{
  "id": 256056,
  "title": "The hand label on the train is not a problem, right?",
  "url": "/competitions/google-smartphone-decimeter-challenge/discussion/256056",
  "author_name": "",
  "post_date": "2021-07-30T15:31:03.552230Z",
  "votes": 9,
  "comment_count": 2,
  "views": 0,
  "content": "<p>In <a href=\"https://www.kaggle.com/columbia2131/area-knn-prediction-train-hand-label\" target=\"_blank\">this notebook</a>, I did hand labeling on the train data, and then using machine learning to predict the label for the test.<br>\nThis hand labeling of train data is not prohibited, so I see no problem with it.  However I am not sure, so cloud you tell me know if there is a problem or not, <a href=\"https://www.kaggle.com/sohier\" target=\"_blank\">@sohier</a> ?</p>",
  "messages": [
    {
      "id": "1405215",
      "postDate": "07/30/2021 15:31:03",
      "content": "<p>In <a href=\"https://www.kaggle.com/columbia2131/area-knn-prediction-train-hand-label\" target=\"_blank\">this notebook</a>, I did hand labeling on the train data, and then using machine learning to predict the label for the test.<br>\nThis hand labeling of train data is not prohibited, so I see no problem with it.  However I am not sure, so cloud you tell me know if there is a problem or not, <a href=\"https://www.kaggle.com/sohier\" target=\"_blank\">@sohier</a> ?</p>",
      "rawMarkdown": "In [this notebook](https://www.kaggle.com/columbia2131/area-knn-prediction-train-hand-label), I did hand labeling on the train data, and then using machine learning to predict the label for the test.\nThis hand labeling of train data is not prohibited, so I see no problem with it.  However I am not sure, so cloud you tell me know if there is a problem or not, @sohier ?",
      "votes": null
    },
    {
      "id": "1405451",
      "postDate": "07/30/2021 21:19:27",
      "content": "<p>Re-labeling the training set by hand is not a problem. Hand-labeling of any part of the test set is prohibited by the rules.</p>",
      "rawMarkdown": "Re-labeling the training set by hand is not a problem. Hand-labeling of any part of the test set is prohibited by the rules.",
      "votes": null
    },
    {
      "id": "1405524",
      "postDate": "07/31/2021 01:54:43",
      "content": "<p>Thank you for the quick response!</p>",
      "rawMarkdown": "Thank you for the quick response!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1405451,
      "author_name": "juliaelliott",
      "author_url": "",
      "post_date": "07/30/2021 21:19:27",
      "content": "<p>Re-labeling the training set by hand is not a problem. Hand-labeling of any part of the test set is prohibited by the rules.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1405524,
          "author_name": "columbia2131",
          "author_url": "",
          "post_date": "07/31/2021 01:54:43",
          "content": "<p>Thank you for the quick response!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1405215": "In [this notebook](https://www.kaggle.com/columbia2131/area-knn-prediction-train-hand-label), I did hand labeling on the train data, and then using machine learning to predict the label for the test.\nThis hand labeling of train data is not prohibited, so I see no problem with it.  However I am not sure, so cloud you tell me know if there is a problem or not, @sohier ?",
    "1405451": "Re-labeling the training set by hand is not a problem. Hand-labeling of any part of the test set is prohibited by the rules.",
    "1405524": "Thank you for the quick response!"
  },
  "source": "meta"
}