{
  "id": 89593,
  "title": "Is this process permitted in this competition?",
  "url": "/competitions/imet-2019-fgvc6/discussion/89593",
  "author_name": "",
  "post_date": "2019-04-16T04:09:27.862974200Z",
  "votes": 3,
  "comment_count": 8,
  "views": 0,
  "content": "<p>I want to check something before much effort to this competition.</p>\n\n<p>After reading discussions, I set my plan like this.</p>\n\n<p>Please verify my plan, Thanks!</p>\n\n<p>(1). Develop my model (pretrained or new own model)</p>\n\n<p>(2) Train the model with competition datasets (not using any external data)</p>\n\n<p>I will train the model by using my GPU (offline)</p>\n\n<p>(3). Upload weights on the kernel in this competition</p>\n\n<p>(4) Not training in kernel. Just predict the labels of testset using uploaded weight.</p>\n\n<p>Is this whole process(1) - (4) OK for this competition?</p>\n\n<p>Thanks!</p>",
  "messages": [
    {
      "id": "517468",
      "postDate": "04/16/2019 04:09:27",
      "content": "<p>I want to check something before much effort to this competition.</p>\n\n<p>After reading discussions, I set my plan like this.</p>\n\n<p>Please verify my plan, Thanks!</p>\n\n<p>(1). Develop my model (pretrained or new own model)</p>\n\n<p>(2) Train the model with competition datasets (not using any external data)</p>\n\n<p>I will train the model by using my GPU (offline)</p>\n\n<p>(3). Upload weights on the kernel in this competition</p>\n\n<p>(4) Not training in kernel. Just predict the labels of testset using uploaded weight.</p>\n\n<p>Is this whole process(1) - (4) OK for this competition?</p>\n\n<p>Thanks!</p>",
      "rawMarkdown": "I want to check something before much effort to this competition.\n\nAfter reading discussions, I set my plan like this.\n\nPlease verify my plan, Thanks!\n\n(1). Develop my model (pretrained or new own model)\n\n(2) Train the model with competition datasets (not using any external data)\n\nI will train the model by using my GPU (offline)\n\n(3). Upload weights on the kernel in this competition\n\n(4) Not training in kernel. Just predict the labels of testset using uploaded weight.\n\nIs this whole process(1) - (4) OK for this competition?\n\nThanks!",
      "votes": null
    },
    {
      "id": "517473",
      "postDate": "04/16/2019 04:20:33",
      "content": "<p>Based on the discussions, I think  this is totally fine, I'm working in this way.  And I'm using the pretrained models from <a href=\"https://www.kaggle.com/igorkrashenyi/pytorch-model-zoo\">https://www.kaggle.com/igorkrashenyi/pytorch-model-zoo</a>. (Really thanks to this great dataset.</p>",
      "rawMarkdown": "Based on the discussions, I think  this is totally fine, I'm working in this way.  And I'm using the pretrained models from https://www.kaggle.com/igorkrashenyi/pytorch-model-zoo. (Really thanks to this great dataset.",
      "votes": null
    },
    {
      "id": "517783",
      "postDate": "04/16/2019 14:29:43",
      "content": "<p>Hi YouHan, I also believe that your plan is OK.</p>",
      "rawMarkdown": "Hi YouHan, I also believe that your plan is OK.",
      "votes": null
    },
    {
      "id": "518312",
      "postDate": "04/17/2019 04:37:22",
      "content": "<p>Very thanks ! I feel a sense of relief. I will keep do this!\nHope you get a good score!</p>",
      "rawMarkdown": "Very thanks ! I feel a sense of relief. I will keep do this!\nHope you get a good score!",
      "votes": null
    },
    {
      "id": "518313",
      "postDate": "04/17/2019 04:37:49",
      "content": "<p>Oh, Good to hear! Let's train models :) Thanks!</p>",
      "rawMarkdown": "Oh, Good to hear! Let's train models :) Thanks!",
      "votes": null
    },
    {
      "id": "518506",
      "postDate": "04/17/2019 11:42:00",
      "content": "<p>I see no problem with your plan :)</p>",
      "rawMarkdown": "I see no problem with your plan :)",
      "votes": null
    },
    {
      "id": "523308",
      "postDate": "04/26/2019 01:46:46",
      "content": "<p>Hello, I'm new to Kaggle competitions, and I‘m wondering how to upload my models trained offline in the kernels.\nLooking forward to your answer, Thanks!</p>",
      "rawMarkdown": "Hello, I'm new to Kaggle competitions, and I‘m wondering how to upload my models trained offline in the kernels.\nLooking forward to your answer, Thanks!",
      "votes": null
    },
    {
      "id": "523310",
      "postDate": "04/26/2019 01:53:12",
      "content": "<p>Upload models to your own dataset first, in the top of the page you can see \"Dataset\"---&gt;New Dataset. After you create your kernel, you can add your dataset and then load your models. Have fun in kaggle!!</p>",
      "rawMarkdown": "Upload models to your own dataset first, in the top of the page you can see \"Dataset\"---&gt;New Dataset. After you create your kernel, you can add your dataset and then load your models. Have fun in kaggle!!",
      "votes": null
    },
    {
      "id": "523390",
      "postDate": "04/26/2019 06:37:56",
      "content": "<p>Thanks for your reply! I upload my models successfully in this way.</p>",
      "rawMarkdown": "Thanks for your reply! I upload my models successfully in this way.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 517473,
      "author_name": "jionie",
      "author_url": "",
      "post_date": "04/16/2019 04:20:33",
      "content": "<p>Based on the discussions, I think  this is totally fine, I'm working in this way.  And I'm using the pretrained models from <a href=\"https://www.kaggle.com/igorkrashenyi/pytorch-model-zoo\">https://www.kaggle.com/igorkrashenyi/pytorch-model-zoo</a>. (Really thanks to this great dataset.</p>",
      "votes": null,
      "replies": [
        {
          "id": 518313,
          "author_name": "youhanlee",
          "author_url": "",
          "post_date": "04/17/2019 04:37:49",
          "content": "<p>Oh, Good to hear! Let's train models :) Thanks!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 517783,
      "author_name": "ratthachat",
      "author_url": "",
      "post_date": "04/16/2019 14:29:43",
      "content": "<p>Hi YouHan, I also believe that your plan is OK.</p>",
      "votes": null,
      "replies": [
        {
          "id": 518312,
          "author_name": "youhanlee",
          "author_url": "",
          "post_date": "04/17/2019 04:37:22",
          "content": "<p>Very thanks ! I feel a sense of relief. I will keep do this!\nHope you get a good score!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 518506,
      "author_name": "endo9e7a6",
      "author_url": "",
      "post_date": "04/17/2019 11:42:00",
      "content": "<p>I see no problem with your plan :)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 523308,
      "author_name": "haolanx",
      "author_url": "",
      "post_date": "04/26/2019 01:46:46",
      "content": "<p>Hello, I'm new to Kaggle competitions, and I‘m wondering how to upload my models trained offline in the kernels.\nLooking forward to your answer, Thanks!</p>",
      "votes": null,
      "replies": [
        {
          "id": 523310,
          "author_name": "jionie",
          "author_url": "",
          "post_date": "04/26/2019 01:53:12",
          "content": "<p>Upload models to your own dataset first, in the top of the page you can see \"Dataset\"---&gt;New Dataset. After you create your kernel, you can add your dataset and then load your models. Have fun in kaggle!!</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 523390,
          "author_name": "haolanx",
          "author_url": "",
          "post_date": "04/26/2019 06:37:56",
          "content": "<p>Thanks for your reply! I upload my models successfully in this way.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "517468": "I want to check something before much effort to this competition.\n\nAfter reading discussions, I set my plan like this.\n\nPlease verify my plan, Thanks!\n\n(1). Develop my model (pretrained or new own model)\n\n(2) Train the model with competition datasets (not using any external data)\n\nI will train the model by using my GPU (offline)\n\n(3). Upload weights on the kernel in this competition\n\n(4) Not training in kernel. Just predict the labels of testset using uploaded weight.\n\nIs this whole process(1) - (4) OK for this competition?\n\nThanks!",
    "517473": "Based on the discussions, I think  this is totally fine, I'm working in this way.  And I'm using the pretrained models from https://www.kaggle.com/igorkrashenyi/pytorch-model-zoo. (Really thanks to this great dataset.",
    "517783": "Hi YouHan, I also believe that your plan is OK.",
    "518312": "Very thanks ! I feel a sense of relief. I will keep do this!\nHope you get a good score!",
    "518313": "Oh, Good to hear! Let's train models :) Thanks!",
    "518506": "I see no problem with your plan :)",
    "523308": "Hello, I'm new to Kaggle competitions, and I‘m wondering how to upload my models trained offline in the kernels.\nLooking forward to your answer, Thanks!",
    "523310": "Upload models to your own dataset first, in the top of the page you can see \"Dataset\"---&gt;New Dataset. After you create your kernel, you can add your dataset and then load your models. Have fun in kaggle!!",
    "523390": "Thanks for your reply! I upload my models successfully in this way."
  },
  "source": "meta"
}