{
  "id": 78623,
  "title": "Fail to get good accuracy on time_failure",
  "url": "/competitions/LANL-Earthquake-Prediction/discussion/78623",
  "author_name": "",
  "post_date": "2019-01-26T05:48:43.661633400Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi all,\nI was checking different kernels and discussions and from this one <a href=\"https://www.kaggle.com/artgor/seismic-data-eda-and-baseline\">https://www.kaggle.com/artgor/seismic-data-eda-and-baseline</a> I noticed I didn't have the same accuracy for time_to_failure while running the same command than on this link. Does anyone have any idea how I could get a better accuracy ?\nI tried np.around which usually do the trick but it doesn't work here...</p>\n\n<p>time = data['time_to_failure']\ntime.head()\n0    1.4691\n1    1.4691\n2    1.4691\n3    1.4691\n4    1.4691</p>",
  "messages": [
    {
      "id": "461460",
      "postDate": "01/26/2019 05:48:43",
      "content": "<p>Hi all,\nI was checking different kernels and discussions and from this one <a href=\"https://www.kaggle.com/artgor/seismic-data-eda-and-baseline\">https://www.kaggle.com/artgor/seismic-data-eda-and-baseline</a> I noticed I didn't have the same accuracy for time_to_failure while running the same command than on this link. Does anyone have any idea how I could get a better accuracy ?\nI tried np.around which usually do the trick but it doesn't work here...</p>\n\n<p>time = data['time_to_failure']\ntime.head()\n0    1.4691\n1    1.4691\n2    1.4691\n3    1.4691\n4    1.4691</p>",
      "rawMarkdown": "Hi all,\nI was checking different kernels and discussions and from this one https://www.kaggle.com/artgor/seismic-data-eda-and-baseline I noticed I didn't have the same accuracy for time_to_failure while running the same command than on this link. Does anyone have any idea how I could get a better accuracy ?\nI tried np.around which usually do the trick but it doesn't work here...\n\ntime = data['time_to_failure']\ntime.head()\n0    1.4691\n1    1.4691\n2    1.4691\n3    1.4691\n4    1.4691",
      "votes": null
    },
    {
      "id": "461504",
      "postDate": "01/26/2019 08:01:19",
      "content": "<p>Rounding target values in this competition isn't a good idea - the values are vary precise, you need 16 figures after the comma.</p>",
      "rawMarkdown": "Rounding target values in this competition isn't a good idea - the values are vary precise, you need 16 figures after the comma.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 461504,
      "author_name": "artgor",
      "author_url": "",
      "post_date": "01/26/2019 08:01:19",
      "content": "<p>Rounding target values in this competition isn't a good idea - the values are vary precise, you need 16 figures after the comma.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "461460": "Hi all,\nI was checking different kernels and discussions and from this one https://www.kaggle.com/artgor/seismic-data-eda-and-baseline I noticed I didn't have the same accuracy for time_to_failure while running the same command than on this link. Does anyone have any idea how I could get a better accuracy ?\nI tried np.around which usually do the trick but it doesn't work here...\n\ntime = data['time_to_failure']\ntime.head()\n0    1.4691\n1    1.4691\n2    1.4691\n3    1.4691\n4    1.4691",
    "461504": "Rounding target values in this competition isn't a good idea - the values are vary precise, you need 16 figures after the comma."
  },
  "source": "meta"
}