{
  "id": 578350,
  "title": "How to handle missing data?",
  "url": "/competitions/stanford-rna-3d-folding/discussion/578350",
  "author_name": "Shebagi Mitra",
  "post_date": "2025-05-10T08:36:17.002000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>It's my first competition, so please guide me if I'm making some major mistake. I'm trying to predict data for my test set, which are the 12 CASP15 targets and their labels are given in the validation sets and those are required to predict to make a successful submission I believe.</p>\n<p>But ID's such as R1116_1, R1116_48-56 and some others don't have any labels in the validation_labels. How am I supposed to predict them? My model predicted the values for other labels just fine, but these ones it gave some ridiculous value which resulted in some insane values for mean absolute and square errors. I also cannot just delete those since the submission file demands I have predictions for these. So how am I supposed to handle them? I'm running a linear regression model and during the predictions on the test set, for x labels, I'm using y and z labels and other categorical labels to predict x. And same goes for the others in a similar manner. I'm not sure what to do with these missing values and I can't make a successful submission either because of it I believe. Please help. Thank you for reading!</p>",
  "messages": [
    {
      "id": 3198944,
      "postDate": "2025-05-10T08:36:17.003Z",
      "content": "<p>It's my first competition, so please guide me if I'm making some major mistake. I'm trying to predict data for my test set, which are the 12 CASP15 targets and their labels are given in the validation sets and those are required to predict to make a successful submission I believe.</p>\n<p>But ID's such as R1116_1, R1116_48-56 and some others don't have any labels in the validation_labels. How am I supposed to predict them? My model predicted the values for other labels just fine, but these ones it gave some ridiculous value which resulted in some insane values for mean absolute and square errors. I also cannot just delete those since the submission file demands I have predictions for these. So how am I supposed to handle them? I'm running a linear regression model and during the predictions on the test set, for x labels, I'm using y and z labels and other categorical labels to predict x. And same goes for the others in a similar manner. I'm not sure what to do with these missing values and I can't make a successful submission either because of it I believe. Please help. Thank you for reading!</p>",
      "rawMarkdown": "It's my first competition, so please guide me if I'm making some major mistake. I'm trying to predict data for my test set, which are the 12 CASP15 targets and their labels are given in the validation sets and those are required to predict to make a successful submission I believe.\n\nBut ID's such as R1116_1, R1116_48-56 and some others don't have any labels in the validation_labels. How am I supposed to predict them? My model predicted the values for other labels just fine, but these ones it gave some ridiculous value which resulted in some insane values for mean absolute and square errors. I also cannot just delete those since the submission file demands I have predictions for these. So how am I supposed to handle them? I'm running a linear regression model and during the predictions on the test set, for x labels, I'm using y and z labels and other categorical labels to predict x. And same goes for the others in a similar manner. I'm not sure what to do with these missing values and I can't make a successful submission either because of it I believe. Please help. Thank you for reading!"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "3198944": "It's my first competition, so please guide me if I'm making some major mistake. I'm trying to predict data for my test set, which are the 12 CASP15 targets and their labels are given in the validation sets and those are required to predict to make a successful submission I believe.\n\nBut ID's such as R1116_1, R1116_48-56 and some others don't have any labels in the validation_labels. How am I supposed to predict them? My model predicted the values for other labels just fine, but these ones it gave some ridiculous value which resulted in some insane values for mean absolute and square errors. I also cannot just delete those since the submission file demands I have predictions for these. So how am I supposed to handle them? I'm running a linear regression model and during the predictions on the test set, for x labels, I'm using y and z labels and other categorical labels to predict x. And same goes for the others in a similar manner. I'm not sure what to do with these missing values and I can't make a successful submission either because of it I believe. Please help. Thank you for reading!"
  }
}