{
  "id": 209187,
  "title": "Pickling and loading nested defaultdict with a lambda function as the default value",
  "url": "/competitions/riiid-test-answer-prediction/discussion/209187",
  "author_name": "Shuhao Cao",
  "post_date": "2021-01-06T15:52:19.356000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p><a href=\"https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump\" target=\"_blank\">https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump</a></p>\n<p>To download the pickled files for fast inference, please go to Version 8: <a href=\"https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump?scriptVersionId=50800149\" target=\"_blank\">https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump?scriptVersionId=50800149</a></p>\n<p>The data can be used to do inference for <a href=\"https://www.kaggle.com/ragnar123/riiid-model-lgbm\" target=\"_blank\">https://www.kaggle.com/ragnar123/riiid-model-lgbm</a> and <a href=\"https://www.kaggle.com/its7171/lgbm-with-loop-feature-engineering\" target=\"_blank\">https://www.kaggle.com/its7171/lgbm-with-loop-feature-engineering</a>. If your current notebook running time + submission time is &lt; 6 hours (including). Adding it to your inference ensembling pipeline is good (I tested myself the submission is &lt; 2 hours if you load the pickled file directly).</p>",
  "messages": [
    {
      "id": 1141290,
      "postDate": "2021-01-06T15:52:19.357Z",
      "content": "<p><a href=\"https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump\" target=\"_blank\">https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump</a></p>\n<p>To download the pickled files for fast inference, please go to Version 8: <a href=\"https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump?scriptVersionId=50800149\" target=\"_blank\">https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump?scriptVersionId=50800149</a></p>\n<p>The data can be used to do inference for <a href=\"https://www.kaggle.com/ragnar123/riiid-model-lgbm\" target=\"_blank\">https://www.kaggle.com/ragnar123/riiid-model-lgbm</a> and <a href=\"https://www.kaggle.com/its7171/lgbm-with-loop-feature-engineering\" target=\"_blank\">https://www.kaggle.com/its7171/lgbm-with-loop-feature-engineering</a>. If your current notebook running time + submission time is &lt; 6 hours (including). Adding it to your inference ensembling pipeline is good (I tested myself the submission is &lt; 2 hours if you load the pickled file directly).</p>",
      "rawMarkdown": "https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump\n\nTo download the pickled files for fast inference, please go to Version 8: https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump?scriptVersionId=50800149\n\nThe data can be used to do inference for https://www.kaggle.com/ragnar123/riiid-model-lgbm and https://www.kaggle.com/its7171/lgbm-with-loop-feature-engineering. If your current notebook running time + submission time is < 6 hours (including). Adding it to your inference ensembling pipeline is good (I tested myself the submission is < 2 hours if you load the pickled file directly).",
      "votes": -1
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1141290": "https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump\n\nTo download the pickled files for fast inference, please go to Version 8: https://www.kaggle.com/scaomath/riiid-lgbm-model-2-loop-user-dict-dump?scriptVersionId=50800149\n\nThe data can be used to do inference for https://www.kaggle.com/ragnar123/riiid-model-lgbm and https://www.kaggle.com/its7171/lgbm-with-loop-feature-engineering. If your current notebook running time + submission time is < 6 hours (including). Adding it to your inference ensembling pipeline is good (I tested myself the submission is < 2 hours if you load the pickled file directly)."
  }
}