{
  "id": 71562,
  "title": "What tricks, resources did you use for this large sized data?",
  "url": "/competitions/airbus-ship-detection/discussion/71562",
  "author_name": "YaGana Sheriff-Hussaini",
  "post_date": "2018-11-14T20:22:41.187000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Since this competition's data requires a lot of resources to process, I want to pose this question to get tips from the top teams for future use.</p>\n\n<ul>\n<li><p>What tricks and resources i.e. HW, cloud etc. did you use for this\ncompetition. </p>\n\n<ul><li>How long did you train to get the best result?</li></ul></li>\n<li><p>If you are working on limited resources like me, how did you divide\nthe work load?</p></li>\n</ul>\n\n<p>I do not have a GPU machine and relied on Kaggle kernel and a cloud account for competitions. In this one I must admit I was discouraged by the leak earlier on so decided not to spend too much time on it. I did something similar to what @Lafoss did in <a href=\"https://www.kaggle.com/iafoss/fine-tuning-resnet34-on-ship-detection/notebook\">this notebook</a> before the leak. So I adapted it to the new data and submitted.</p>",
  "messages": [
    {
      "id": 421266,
      "postDate": "2018-11-14T20:22:41.187Z",
      "content": "<p>Since this competition's data requires a lot of resources to process, I want to pose this question to get tips from the top teams for future use.</p>\n\n<ul>\n<li><p>What tricks and resources i.e. HW, cloud etc. did you use for this\ncompetition. </p>\n\n<ul><li>How long did you train to get the best result?</li></ul></li>\n<li><p>If you are working on limited resources like me, how did you divide\nthe work load?</p></li>\n</ul>\n\n<p>I do not have a GPU machine and relied on Kaggle kernel and a cloud account for competitions. In this one I must admit I was discouraged by the leak earlier on so decided not to spend too much time on it. I did something similar to what @Lafoss did in <a href=\"https://www.kaggle.com/iafoss/fine-tuning-resnet34-on-ship-detection/notebook\">this notebook</a> before the leak. So I adapted it to the new data and submitted.</p>",
      "rawMarkdown": "Since this competition's data requires a lot of resources to process, I want to pose this question to get tips from the top teams for future use.\n\n - What tricks and resources i.e. HW, cloud etc. did you use for this\n   competition. \n   \n  - How long did you train to get the best result?\n   \n -  If you are working on limited resources like me, how did you divide\n   the work load?\n\nI do not have a GPU machine and relied on Kaggle kernel and a cloud account for competitions. In this one I must admit I was discouraged by the leak earlier on so decided not to spend too much time on it. I did something similar to what @Lafoss did in [this notebook][1] before the leak. So I adapted it to the new data and submitted.\n\n\n  [1]: https://www.kaggle.com/iafoss/fine-tuning-resnet34-on-ship-detection/notebook"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "421266": "Since this competition's data requires a lot of resources to process, I want to pose this question to get tips from the top teams for future use.\n\n - What tricks and resources i.e. HW, cloud etc. did you use for this\n   competition. \n   \n  - How long did you train to get the best result?\n   \n -  If you are working on limited resources like me, how did you divide\n   the work load?\n\nI do not have a GPU machine and relied on Kaggle kernel and a cloud account for competitions. In this one I must admit I was discouraged by the leak earlier on so decided not to spend too much time on it. I did something similar to what @Lafoss did in [this notebook][1] before the leak. So I adapted it to the new data and submitted.\n\n\n  [1]: https://www.kaggle.com/iafoss/fine-tuning-resnet34-on-ship-detection/notebook"
  }
}