{
  "id": 174995,
  "title": "Autonencoder approach for Anomaly detection - notebook link ( TPU )",
  "url": "/competitions/siim-isic-melanoma-classification/discussion/174995",
  "author_name": "Rajnish Chauhan",
  "post_date": "2020-08-16T17:02:33.638000",
  "votes": 3,
  "comment_count": 0,
  "views": 0,
  "content": "<p>This problem could be treated as anomaly and autonecoder approach could work.</p>\n<p>Here is notebook for autoencoder<br>\n<br>\nIt runs on TPU now and one epoch take just 25 min on TPU 😊<br>\nUpto you to check how to use it. 😊</p>\n<ul>\n<li><a href=\"https://www.kaggle.com/rajnishe/rc-siim-autoencoder-model\" target=\"_blank\">https://www.kaggle.com/rajnishe/rc-siim-autoencoder-model</a></li>\n</ul>",
  "messages": [
    {
      "id": 972592,
      "postDate": "2020-08-16T17:02:33.637Z",
      "content": "<p>This problem could be treated as anomaly and autonecoder approach could work.</p>\n<p>Here is notebook for autoencoder<br>\n<br>\nIt runs on TPU now and one epoch take just 25 min on TPU 😊<br>\nUpto you to check how to use it. 😊</p>\n<ul>\n<li><a href=\"https://www.kaggle.com/rajnishe/rc-siim-autoencoder-model\" target=\"_blank\">https://www.kaggle.com/rajnishe/rc-siim-autoencoder-model</a></li>\n</ul>",
      "rawMarkdown": "This problem could be treated as anomaly and autonecoder approach could work.\n\nHere is notebook for autoencoder\n~~But this take a lot of time , one epoch complete in 4 hrs on all train images of size 384.~~\nIt runs on TPU now and one epoch take just 25 min on TPU 😊\nUpto you to check how to use it. 😊\n- https://www.kaggle.com/rajnishe/rc-siim-autoencoder-model",
      "votes": 3
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "972592": "This problem could be treated as anomaly and autonecoder approach could work.\n\nHere is notebook for autoencoder\n~~But this take a lot of time , one epoch complete in 4 hrs on all train images of size 384.~~\nIt runs on TPU now and one epoch take just 25 min on TPU 😊\nUpto you to check how to use it. 😊\n- https://www.kaggle.com/rajnishe/rc-siim-autoencoder-model"
  }
}