{
  "id": 173111,
  "title": "What exactly is the 100k private training set?",
  "url": "/competitions/landmark-recognition-2020/discussion/173111",
  "author_name": "",
  "post_date": "2020-08-07T21:29:23.791563200Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>The data tab says \"To facilitate recognition-by-retrieval approaches, the private training set contains only a 100k subset of the total public training set.\" However when I \"Save &amp; Run All\" my notebook, it still sees 1.5M rows in train.csv. When is this number reduced to 100k?</p>",
  "messages": [
    {
      "id": "962163",
      "postDate": "08/07/2020 21:29:23",
      "content": "<p>The data tab says \"To facilitate recognition-by-retrieval approaches, the private training set contains only a 100k subset of the total public training set.\" However when I \"Save &amp; Run All\" my notebook, it still sees 1.5M rows in train.csv. When is this number reduced to 100k?</p>",
      "rawMarkdown": "The data tab says \"To facilitate recognition-by-retrieval approaches, the private training set contains only a 100k subset of the total public training set.\" However when I \"Save &amp; Run All\" my notebook, it still sees 1.5M rows in train.csv. When is this number reduced to 100k?",
      "votes": null
    },
    {
      "id": "962169",
      "postDate": "08/07/2020 21:43:04",
      "content": "<p>Ah I see, it seems like after we hit \"submit\" then our notebook is run yet again, presumably with the reduced set.</p>",
      "rawMarkdown": "Ah I see, it seems like after we hit \"submit\" then our notebook is run yet again, presumably with the reduced set.",
      "votes": null
    },
    {
      "id": "982184",
      "postDate": "08/23/2020 06:03:22",
      "content": "<p>What do you mean? Can you please explain this in detail? </p>",
      "rawMarkdown": "What do you mean? Can you please explain this in detail?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 962169,
      "author_name": "usmannkhan",
      "author_url": "",
      "post_date": "08/07/2020 21:43:04",
      "content": "<p>Ah I see, it seems like after we hit \"submit\" then our notebook is run yet again, presumably with the reduced set.</p>",
      "votes": null,
      "replies": [
        {
          "id": 982184,
          "author_name": "aroraaman",
          "author_url": "",
          "post_date": "08/23/2020 06:03:22",
          "content": "<p>What do you mean? Can you please explain this in detail? </p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "962163": "The data tab says \"To facilitate recognition-by-retrieval approaches, the private training set contains only a 100k subset of the total public training set.\" However when I \"Save &amp; Run All\" my notebook, it still sees 1.5M rows in train.csv. When is this number reduced to 100k?",
    "962169": "Ah I see, it seems like after we hit \"submit\" then our notebook is run yet again, presumably with the reduced set.",
    "982184": "What do you mean? Can you please explain this in detail?"
  },
  "source": "meta"
}