{
  "id": 30218,
  "title": "What are you using for this competition?",
  "url": "/competitions/intel-mobileodt-cervical-cancer-screening/discussion/30218",
  "author_name": "",
  "post_date": "2017-03-16T22:54:21.365368300Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I'm curious what people are using for this. Is anyone using h2o? Intel tools? Anyone using R for this competition? Are you pre-training your model or using an existing​ pretrained model?</p>",
  "messages": [
    {
      "id": "168352",
      "postDate": "03/16/2017 22:54:21",
      "content": "<p>I'm curious what people are using for this. Is anyone using h2o? Intel tools? Anyone using R for this competition? Are you pre-training your model or using an existing​ pretrained model?</p>",
      "rawMarkdown": "I'm curious what people are using for this. Is anyone using h2o? Intel tools? Anyone using R for this competition? Are you pre-training your model or using an existing​ pretrained model?",
      "votes": null
    },
    {
      "id": "172547",
      "postDate": "04/04/2017 03:42:55",
      "content": "<p>I'm new to the field, and taking a somewhat naive approach as a starting point; I'm using Octave for my model design and testing for ease of design and debugging.</p>",
      "rawMarkdown": "I'm new to the field, and taking a somewhat naive approach as a starting point; I'm using Octave for my model design and testing for ease of design and debugging.",
      "votes": null
    },
    {
      "id": "172831",
      "postDate": "04/04/2017 23:24:48",
      "content": "<p>Honestly, this challenge really confuses my network. It it sometimes inexplicably confident, sometimes goes 33 33 33. I don't even know. The plan is to compare an untrained model of my own design to the results of untrained inception and resnet50, to see which works best.</p>",
      "rawMarkdown": "Honestly, this challenge really confuses my network. It it sometimes inexplicably confident, sometimes goes 33 33 33. I don't even know. The plan is to compare an untrained model of my own design to the results of untrained inception and resnet50, to see which works best.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 172547,
      "author_name": "mlazich",
      "author_url": "",
      "post_date": "04/04/2017 03:42:55",
      "content": "<p>I'm new to the field, and taking a somewhat naive approach as a starting point; I'm using Octave for my model design and testing for ease of design and debugging.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 172831,
      "author_name": "benchiislett",
      "author_url": "",
      "post_date": "04/04/2017 23:24:48",
      "content": "<p>Honestly, this challenge really confuses my network. It it sometimes inexplicably confident, sometimes goes 33 33 33. I don't even know. The plan is to compare an untrained model of my own design to the results of untrained inception and resnet50, to see which works best.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "168352": "I'm curious what people are using for this. Is anyone using h2o? Intel tools? Anyone using R for this competition? Are you pre-training your model or using an existing​ pretrained model?",
    "172547": "I'm new to the field, and taking a somewhat naive approach as a starting point; I'm using Octave for my model design and testing for ease of design and debugging.",
    "172831": "Honestly, this challenge really confuses my network. It it sometimes inexplicably confident, sometimes goes 33 33 33. I don't even know. The plan is to compare an untrained model of my own design to the results of untrained inception and resnet50, to see which works best."
  },
  "source": "meta"
}