{
  "id": 302829,
  "title": "Timeout Error While using Pre-traind model",
  "url": "/competitions/g-research-crypto-forecasting/discussion/302829",
  "author_name": "",
  "post_date": "2022-01-24T13:54:47.103491400Z",
  "votes": 9,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I use a pretrainde models to infer test dataframe while the infer time in the notebook is 177 seconds when i submit my notebook it gives timeout error while the original notebook in which i trained the model and use it in inference it ran in 6600 seconds and pass successfully in submission is there any one has same issues</p>",
  "messages": [
    {
      "id": "1662727",
      "postDate": "01/24/2022 13:54:47",
      "content": "<p>I use a pretrainde models to infer test dataframe while the infer time in the notebook is 177 seconds when i submit my notebook it gives timeout error while the original notebook in which i trained the model and use it in inference it ran in 6600 seconds and pass successfully in submission is there any one has same issues</p>",
      "rawMarkdown": "I use a pretrainde models to infer test dataframe while the infer time in the notebook is 177 seconds when i submit my notebook it gives timeout error while the original notebook in which i trained the model and use it in inference it ran in 6600 seconds and pass successfully in submission is there any one has same issues",
      "votes": null
    },
    {
      "id": "1663929",
      "postDate": "01/25/2022 14:40:36",
      "content": "<p>I am having the same problem. Some discussions say that removing print() and display() in the iteration solves the problem, but it did not.</p>",
      "rawMarkdown": "I am having the same problem. Some discussions say that removing print() and display() in the iteration solves the problem, but it did not.",
      "votes": null
    },
    {
      "id": "1664103",
      "postDate": "01/25/2022 17:36:55",
      "content": "<p>I would try make sure that each iteration on the API is less than 0.25s, else it is likely that you will time out.</p>",
      "rawMarkdown": "I would try make sure that each iteration on the API is less than 0.25s, else it is likely that you will time out.",
      "votes": null
    },
    {
      "id": "1664321",
      "postDate": "01/25/2022 21:16:11",
      "content": "<p>If you need a tool to assess how your model is doing during submission, try importing my local_api module, which emulates the submission api locally. A detailed introduction is <a href=\"https://www.kaggle.com/jagofc/detailed-local-api-introduction\" target=\"_blank\">here</a>. It gives a pretty good idea on timing.</p>",
      "rawMarkdown": "If you need a tool to assess how your model is doing during submission, try importing my local_api module, which emulates the submission api locally. A detailed introduction is [here](https://www.kaggle.com/jagofc/detailed-local-api-introduction). It gives a pretty good idea on timing.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1663929,
      "author_name": "naokinaokinaoki",
      "author_url": "",
      "post_date": "01/25/2022 14:40:36",
      "content": "<p>I am having the same problem. Some discussions say that removing print() and display() in the iteration solves the problem, but it did not.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1664103,
      "author_name": "jonnydossantos",
      "author_url": "",
      "post_date": "01/25/2022 17:36:55",
      "content": "<p>I would try make sure that each iteration on the API is less than 0.25s, else it is likely that you will time out.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1664321,
      "author_name": "jagofc",
      "author_url": "",
      "post_date": "01/25/2022 21:16:11",
      "content": "<p>If you need a tool to assess how your model is doing during submission, try importing my local_api module, which emulates the submission api locally. A detailed introduction is <a href=\"https://www.kaggle.com/jagofc/detailed-local-api-introduction\" target=\"_blank\">here</a>. It gives a pretty good idea on timing.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1662727": "I use a pretrainde models to infer test dataframe while the infer time in the notebook is 177 seconds when i submit my notebook it gives timeout error while the original notebook in which i trained the model and use it in inference it ran in 6600 seconds and pass successfully in submission is there any one has same issues",
    "1663929": "I am having the same problem. Some discussions say that removing print() and display() in the iteration solves the problem, but it did not.",
    "1664103": "I would try make sure that each iteration on the API is less than 0.25s, else it is likely that you will time out.",
    "1664321": "If you need a tool to assess how your model is doing during submission, try importing my local_api module, which emulates the submission api locally. A detailed introduction is [here](https://www.kaggle.com/jagofc/detailed-local-api-introduction). It gives a pretty good idea on timing."
  },
  "source": "meta"
}