{
  "id": 168073,
  "title": "Learn Deep learning by competing in SIIM-ISIC Melanoma Classification.",
  "url": "/competitions/siim-isic-melanoma-classification/discussion/168073",
  "author_name": "",
  "post_date": "2020-07-19T04:50:45.916186900Z",
  "votes": 10,
  "comment_count": 5,
  "views": 0,
  "content": "<p>Hi Friends,</p>\n\n<p><a href=\"https://jarvislabs.ai/\">We</a> are working on creating a free online course to help anyone new to deep learning or Kaggle to get a medal. The course is inspired by fast.ai's top-down approach to learning. We launched the first set of videos <a href=\"https://www.youtube.com/playlist?list=PLexqeSjf_hzPUygXwHyDUkIpU9d1xFCXs\">here</a> that explains the kernel I posted recently <a href=\"https://www.kaggle.com/vishnus/a-simple-pytorch-starter-code-single-fold-93\">here</a>. We are planning to use PyTorch for the course and may migrate to PyTorch lightning for using features like mixed-precision, gradient accumulation.</p>\n\n<p>We will be adding more videos to explain each component of the Kernel in detail. </p>\n\n<p>In addition to the existing kernel, I will be sharing more videos that explain potential ideas that can be tried to improve the LB score. Since it is an ongoing competition I will share these ideas responsibly that means we will not provide a complete solution. </p>\n\n<p>If you are interested please watch through the videos and let us know how we can help you in achieving your 1st medal (Ex: If you do not understand some part of the video, or if you want to understand some concept and If I know it). </p>\n\n<p>Note: This is the first time we are doing this, so please share your feedback on how we can improve the course to make it more useful.</p>",
  "messages": [
    {
      "id": "935101",
      "postDate": "07/19/2020 04:50:45",
      "content": "<p>Hi Friends,</p>\n\n<p><a href=\"https://jarvislabs.ai/\">We</a> are working on creating a free online course to help anyone new to deep learning or Kaggle to get a medal. The course is inspired by fast.ai's top-down approach to learning. We launched the first set of videos <a href=\"https://www.youtube.com/playlist?list=PLexqeSjf_hzPUygXwHyDUkIpU9d1xFCXs\">here</a> that explains the kernel I posted recently <a href=\"https://www.kaggle.com/vishnus/a-simple-pytorch-starter-code-single-fold-93\">here</a>. We are planning to use PyTorch for the course and may migrate to PyTorch lightning for using features like mixed-precision, gradient accumulation.</p>\n\n<p>We will be adding more videos to explain each component of the Kernel in detail. </p>\n\n<p>In addition to the existing kernel, I will be sharing more videos that explain potential ideas that can be tried to improve the LB score. Since it is an ongoing competition I will share these ideas responsibly that means we will not provide a complete solution. </p>\n\n<p>If you are interested please watch through the videos and let us know how we can help you in achieving your 1st medal (Ex: If you do not understand some part of the video, or if you want to understand some concept and If I know it). </p>\n\n<p>Note: This is the first time we are doing this, so please share your feedback on how we can improve the course to make it more useful.</p>",
      "rawMarkdown": "Hi Friends,\n\n[We](https://jarvislabs.ai/) are working on creating a free online course to help anyone new to deep learning or Kaggle to get a medal. The course is inspired by fast.ai's top-down approach to learning. We launched the first set of videos [here](https://www.youtube.com/playlist?list=PLexqeSjf_hzPUygXwHyDUkIpU9d1xFCXs) that explains the kernel I posted recently [here](https://www.kaggle.com/vishnus/a-simple-pytorch-starter-code-single-fold-93). We are planning to use PyTorch for the course and may migrate to PyTorch lightning for using features like mixed-precision, gradient accumulation.\n\nWe will be adding more videos to explain each component of the Kernel in detail. \n\nIn addition to the existing kernel, I will be sharing more videos that explain potential ideas that can be tried to improve the LB score. Since it is an ongoing competition I will share these ideas responsibly that means we will not provide a complete solution. \n\nIf you are interested please watch through the videos and let us know how we can help you in achieving your 1st medal (Ex: If you do not understand some part of the video, or if you want to understand some concept and If I know it). \n\nNote: This is the first time we are doing this, so please share your feedback on how we can improve the course to make it more useful.",
      "votes": null
    },
    {
      "id": "935383",
      "postDate": "07/19/2020 10:46:53",
      "content": "<p>Sounds fishy - \"let us know how we can help you in achieving your 1st medal\" </p>",
      "rawMarkdown": "Sounds fishy - \"let us know how we can help you in achieving your 1st medal\"",
      "votes": null
    },
    {
      "id": "935390",
      "postDate": "07/19/2020 10:56:31",
      "content": "<p>The idea is give enough content that can help in getting a medal for someone who is completely new to Kaggle. It could be myself from 2/3 years back. No bad intentions. All the content is public and no private sharing. </p>",
      "rawMarkdown": "The idea is give enough content that can help in getting a medal for someone who is completely new to Kaggle. It could be myself from 2/3 years back. No bad intentions. All the content is public and no private sharing.",
      "votes": null
    },
    {
      "id": "935398",
      "postDate": "07/19/2020 11:04:19",
      "content": "<p>In that case I really wish you good luck :)</p>\n\n<p>I apologise for misunderstanding at first.</p>",
      "rawMarkdown": "In that case I really wish you good luck :)\n\nI apologise for misunderstanding at first.",
      "votes": null
    },
    {
      "id": "935407",
      "postDate": "07/19/2020 11:10:53",
      "content": "<p>No problem. </p>",
      "rawMarkdown": "No problem.",
      "votes": null
    },
    {
      "id": "935571",
      "postDate": "07/19/2020 13:53:14",
      "content": "<p>Added a few examples so that others do not misunderstand.</p>",
      "rawMarkdown": "Added a few examples so that others do not misunderstand.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 935383,
      "author_name": "aroraaman",
      "author_url": "",
      "post_date": "07/19/2020 10:46:53",
      "content": "<p>Sounds fishy - \"let us know how we can help you in achieving your 1st medal\" </p>",
      "votes": null,
      "replies": [
        {
          "id": 935390,
          "author_name": "vishnus",
          "author_url": "",
          "post_date": "07/19/2020 10:56:31",
          "content": "<p>The idea is give enough content that can help in getting a medal for someone who is completely new to Kaggle. It could be myself from 2/3 years back. No bad intentions. All the content is public and no private sharing. </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 935398,
          "author_name": "aroraaman",
          "author_url": "",
          "post_date": "07/19/2020 11:04:19",
          "content": "<p>In that case I really wish you good luck :)</p>\n\n<p>I apologise for misunderstanding at first.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 935407,
          "author_name": "vishnus",
          "author_url": "",
          "post_date": "07/19/2020 11:10:53",
          "content": "<p>No problem. </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 935571,
          "author_name": "vishnus",
          "author_url": "",
          "post_date": "07/19/2020 13:53:14",
          "content": "<p>Added a few examples so that others do not misunderstand.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "935101": "Hi Friends,\n\n[We](https://jarvislabs.ai/) are working on creating a free online course to help anyone new to deep learning or Kaggle to get a medal. The course is inspired by fast.ai's top-down approach to learning. We launched the first set of videos [here](https://www.youtube.com/playlist?list=PLexqeSjf_hzPUygXwHyDUkIpU9d1xFCXs) that explains the kernel I posted recently [here](https://www.kaggle.com/vishnus/a-simple-pytorch-starter-code-single-fold-93). We are planning to use PyTorch for the course and may migrate to PyTorch lightning for using features like mixed-precision, gradient accumulation.\n\nWe will be adding more videos to explain each component of the Kernel in detail. \n\nIn addition to the existing kernel, I will be sharing more videos that explain potential ideas that can be tried to improve the LB score. Since it is an ongoing competition I will share these ideas responsibly that means we will not provide a complete solution. \n\nIf you are interested please watch through the videos and let us know how we can help you in achieving your 1st medal (Ex: If you do not understand some part of the video, or if you want to understand some concept and If I know it). \n\nNote: This is the first time we are doing this, so please share your feedback on how we can improve the course to make it more useful.",
    "935383": "Sounds fishy - \"let us know how we can help you in achieving your 1st medal\"",
    "935390": "The idea is give enough content that can help in getting a medal for someone who is completely new to Kaggle. It could be myself from 2/3 years back. No bad intentions. All the content is public and no private sharing.",
    "935398": "In that case I really wish you good luck :)\n\nI apologise for misunderstanding at first.",
    "935407": "No problem.",
    "935571": "Added a few examples so that others do not misunderstand."
  },
  "source": "meta"
}