{
  "id": 501903,
  "title": "How to submit parquet file to Kaggle?",
  "url": "/competitions/leap-atmospheric-physics-ai-climsim/discussion/501903",
  "author_name": "DennisSakva",
  "post_date": "2024-05-11T07:34:47.948000",
  "votes": 7,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I get the \"ID column sample_id not found in submission\" error, however, when I read the file the column is there.<br>\nThe code I use to save it</p>\n<pre><code>submission=path+\ntest_predicts.to_parquet(submission, index=)\n</code></pre>",
  "messages": [
    {
      "id": 2806619,
      "postDate": "2024-05-11T07:34:47.950Z",
      "content": "<p>I get the \"ID column sample_id not found in submission\" error, however, when I read the file the column is there.<br>\nThe code I use to save it</p>\n<pre><code>submission=path+\ntest_predicts.to_parquet(submission, index=)\n</code></pre>",
      "rawMarkdown": "I get the \"ID column sample_id not found in submission\" error, however, when I read the file the column is there.\nThe code I use to save it\n```python\nsubmission=path+'submission.parquet'\ntest_predicts.to_parquet(submission, index=True)\n```",
      "votes": 7
    },
    {
      "id": 2806654,
      "postDate": "2024-05-11T07:52:11.700Z",
      "content": "<p>Apparently, Kaggle ignores parquet index. Add non-index \"sample_id\" column and it works. Much faster than saving a compressed csv, comparable size though.</p>",
      "rawMarkdown": "Apparently, Kaggle ignores parquet index. Add non-index \"sample_id\" column and it works. Much faster than saving a compressed csv, comparable size though.",
      "votes": 4
    }
  ],
  "comments": [
    {
      "id": 2806654,
      "author_name": "DennisSakva",
      "author_url": "",
      "post_date": "2024-05-11T07:52:11.700000",
      "content": "<p>Apparently, Kaggle ignores parquet index. Add non-index \"sample_id\" column and it works. Much faster than saving a compressed csv, comparable size though.</p>",
      "votes": 4,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2806619": "I get the \"ID column sample_id not found in submission\" error, however, when I read the file the column is there.\nThe code I use to save it\n```python\nsubmission=path+'submission.parquet'\ntest_predicts.to_parquet(submission, index=True)\n```",
    "2806654": "Apparently, Kaggle ignores parquet index. Add non-index \"sample_id\" column and it works. Much faster than saving a compressed csv, comparable size though."
  }
}