{
  "id": 118115,
  "title": "95th Place Solution",
  "url": "/competitions/understanding_cloud_organization/writeups/uncloud-95th-place-solution",
  "author_name": "",
  "post_date": "2019-11-19T15:21:56.337260100Z",
  "votes": 13,
  "comment_count": 5,
  "views": 0,
  "content": "<p>It has been great to work with <a href=\"/muhakabartay\">@muhakabartay</a>, <a href=\"/carlolepelaars\">@carlolepelaars</a>, <a href=\"/roydatascience\">@roydatascience</a>, <a href=\"/lightnezzofbeing\">@lightnezzofbeing</a> . We have open sourced the code in github @  <a href=\"https://github.com/LightnessOfBeing/Cl0uds\">https://github.com/LightnessOfBeing/Cl0uds</a>. </p>",
  "messages": [
    {
      "id": "676889",
      "postDate": "11/19/2019 15:21:56",
      "content": "<p>It has been great to work with <a href=\"/muhakabartay\">@muhakabartay</a>, <a href=\"/carlolepelaars\">@carlolepelaars</a>, <a href=\"/roydatascience\">@roydatascience</a>, <a href=\"/lightnezzofbeing\">@lightnezzofbeing</a> . We have open sourced the code in github @  <a href=\"https://github.com/LightnessOfBeing/Cl0uds\">https://github.com/LightnessOfBeing/Cl0uds</a>. </p>",
      "rawMarkdown": "It has been great to work with @muhakabartay, @carlolepelaars, @roydatascience, @lightnezzofbeing . We have open sourced the code in github @  https://github.com/LightnessOfBeing/Cl0uds.",
      "votes": null
    },
    {
      "id": "676901",
      "postDate": "11/19/2019 15:33:03",
      "content": "<p>Congratulations\nThanks for Sharing your Approach <a href=\"/manojprabhaakr\">@manojprabhaakr</a> </p>",
      "rawMarkdown": "Congratulations\nThanks for Sharing your Approach @manojprabhaakr",
      "votes": null
    },
    {
      "id": "676915",
      "postDate": "11/19/2019 15:50:57",
      "content": "<p>Fully agree, it was great to work with this team!</p>",
      "rawMarkdown": "Fully agree, it was great to work with this team!",
      "votes": null
    },
    {
      "id": "677259",
      "postDate": "11/20/2019 01:15:44",
      "content": "<p>Congratulations!</p>",
      "rawMarkdown": "Congratulations!",
      "votes": null
    },
    {
      "id": "678076",
      "postDate": "11/21/2019 01:09:08",
      "content": "<p>Congrats Manoj and team. Great job. I checked out your code. I like how you use pseudo labels. That's smart to only use them during stage 1 of training. It helps your net's learning move in the right direction but then the model is fine tuned with the true training images.</p>",
      "rawMarkdown": "Congrats Manoj and team. Great job. I checked out your code. I like how you use pseudo labels. That's smart to only use them during stage 1 of training. It helps your net's learning move in the right direction but then the model is fine tuned with the true training images.",
      "votes": null
    },
    {
      "id": "680156",
      "postDate": "11/24/2019 06:02:13",
      "content": "<p>thanks for sharing the solution appreciate it !</p>",
      "rawMarkdown": "thanks for sharing the solution appreciate it !",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 676901,
      "author_name": "veeralakrishna",
      "author_url": "",
      "post_date": "11/19/2019 15:33:03",
      "content": "<p>Congratulations\nThanks for Sharing your Approach <a href=\"/manojprabhaakr\">@manojprabhaakr</a> </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 676915,
      "author_name": "carlolepelaars",
      "author_url": "",
      "post_date": "11/19/2019 15:50:57",
      "content": "<p>Fully agree, it was great to work with this team!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 677259,
      "author_name": "hanjoonchoe",
      "author_url": "",
      "post_date": "11/20/2019 01:15:44",
      "content": "<p>Congratulations!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 678076,
      "author_name": "cdeotte",
      "author_url": "",
      "post_date": "11/21/2019 01:09:08",
      "content": "<p>Congrats Manoj and team. Great job. I checked out your code. I like how you use pseudo labels. That's smart to only use them during stage 1 of training. It helps your net's learning move in the right direction but then the model is fine tuned with the true training images.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 680156,
      "author_name": "ravijoe",
      "author_url": "",
      "post_date": "11/24/2019 06:02:13",
      "content": "<p>thanks for sharing the solution appreciate it !</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "676889": "It has been great to work with @muhakabartay, @carlolepelaars, @roydatascience, @lightnezzofbeing . We have open sourced the code in github @  https://github.com/LightnessOfBeing/Cl0uds.",
    "676901": "Congratulations\nThanks for Sharing your Approach @manojprabhaakr",
    "676915": "Fully agree, it was great to work with this team!",
    "677259": "Congratulations!",
    "678076": "Congrats Manoj and team. Great job. I checked out your code. I like how you use pseudo labels. That's smart to only use them during stage 1 of training. It helps your net's learning move in the right direction but then the model is fine tuned with the true training images.",
    "680156": "thanks for sharing the solution appreciate it !"
  },
  "source": "meta"
}