{
  "id": 52720,
  "title": "confusion about stage2 ..",
  "url": "/competitions/data-science-bowl-2018/discussion/52720",
  "author_name": "",
  "post_date": "2018-03-22T13:23:28.382225400Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>this link :   <a href=\"https://www.kaggle.com/two-stage-faq\">https://www.kaggle.com/two-stage-faq</a> <br>\nexplain the reason of \"Two-Stage Competition FAQ \"  </p>\n\n<p>Kaggle sometimes runs what are called two-stage competitions. In the first phase, participants train models and submit to a small/temporary leaderboard. In the second phase, they use the same models developed in the first phase to predict on an unseen test set. </p>\n\n<p>But my question is , if i have two models for this  competition ， one for color images, and the other for grayscal images , then i have to upload both of them ..... , would official of the competition run both models excatly as i did in my local workstation ?   like model1 to predict grayscale images, model2 to predict color images?  </p>\n\n<p>very thanks than words can say, if some guy could dispel my confusion .\nand by the way , forget my poor english .</p>",
  "messages": [
    {
      "id": "301202",
      "postDate": "03/22/2018 13:23:28",
      "content": "<p>this link :   <a href=\"https://www.kaggle.com/two-stage-faq\">https://www.kaggle.com/two-stage-faq</a> <br>\nexplain the reason of \"Two-Stage Competition FAQ \"  </p>\n\n<p>Kaggle sometimes runs what are called two-stage competitions. In the first phase, participants train models and submit to a small/temporary leaderboard. In the second phase, they use the same models developed in the first phase to predict on an unseen test set. </p>\n\n<p>But my question is , if i have two models for this  competition ， one for color images, and the other for grayscal images , then i have to upload both of them ..... , would official of the competition run both models excatly as i did in my local workstation ?   like model1 to predict grayscale images, model2 to predict color images?  </p>\n\n<p>very thanks than words can say, if some guy could dispel my confusion .\nand by the way , forget my poor english .</p>",
      "rawMarkdown": "this link :   https://www.kaggle.com/two-stage-faq  \nexplain the reason of \"Two-Stage Competition FAQ \"  \n\nKaggle sometimes runs what are called two-stage competitions. In the first phase, participants train models and submit to a small/temporary leaderboard. In the second phase, they use the same models developed in the first phase to predict on an unseen test set. \n\nBut my question is , if i have two models for this  competition ， one for color images, and the other for grayscal images , then i have to upload both of them ..... , would official of the competition run both models excatly as i did in my local workstation ?   like model1 to predict grayscale images, model2 to predict color images?  \n\nvery thanks than words can say, if some guy could dispel my confusion .\nand by the way , forget my poor english .",
      "votes": null
    },
    {
      "id": "303204",
      "postDate": "03/25/2018 18:39:07",
      "content": "<p>Do you select images manually to feed color or gray networks? If yes, it won't work. Your pipeline should get the images and produce the output, no manual processing is allowed.</p>",
      "rawMarkdown": "Do you select images manually to feed color or gray networks? If yes, it won't work. Your pipeline should get the images and produce the output, no manual processing is allowed.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 303204,
      "author_name": "mrtzmilani",
      "author_url": "",
      "post_date": "03/25/2018 18:39:07",
      "content": "<p>Do you select images manually to feed color or gray networks? If yes, it won't work. Your pipeline should get the images and produce the output, no manual processing is allowed.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "301202": "this link :   https://www.kaggle.com/two-stage-faq  \nexplain the reason of \"Two-Stage Competition FAQ \"  \n\nKaggle sometimes runs what are called two-stage competitions. In the first phase, participants train models and submit to a small/temporary leaderboard. In the second phase, they use the same models developed in the first phase to predict on an unseen test set. \n\nBut my question is , if i have two models for this  competition ， one for color images, and the other for grayscal images , then i have to upload both of them ..... , would official of the competition run both models excatly as i did in my local workstation ?   like model1 to predict grayscale images, model2 to predict color images?  \n\nvery thanks than words can say, if some guy could dispel my confusion .\nand by the way , forget my poor english .",
    "303204": "Do you select images manually to feed color or gray networks? If yes, it won't work. Your pipeline should get the images and produce the output, no manual processing is allowed."
  },
  "source": "meta"
}