{
  "id": 133181,
  "title": "[placeholder] tpu starter-kit",
  "url": "/competitions/flower-classification-with-tpus/discussion/133181",
  "author_name": "",
  "post_date": "2020-03-01T06:20:12.400844500Z",
  "votes": 5,
  "comment_count": 6,
  "views": 0,
  "content": "<p>this is probably a good project for TPU</p>\n\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F113660%2Ff38847066978751fa3e64aa860a9bdd1%2FSelection_101.png?generation=1583043596439774&amp;alt=media\" alt=\"\"></p>",
  "messages": [
    {
      "id": "760335",
      "postDate": "03/01/2020 06:20:12",
      "content": "<p>this is probably a good project for TPU</p>\n\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F113660%2Ff38847066978751fa3e64aa860a9bdd1%2FSelection_101.png?generation=1583043596439774&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "this is probably a good project for TPU\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F113660%2Ff38847066978751fa3e64aa860a9bdd1%2FSelection_101.png?generation=1583043596439774&amp;alt=media)",
      "votes": null
    },
    {
      "id": "760336",
      "postDate": "03/01/2020 06:20:35",
      "content": "<p>to be updated ...</p>",
      "rawMarkdown": "to be updated ...",
      "votes": null
    },
    {
      "id": "760585",
      "postDate": "03/01/2020 13:46:40",
      "content": "<p>We´ll be waiting for your update.</p>",
      "rawMarkdown": "We´ll be waiting for your update.",
      "votes": null
    },
    {
      "id": "797988",
      "postDate": "04/05/2020 04:52:59",
      "content": "<p>google just open source this code yesterday!\n<a href=\"https://github.com/google-research/simclr\">https://github.com/google-research/simclr</a></p>",
      "rawMarkdown": "google just open source this code yesterday!\nhttps://github.com/google-research/simclr",
      "votes": null
    },
    {
      "id": "798497",
      "postDate": "04/05/2020 14:51:36",
      "content": "<p>Thanks for sharing it Heng Cherkeng and for always help us Kagglers that want to learn. Your experience, knowledge and kindness is highly appreciated in our community. </p>",
      "rawMarkdown": "Thanks for sharing it Heng Cherkeng and for always help us Kagglers that want to learn. Your experience, knowledge and kindness is highly appreciated in our community.",
      "votes": null
    },
    {
      "id": "798576",
      "postDate": "04/05/2020 16:50:18",
      "content": "<p>this seems to a better paper that beats simCLR</p>\n\n<p><a href=\"https://arxiv.org/abs/2003.10580\">https://arxiv.org/abs/2003.10580</a>\nMeta Pseudo Labels\nHieu Pham, Qizhe Xie, Zihang Dai, Quoc V. Le</p>",
      "rawMarkdown": "this seems to a better paper that beats simCLR\n\nhttps://arxiv.org/abs/2003.10580\nMeta Pseudo Labels\nHieu Pham, Qizhe Xie, Zihang Dai, Quoc V. Le",
      "votes": null
    },
    {
      "id": "799011",
      "postDate": "04/06/2020 04:44:01",
      "content": "<p>Been wanting to learn about CPC, CPC-v2, MoCo, SimCLR, MoCo-v2, and other self-supervised learning techniques?</p>\n\n<p>Wonderful lecture by \n@Aravind7694\n for the \n@UCBerkeley\n Deep Unsupervised Learning class:</p>\n\n<p><a href=\"https://www.youtube.com/watch?v=dMUes74-nYY\">https://www.youtube.com/watch?v=dMUes74-nYY</a></p>",
      "rawMarkdown": "Been wanting to learn about CPC, CPC-v2, MoCo, SimCLR, MoCo-v2, and other self-supervised learning techniques?\n\nWonderful lecture by \n@Aravind7694\n for the \n@UCBerkeley\n Deep Unsupervised Learning class:\n\nhttps://www.youtube.com/watch?v=dMUes74-nYY",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 760336,
      "author_name": "hengck23",
      "author_url": "",
      "post_date": "03/01/2020 06:20:35",
      "content": "<p>to be updated ...</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 760585,
      "author_name": "mpwolke",
      "author_url": "",
      "post_date": "03/01/2020 13:46:40",
      "content": "<p>We´ll be waiting for your update.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 797988,
      "author_name": "hengck23",
      "author_url": "",
      "post_date": "04/05/2020 04:52:59",
      "content": "<p>google just open source this code yesterday!\n<a href=\"https://github.com/google-research/simclr\">https://github.com/google-research/simclr</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 798497,
          "author_name": "mpwolke",
          "author_url": "",
          "post_date": "04/05/2020 14:51:36",
          "content": "<p>Thanks for sharing it Heng Cherkeng and for always help us Kagglers that want to learn. Your experience, knowledge and kindness is highly appreciated in our community. </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 799011,
          "author_name": "hengck23",
          "author_url": "",
          "post_date": "04/06/2020 04:44:01",
          "content": "<p>Been wanting to learn about CPC, CPC-v2, MoCo, SimCLR, MoCo-v2, and other self-supervised learning techniques?</p>\n\n<p>Wonderful lecture by \n@Aravind7694\n for the \n@UCBerkeley\n Deep Unsupervised Learning class:</p>\n\n<p><a href=\"https://www.youtube.com/watch?v=dMUes74-nYY\">https://www.youtube.com/watch?v=dMUes74-nYY</a></p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 798576,
      "author_name": "hengck23",
      "author_url": "",
      "post_date": "04/05/2020 16:50:18",
      "content": "<p>this seems to a better paper that beats simCLR</p>\n\n<p><a href=\"https://arxiv.org/abs/2003.10580\">https://arxiv.org/abs/2003.10580</a>\nMeta Pseudo Labels\nHieu Pham, Qizhe Xie, Zihang Dai, Quoc V. Le</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "760335": "this is probably a good project for TPU\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F113660%2Ff38847066978751fa3e64aa860a9bdd1%2FSelection_101.png?generation=1583043596439774&amp;alt=media)",
    "760336": "to be updated ...",
    "760585": "We´ll be waiting for your update.",
    "797988": "google just open source this code yesterday!\nhttps://github.com/google-research/simclr",
    "798497": "Thanks for sharing it Heng Cherkeng and for always help us Kagglers that want to learn. Your experience, knowledge and kindness is highly appreciated in our community.",
    "798576": "this seems to a better paper that beats simCLR\n\nhttps://arxiv.org/abs/2003.10580\nMeta Pseudo Labels\nHieu Pham, Qizhe Xie, Zihang Dai, Quoc V. Le",
    "799011": "Been wanting to learn about CPC, CPC-v2, MoCo, SimCLR, MoCo-v2, and other self-supervised learning techniques?\n\nWonderful lecture by \n@Aravind7694\n for the \n@UCBerkeley\n Deep Unsupervised Learning class:\n\nhttps://www.youtube.com/watch?v=dMUes74-nYY"
  },
  "source": "meta"
}