{
  "id": 105854,
  "title": "Additional data section - any \"molecularly-transferable\" information in there?",
  "url": "/competitions/champs-scalar-coupling/discussion/105854",
  "author_name": "Michel ML",
  "post_date": "2019-08-26T19:29:55.272000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Disclaimer: I'm very late to this competition</p>\n\n<p>This is a question related to the competition's additional data: <a href=\"https://www.kaggle.com/c/champs-scalar-coupling/data\">https://www.kaggle.com/c/champs-scalar-coupling/data</a>  </p>\n\n<p>It is said that the additional data is only available for the train molecules, I wonder if anyone would be willing to share if some of this additional data is molecule-independent or transferable from one type of molecule to another one.</p>\n\n<p>Specific questions:\n- Is there something stable at the atomic level in this additional data that we could use both in the train and test scenarios?\n- Is there such a thing as families of molecules? Where, for example, certain molecules of one family could be found both in train and test - and would have the same <code>potential_energy</code> ?</p>\n\n<p>Links to relevant information are welcome too </p>",
  "messages": [
    {
      "id": 608467,
      "postDate": "2019-08-26T19:29:55.273Z",
      "content": "<p>Disclaimer: I'm very late to this competition</p>\n\n<p>This is a question related to the competition's additional data: <a href=\"https://www.kaggle.com/c/champs-scalar-coupling/data\">https://www.kaggle.com/c/champs-scalar-coupling/data</a>  </p>\n\n<p>It is said that the additional data is only available for the train molecules, I wonder if anyone would be willing to share if some of this additional data is molecule-independent or transferable from one type of molecule to another one.</p>\n\n<p>Specific questions:\n- Is there something stable at the atomic level in this additional data that we could use both in the train and test scenarios?\n- Is there such a thing as families of molecules? Where, for example, certain molecules of one family could be found both in train and test - and would have the same <code>potential_energy</code> ?</p>\n\n<p>Links to relevant information are welcome too </p>",
      "rawMarkdown": "Disclaimer: I'm very late to this competition\n\nThis is a question related to the competition's additional data: https://www.kaggle.com/c/champs-scalar-coupling/data  \n\nIt is said that the additional data is only available for the train molecules, I wonder if anyone would be willing to share if some of this additional data is molecule-independent or transferable from one type of molecule to another one.\n\nSpecific questions:\n- Is there something stable at the atomic level in this additional data that we could use both in the train and test scenarios?\n- Is there such a thing as families of molecules? Where, for example, certain molecules of one family could be found both in train and test - and would have the same `potential_energy` ?\n\nLinks to relevant information are welcome too "
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "608467": "Disclaimer: I'm very late to this competition\n\nThis is a question related to the competition's additional data: https://www.kaggle.com/c/champs-scalar-coupling/data  \n\nIt is said that the additional data is only available for the train molecules, I wonder if anyone would be willing to share if some of this additional data is molecule-independent or transferable from one type of molecule to another one.\n\nSpecific questions:\n- Is there something stable at the atomic level in this additional data that we could use both in the train and test scenarios?\n- Is there such a thing as families of molecules? Where, for example, certain molecules of one family could be found both in train and test - and would have the same `potential_energy` ?\n\nLinks to relevant information are welcome too "
  }
}