the products a customer bought, and assume it's more likely a customer would buy one of those.\n\nIn shoppee competition notebooks you can find code to find similar images.",
    "1685047": "That helps, thank you😊"
  },
  "source": "meta"
}