{
  "id": 312292,
  "title": "Customers that buy an article more than once and evaluation metric MAP@12",
  "url": "/competitions/h-and-m-personalized-fashion-recommendations/discussion/312292",
  "author_name": "",
  "post_date": "2022-03-11T08:26:01.611976100Z",
  "votes": 4,
  "comment_count": 1,
  "views": 0,
  "content": "<p>In the sales transactions data, repeated purchases of the same article by a customer on the same days are represented by several rows.</p>\n<p>Has anyone checked, how repeated purchases affect the evaluation metric? Would it make sense to predict the same article more than once for a single customer?</p>",
  "messages": [
    {
      "id": "1718869",
      "postDate": "03/11/2022 08:26:01",
      "content": "<p>In the sales transactions data, repeated purchases of the same article by a customer on the same days are represented by several rows.</p>\n<p>Has anyone checked, how repeated purchases affect the evaluation metric? Would it make sense to predict the same article more than once for a single customer?</p>",
      "rawMarkdown": "In the sales transactions data, repeated purchases of the same article by a customer on the same days are represented by several rows.\n\nHas anyone checked, how repeated purchases affect the evaluation metric? Would it make sense to predict the same article more than once for a single customer?",
      "votes": null
    },
    {
      "id": "1719514",
      "postDate": "03/11/2022 20:34:45",
      "content": "<p>I had the same thought, but it will hurt competition metric because the metric code ignores duplicate predictions. See code in GitHub <a href=\"https://github.com/benhamner/Metrics/blob/9a637aea795dc6f2333f022b0863398de0a1ca77/Python/ml_metrics/average_precision.py#L32\" target=\"_blank\">here</a></p>",
      "rawMarkdown": "I had the same thought, but it will hurt competition metric because the metric code ignores duplicate predictions. See code in GitHub [here][1]\n\n[1]: https://github.com/benhamner/Metrics/blob/9a637aea795dc6f2333f022b0863398de0a1ca77/Python/ml_metrics/average_precision.py#L32",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1719514,
      "author_name": "cdeotte",
      "author_url": "",
      "post_date": "03/11/2022 20:34:45",
      "content": "<p>I had the same thought, but it will hurt competition metric because the metric code ignores duplicate predictions. See code in GitHub <a href=\"https://github.com/benhamner/Metrics/blob/9a637aea795dc6f2333f022b0863398de0a1ca77/Python/ml_metrics/average_precision.py#L32\" target=\"_blank\">here</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1718869": "In the sales transactions data, repeated purchases of the same article by a customer on the same days are represented by several rows.\n\nHas anyone checked, how repeated purchases affect the evaluation metric? Would it make sense to predict the same article more than once for a single customer?",
    "1719514": "I had the same thought, but it will hurt competition metric because the metric code ignores duplicate predictions. See code in GitHub [here][1]\n\n[1]: https://github.com/benhamner/Metrics/blob/9a637aea795dc6f2333f022b0863398de0a1ca77/Python/ml_metrics/average_precision.py#L32"
  },
  "source": "meta"
}