{
  "id": 202764,
  "title": "Does anyone have a sense of the score of the labels?",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/202764",
  "author_name": "Alexander Soare",
  "post_date": "2020-12-11T21:47:13.683000",
  "votes": 2,
  "comment_count": 0,
  "views": 0,
  "content": "<p>There's a lot of talk about label noise. Wondering if anyone has looked enough into this to get a sense of what the labels score in terms of the competition metric when compared to the \"true truth\". On an assumption that the label noise is random, this is basically our upper bound for the lb.</p>",
  "messages": [
    {
      "id": 1109603,
      "postDate": "2020-12-11T21:47:13.683Z",
      "content": "<p>There's a lot of talk about label noise. Wondering if anyone has looked enough into this to get a sense of what the labels score in terms of the competition metric when compared to the \"true truth\". On an assumption that the label noise is random, this is basically our upper bound for the lb.</p>",
      "rawMarkdown": "There's a lot of talk about label noise. Wondering if anyone has looked enough into this to get a sense of what the labels score in terms of the competition metric when compared to the \"true truth\". On an assumption that the label noise is random, this is basically our upper bound for the lb.\n\n",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1109603": "There's a lot of talk about label noise. Wondering if anyone has looked enough into this to get a sense of what the labels score in terms of the competition metric when compared to the \"true truth\". On an assumption that the label noise is random, this is basically our upper bound for the lb.\n\n"
  }
}