{
  "id": 169885,
  "title": "Plot metric",
  "url": "/competitions/osic-pulmonary-fibrosis-progression/discussion/169885",
  "author_name": "",
  "post_date": "2020-07-25T14:55:35.613134600Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I tried to depict what the metric looks like and I have a question if this can help in any way, for example, in choosing a loss function<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332114%2Fce2520d447ea921355caac4d368dfd27%2Fmetric.png?generation=1595688924487898&amp;alt=media\" alt=\"\"></p>",
  "messages": [
    {
      "id": "945090",
      "postDate": "07/25/2020 14:55:35",
      "content": "<p>I tried to depict what the metric looks like and I have a question if this can help in any way, for example, in choosing a loss function<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332114%2Fce2520d447ea921355caac4d368dfd27%2Fmetric.png?generation=1595688924487898&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "I tried to depict what the metric looks like and I have a question if this can help in any way, for example, in choosing a loss function![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332114%2Fce2520d447ea921355caac4d368dfd27%2Fmetric.png?generation=1595688924487898&amp;alt=media)",
      "votes": null
    },
    {
      "id": "945311",
      "postDate": "07/25/2020 18:18:32",
      "content": "<p>Is it laplace log-likelihood? What does the plot depict?</p>\n\n<p><a href=\"/konar616\">@konar616</a> </p>",
      "rawMarkdown": "Is it laplace log-likelihood? What does the plot depict?\n\n@konar616",
      "votes": null
    },
    {
      "id": "949539",
      "postDate": "07/28/2020 17:42:01",
      "content": "<p><a href=\"/kurianbenoy\">@kurianbenoy</a>  Sorry for the long answer, there is a mistake, the graph looks different. I just depicted the metric as a function of two arguments (delta, confidence). They wrote that the theoretical minimum is around 4.59</p>\n\n<p>This is a new schedule:<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332114%2F3cecc709b9ac77c7c82cd4ce67e51218%2F9b9c89e67e5bd7025cd5da35717bf3fb.png?generation=1595958113084330&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "kurianbenoy  Sorry for the long answer, there is a mistake, the graph looks different. I just depicted the metric as a function of two arguments (delta, confidence). They wrote that the theoretical minimum is around 4.59\n\nThis is a new schedule:![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332114%2F3cecc709b9ac77c7c82cd4ce67e51218%2F9b9c89e67e5bd7025cd5da35717bf3fb.png?generation=1595958113084330&amp;alt=media)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 945311,
      "author_name": "kurianbenoy",
      "author_url": "",
      "post_date": "07/25/2020 18:18:32",
      "content": "<p>Is it laplace log-likelihood? What does the plot depict?</p>\n\n<p><a href=\"/konar616\">@konar616</a> </p>",
      "votes": null,
      "replies": [
        {
          "id": 949539,
          "author_name": "",
          "author_url": "",
          "post_date": "07/28/2020 17:42:01",
          "content": "<p><a href=\"/kurianbenoy\">@kurianbenoy</a>  Sorry for the long answer, there is a mistake, the graph looks different. I just depicted the metric as a function of two arguments (delta, confidence). They wrote that the theoretical minimum is around 4.59</p>\n\n<p>This is a new schedule:<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332114%2F3cecc709b9ac77c7c82cd4ce67e51218%2F9b9c89e67e5bd7025cd5da35717bf3fb.png?generation=1595958113084330&amp;alt=media\" alt=\"\"></p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "945090": "I tried to depict what the metric looks like and I have a question if this can help in any way, for example, in choosing a loss function![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332114%2Fce2520d447ea921355caac4d368dfd27%2Fmetric.png?generation=1595688924487898&amp;alt=media)",
    "945311": "Is it laplace log-likelihood? What does the plot depict?\n\n@konar616",
    "949539": "kurianbenoy  Sorry for the long answer, there is a mistake, the graph looks different. I just depicted the metric as a function of two arguments (delta, confidence). They wrote that the theoretical minimum is around 4.59\n\nThis is a new schedule:![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332114%2F3cecc709b9ac77c7c82cd4ce67e51218%2F9b9c89e67e5bd7025cd5da35717bf3fb.png?generation=1595958113084330&amp;alt=media)"
  },
  "source": "meta"
}