{
  "id": 272472,
  "title": "PyTorch 1D CNN baseline train+inference",
  "url": "/competitions/g2net-gravitational-wave-detection/discussion/272472",
  "author_name": "",
  "post_date": "2021-09-15T15:42:52.036577600Z",
  "votes": 32,
  "comment_count": 2,
  "views": 0,
  "content": "<p><a href=\"https://www.kaggle.com/scaomath/g2net-1d-cnn-in-pytorch-baseline-train-inference\" target=\"_blank\">https://www.kaggle.com/scaomath/g2net-1d-cnn-in-pytorch-baseline-train-inference</a></p>\n<ul>\n<li>Inspired by <a href=\"https://www.kaggle.com/hengck23\" target=\"_blank\">@hengck23</a> 's comment at <a href=\"https://www.kaggle.com/c/g2net-gravitational-wave-detection/discussion/268553\" target=\"_blank\">https://www.kaggle.com/c/g2net-gravitational-wave-detection/discussion/268553</a>.</li>\n<li>The model is modified from <a href=\"https://www.kaggle.com/kit716/grav-wave-detection\" target=\"_blank\">https://www.kaggle.com/kit716/grav-wave-detection</a>, I simply added more channels, filter size unchanged.</li>\n<li>Used <a href=\"https://www.kaggle.com/hidehisaarai1213\" target=\"_blank\">@hidehisaarai1213</a> 's tfrecords dataloader.</li>\n<li>Pipeline is modified from <a href=\"https://www.kaggle.com/yasufuminakama\" target=\"_blank\">@yasufuminakama</a> 's, but the inference is improved from the original one's \"iter on loader first then load model\" to \"load model first then iter the loader\" thus much faster.</li>\n</ul>\n<p>Small model, no need to use TPU, CV 0.86xx.</p>",
  "messages": [
    {
      "id": "1514010",
      "postDate": "09/15/2021 15:42:52",
      "content": "<p><a href=\"https://www.kaggle.com/scaomath/g2net-1d-cnn-in-pytorch-baseline-train-inference\" target=\"_blank\">https://www.kaggle.com/scaomath/g2net-1d-cnn-in-pytorch-baseline-train-inference</a></p>\n<ul>\n<li>Inspired by <a href=\"https://www.kaggle.com/hengck23\" target=\"_blank\">@hengck23</a> 's comment at <a href=\"https://www.kaggle.com/c/g2net-gravitational-wave-detection/discussion/268553\" target=\"_blank\">https://www.kaggle.com/c/g2net-gravitational-wave-detection/discussion/268553</a>.</li>\n<li>The model is modified from <a href=\"https://www.kaggle.com/kit716/grav-wave-detection\" target=\"_blank\">https://www.kaggle.com/kit716/grav-wave-detection</a>, I simply added more channels, filter size unchanged.</li>\n<li>Used <a href=\"https://www.kaggle.com/hidehisaarai1213\" target=\"_blank\">@hidehisaarai1213</a> 's tfrecords dataloader.</li>\n<li>Pipeline is modified from <a href=\"https://www.kaggle.com/yasufuminakama\" target=\"_blank\">@yasufuminakama</a> 's, but the inference is improved from the original one's \"iter on loader first then load model\" to \"load model first then iter the loader\" thus much faster.</li>\n</ul>\n<p>Small model, no need to use TPU, CV 0.86xx.</p>",
      "rawMarkdown": "https://www.kaggle.com/scaomath/g2net-1d-cnn-in-pytorch-baseline-train-inference\n\n- Inspired by @hengck23 's comment at https://www.kaggle.com/c/g2net-gravitational-wave-detection/discussion/268553.\n- The model is modified from https://www.kaggle.com/kit716/grav-wave-detection, I simply added more channels, filter size unchanged.\n- Used @hidehisaarai1213 's tfrecords dataloader.\n- Pipeline is modified from @yasufuminakama 's, but the inference is improved from the original one's \"iter on loader first then load model\" to \"load model first then iter the loader\" thus much faster.\n\nSmall model, no need to use TPU, CV 0.86xx.",
      "votes": null
    },
    {
      "id": "1516164",
      "postDate": "09/18/2021 02:59:36",
      "content": "<p>great! thankyou!</p>",
      "rawMarkdown": "great! thankyou!",
      "votes": null
    },
    {
      "id": "1559720",
      "postDate": "10/27/2021 07:09:15",
      "content": "<p>Hey All,</p>\n<p>Thank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey <a href=\"https://forms.gle/QP9L16niPexozyhu5\" target=\"_blank\">https://forms.gle/QP9L16niPexozyhu5</a>.</p>\n<p>Thank you all,</p>\n<p>Regards,<br>\nChris</p>",
      "rawMarkdown": "Hey All,\n\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.\n\nThank you all,\n\nRegards,\nChris",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1516164,
      "author_name": "faaizhashmi",
      "author_url": "",
      "post_date": "09/18/2021 02:59:36",
      "content": "<p>great! thankyou!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1559720,
      "author_name": "zerafachris",
      "author_url": "",
      "post_date": "10/27/2021 07:09:15",
      "content": "<p>Hey All,</p>\n<p>Thank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey <a href=\"https://forms.gle/QP9L16niPexozyhu5\" target=\"_blank\">https://forms.gle/QP9L16niPexozyhu5</a>.</p>\n<p>Thank you all,</p>\n<p>Regards,<br>\nChris</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1514010": "https://www.kaggle.com/scaomath/g2net-1d-cnn-in-pytorch-baseline-train-inference\n\n- Inspired by @hengck23 's comment at https://www.kaggle.com/c/g2net-gravitational-wave-detection/discussion/268553.\n- The model is modified from https://www.kaggle.com/kit716/grav-wave-detection, I simply added more channels, filter size unchanged.\n- Used @hidehisaarai1213 's tfrecords dataloader.\n- Pipeline is modified from @yasufuminakama 's, but the inference is improved from the original one's \"iter on loader first then load model\" to \"load model first then iter the loader\" thus much faster.\n\nSmall model, no need to use TPU, CV 0.86xx.",
    "1516164": "great! thankyou!",
    "1559720": "Hey All,\n\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.\n\nThank you all,\n\nRegards,\nChris"
  },
  "source": "meta"
}