{
  "id": 199730,
  "title": "Is submission from Google Colab possible for this competition?",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/199730",
  "author_name": "",
  "post_date": "2020-11-27T04:40:21.148805900Z",
  "votes": 2,
  "comment_count": 5,
  "views": 0,
  "content": "<p>My GPU quota is exceeded for this week and I want to submit inference kernel ran on Google Colab. Is there a way to submit it via API for kernels only competitions?</p>",
  "messages": [
    {
      "id": "1092669",
      "postDate": "11/27/2020 04:40:21",
      "content": "<p>My GPU quota is exceeded for this week and I want to submit inference kernel ran on Google Colab. Is there a way to submit it via API for kernels only competitions?</p>",
      "rawMarkdown": "My GPU quota is exceeded for this week and I want to submit inference kernel ran on Google Colab. Is there a way to submit it via API for kernels only competitions?",
      "votes": null
    },
    {
      "id": "1092771",
      "postDate": "11/27/2020 06:41:21",
      "content": "<p>You can submit via CPU. Just train the model first and load it.</p>",
      "rawMarkdown": "You can submit via CPU. Just train the model first and load it.",
      "votes": null
    },
    {
      "id": "1092775",
      "postDate": "11/27/2020 06:44:29",
      "content": "<p>Yeah that's one option but it'll take too much time</p>",
      "rawMarkdown": "Yeah that's one option but it'll take too much time",
      "votes": null
    },
    {
      "id": "1092853",
      "postDate": "11/27/2020 08:14:04",
      "content": "<p>There is only 1 image in the test set. If you are talking about waiting for result is too much time, instead of waiting for it you can continue training.</p>",
      "rawMarkdown": "There is only 1 image in the test set. If you are talking about waiting for result is too much time, instead of waiting for it you can continue training.",
      "votes": null
    },
    {
      "id": "1092856",
      "postDate": "11/27/2020 08:17:11",
      "content": "<p>I'm talking about the time taken by submission to competition. At that time it don't have one image.</p>",
      "rawMarkdown": "I'm talking about the time taken by submission to competition. At that time it don't have one image.",
      "votes": null
    },
    {
      "id": "1093866",
      "postDate": "11/28/2020 04:56:23",
      "content": "<p>You can simply train as much as you wish on Colab, and then upload it for further train, or just submission :</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4892590%2F1ebdbd477f97f83db230638a83362a61%2Funtitled.png?generation=1606539348150604&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "You can simply train as much as you wish on Colab, and then upload it for further train, or just submission :\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4892590%2F1ebdbd477f97f83db230638a83362a61%2Funtitled.png?generation=1606539348150604&alt=media)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1092771,
      "author_name": "aeryss",
      "author_url": "",
      "post_date": "11/27/2020 06:41:21",
      "content": "<p>You can submit via CPU. Just train the model first and load it.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1092775,
          "author_name": "kaushal2896",
          "author_url": "",
          "post_date": "11/27/2020 06:44:29",
          "content": "<p>Yeah that's one option but it'll take too much time</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1092853,
          "author_name": "aeryss",
          "author_url": "",
          "post_date": "11/27/2020 08:14:04",
          "content": "<p>There is only 1 image in the test set. If you are talking about waiting for result is too much time, instead of waiting for it you can continue training.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1092856,
          "author_name": "kaushal2896",
          "author_url": "",
          "post_date": "11/27/2020 08:17:11",
          "content": "<p>I'm talking about the time taken by submission to competition. At that time it don't have one image.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1093866,
      "author_name": "daniboy370",
      "author_url": "",
      "post_date": "11/28/2020 04:56:23",
      "content": "<p>You can simply train as much as you wish on Colab, and then upload it for further train, or just submission :</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4892590%2F1ebdbd477f97f83db230638a83362a61%2Funtitled.png?generation=1606539348150604&amp;alt=media\" alt=\"\"></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1092669": "My GPU quota is exceeded for this week and I want to submit inference kernel ran on Google Colab. Is there a way to submit it via API for kernels only competitions?",
    "1092771": "You can submit via CPU. Just train the model first and load it.",
    "1092775": "Yeah that's one option but it'll take too much time",
    "1092853": "There is only 1 image in the test set. If you are talking about waiting for result is too much time, instead of waiting for it you can continue training.",
    "1092856": "I'm talking about the time taken by submission to competition. At that time it don't have one image.",
    "1093866": "You can simply train as much as you wish on Colab, and then upload it for further train, or just submission :\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4892590%2F1ebdbd477f97f83db230638a83362a61%2Funtitled.png?generation=1606539348150604&alt=media)"
  },
  "source": "meta"
}