{
  "id": 201620,
  "title": "What is the competition metric?",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/201620",
  "author_name": "",
  "post_date": "2020-12-05T20:00:19.929756800Z",
  "votes": null,
  "comment_count": 3,
  "views": 0,
  "content": "<p><a href=\"https://www.kaggle.com/juliaelliott\" target=\"_blank\">@juliaelliott</a>  In the evaluation page, the competition metric is given as <code>categorization accuracy</code> with a link to a binary case. I'm slightly confused with the metric. </p>\n<p>Is the competition metric <code>accuracy</code> or <code>categorical_accuracy</code> in tensorflow? A formal definition would also suffice.</p>",
  "messages": [
    {
      "id": "1103306",
      "postDate": "12/05/2020 20:00:19",
      "content": "<p><a href=\"https://www.kaggle.com/juliaelliott\" target=\"_blank\">@juliaelliott</a>  In the evaluation page, the competition metric is given as <code>categorization accuracy</code> with a link to a binary case. I'm slightly confused with the metric. </p>\n<p>Is the competition metric <code>accuracy</code> or <code>categorical_accuracy</code> in tensorflow? A formal definition would also suffice.</p>",
      "rawMarkdown": "juliaelliott  In the evaluation page, the competition metric is given as `categorization accuracy` with a link to a binary case. I'm slightly confused with the metric. \n\nIs the competition metric `accuracy` or `categorical_accuracy` in tensorflow? A formal definition would also suffice.",
      "votes": null
    },
    {
      "id": "1106143",
      "postDate": "12/08/2020 15:10:01",
      "content": "<p>It's standard accuracy metric, <br>\n<code>Number of Correctly Predicted Datapoints / Total Datapoints</code><br>\nRef: <a href=\"https://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html\" target=\"_blank\">https://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html</a></p>",
      "rawMarkdown": "It's standard accuracy metric, \n` Number of Correctly Predicted Datapoints / Total Datapoints `\nRef: https://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html",
      "votes": null
    },
    {
      "id": "1106160",
      "postDate": "12/08/2020 15:15:00",
      "content": "<p>OK. 1 if the row is predicted correctly, and 0 otherwise.</p>",
      "rawMarkdown": "OK. 1 if the row is predicted correctly, and 0 otherwise.",
      "votes": null
    },
    {
      "id": "1107125",
      "postDate": "12/09/2020 12:14:27",
      "content": "<p>Yes, Correct.</p>",
      "rawMarkdown": "Yes, Correct.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1106143,
      "author_name": "darshanpatel11",
      "author_url": "",
      "post_date": "12/08/2020 15:10:01",
      "content": "<p>It's standard accuracy metric, <br>\n<code>Number of Correctly Predicted Datapoints / Total Datapoints</code><br>\nRef: <a href=\"https://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html\" target=\"_blank\">https://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 1106160,
          "author_name": "tolgadincer",
          "author_url": "",
          "post_date": "12/08/2020 15:15:00",
          "content": "<p>OK. 1 if the row is predicted correctly, and 0 otherwise.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1107125,
          "author_name": "darshanpatel11",
          "author_url": "",
          "post_date": "12/09/2020 12:14:27",
          "content": "<p>Yes, Correct.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1103306": "juliaelliott  In the evaluation page, the competition metric is given as `categorization accuracy` with a link to a binary case. I'm slightly confused with the metric. \n\nIs the competition metric `accuracy` or `categorical_accuracy` in tensorflow? A formal definition would also suffice.",
    "1106143": "It's standard accuracy metric, \n` Number of Correctly Predicted Datapoints / Total Datapoints `\nRef: https://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html",
    "1106160": "OK. 1 if the row is predicted correctly, and 0 otherwise.",
    "1107125": "Yes, Correct."
  },
  "source": "meta"
}