{
  "id": 585622,
  "title": "Problem read_parquet",
  "url": "/competitions/aeroclub-recsys-2025/discussion/585622",
  "author_name": "",
  "post_date": "2025-06-21T20:41:24.652552800Z",
  "votes": 8,
  "comment_count": 1,
  "views": 0,
  "content": "<p>In this competition, if you have not changed the dataset, you will encounter an error:<br>\n❗ Your notebook tried to allocate memory memory than is available. It has restarted</p>\n<p>1 Use polars lib:</p>\n<pre><code> polars  pl\ntrain = pl.read_parquet()\n\n\n train\n\ntest = pl.read_parquet()\n</code></pre>\n<p>2 Just turn on TPU<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F17954496%2F4090241441707625d8ed73222c1c9342%2Fsave.png?generation=1750538268036650&amp;alt=media\" alt=\"showTpu\"></p>",
  "messages": [
    {
      "id": "3229637",
      "postDate": "06/21/2025 20:41:24",
      "content": "<p>In this competition, if you have not changed the dataset, you will encounter an error:<br>\n❗ Your notebook tried to allocate memory memory than is available. It has restarted</p>\n<p>1 Use polars lib:</p>\n<pre><code> polars  pl\ntrain = pl.read_parquet()\n\n\n train\n\ntest = pl.read_parquet()\n</code></pre>\n<p>2 Just turn on TPU<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F17954496%2F4090241441707625d8ed73222c1c9342%2Fsave.png?generation=1750538268036650&amp;alt=media\" alt=\"showTpu\"></p>",
      "rawMarkdown": "In this competition, if you have not changed the dataset, you will encounter an error:\n❗ Your notebook tried to allocate memory memory than is available. It has restarted\n\n1 Use polars lib:\n```Python\nimport polars as pl\ntrain = pl.read_parquet('/kaggle/input/aeroclub-recsys-2025/train.parquet')\n\n# learn model\ndel train\n\ntest = pl.read_parquet('/kaggle/input/aeroclub-recsys-2025/test.parquet')\n```\n\n2 Just turn on TPU\n![showTpu](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F17954496%2F4090241441707625d8ed73222c1c9342%2Fsave.png?generation=1750538268036650&alt=media)",
      "votes": null
    },
    {
      "id": "3229703",
      "postDate": "06/22/2025 01:45:29",
      "content": "<p>Or DuckDB too. Your option returned a better dataframe since it has ALL columns and shape.<br>\nReally, very helpful Antonoof.</p>",
      "rawMarkdown": "Or DuckDB too. Your option returned a better dataframe since it has ALL columns and shape.\nReally, very helpful Antonoof.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3229703,
      "author_name": "mpwolke",
      "author_url": "",
      "post_date": "06/22/2025 01:45:29",
      "content": "<p>Or DuckDB too. Your option returned a better dataframe since it has ALL columns and shape.<br>\nReally, very helpful Antonoof.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3229637": "In this competition, if you have not changed the dataset, you will encounter an error:\n❗ Your notebook tried to allocate memory memory than is available. It has restarted\n\n1 Use polars lib:\n```Python\nimport polars as pl\ntrain = pl.read_parquet('/kaggle/input/aeroclub-recsys-2025/train.parquet')\n\n# learn model\ndel train\n\ntest = pl.read_parquet('/kaggle/input/aeroclub-recsys-2025/test.parquet')\n```\n\n2 Just turn on TPU\n![showTpu](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F17954496%2F4090241441707625d8ed73222c1c9342%2Fsave.png?generation=1750538268036650&alt=media)",
    "3229703": "Or DuckDB too. Your option returned a better dataframe since it has ALL columns and shape.\nReally, very helpful Antonoof."
  },
  "source": "meta"
}