{
  "id": 172231,
  "title": "TF records",
  "url": "/competitions/landmark-recognition-2020/discussion/172231",
  "author_name": "",
  "post_date": "2020-08-04T07:59:16.108119800Z",
  "votes": 25,
  "comment_count": 10,
  "views": 0,
  "content": "<p>If somebody needs tf records(256x256) for this competition use the below links. its in batches of 3. I also reduced to the dataset to classes with 150+ images. so, there is 890 classes in total<br>\n<a href=\"https://www.kaggle.com/khailashsanthakumar/landmarktfv31\" target=\"_blank\">https://www.kaggle.com/khailashsanthakumar/landmarktfv31</a><br>\n<a href=\"https://www.kaggle.com/khailashsanthakumar/landmarktfv32\" target=\"_blank\">https://www.kaggle.com/khailashsanthakumar/landmarktfv32</a><br>\n<a href=\"https://www.kaggle.com/khailashsanthakumar/landmarktfv33\" target=\"_blank\">https://www.kaggle.com/khailashsanthakumar/landmarktfv33</a><br>\n<a href=\"https://www.kaggle.com/khailashsanthakumar/landmarktfrecordtest\" target=\"_blank\">https://www.kaggle.com/khailashsanthakumar/landmarktfrecordtest</a><br>\nIf its useful please upvote the discussion and dataset</p>",
  "messages": [
    {
      "id": "957306",
      "postDate": "08/04/2020 07:59:16",
      "content": "<p>If somebody needs tf records(256x256) for this competition use the below links. its in batches of 3. I also reduced to the dataset to classes with 150+ images. so, there is 890 classes in total<br>\n<a href=\"https://www.kaggle.com/khailashsanthakumar/landmarktfv31\" target=\"_blank\">https://www.kaggle.com/khailashsanthakumar/landmarktfv31</a><br>\n<a href=\"https://www.kaggle.com/khailashsanthakumar/landmarktfv32\" target=\"_blank\">https://www.kaggle.com/khailashsanthakumar/landmarktfv32</a><br>\n<a href=\"https://www.kaggle.com/khailashsanthakumar/landmarktfv33\" target=\"_blank\">https://www.kaggle.com/khailashsanthakumar/landmarktfv33</a><br>\n<a href=\"https://www.kaggle.com/khailashsanthakumar/landmarktfrecordtest\" target=\"_blank\">https://www.kaggle.com/khailashsanthakumar/landmarktfrecordtest</a><br>\nIf its useful please upvote the discussion and dataset</p>",
      "rawMarkdown": "If somebody needs tf records(256x256) for this competition use the below links. its in batches of 3. I also reduced to the dataset to classes with 150+ images. so, there is 890 classes in total\nhttps://www.kaggle.com/khailashsanthakumar/landmarktfv31\nhttps://www.kaggle.com/khailashsanthakumar/landmarktfv32\nhttps://www.kaggle.com/khailashsanthakumar/landmarktfv33\nhttps://www.kaggle.com/khailashsanthakumar/landmarktfrecordtest\nIf its useful please upvote the discussion and dataset",
      "votes": null
    },
    {
      "id": "959368",
      "postDate": "08/05/2020 14:36:38",
      "content": "<p>Thanks khailash</p>",
      "rawMarkdown": "Thanks khailash",
      "votes": null
    },
    {
      "id": "959661",
      "postDate": "08/05/2020 19:43:11",
      "content": "<p>There has been a small error in the dataset. I promise you that it will fixed tomorrow. Big fan!</p>",
      "rawMarkdown": "There has been a small error in the dataset. I promise you that it will fixed tomorrow. Big fan!",
      "votes": null
    },
    {
      "id": "965647",
      "postDate": "08/10/2020 18:34:47",
      "content": "<p>Cheers! Has the error been fixed? And out-of-curiosity, what was it?</p>",
      "rawMarkdown": "Cheers! Has the error been fixed? And out-of-curiosity, what was it?",
      "votes": null
    },
    {
      "id": "966137",
      "postDate": "08/11/2020 07:11:20",
      "content": "<p>The tf records are updated.</p>",
      "rawMarkdown": "The tf records are updated.",
      "votes": null
    },
    {
      "id": "966138",
      "postDate": "08/11/2020 07:12:12",
      "content": "<p>i didn't convert the output to categorical</p>",
      "rawMarkdown": "i didn't convert the output to categorical",
      "votes": null
    },
    {
      "id": "975252",
      "postDate": "08/18/2020 08:06:08",
      "content": "<p><a href=\"https://www.kaggle.com/khailashsanthakumar\" target=\"_blank\">@khailashsanthakumar</a> can you share code how you created tfrecord ?</p>",
      "rawMarkdown": "khailashsanthakumar can you share code how you created tfrecord ?",
      "votes": null
    },
    {
      "id": "982987",
      "postDate": "08/23/2020 23:17:27",
      "content": "<p>Thanks. This would be great for local code debugging!</p>",
      "rawMarkdown": "Thanks. This would be great for local code debugging!",
      "votes": null
    },
    {
      "id": "986363",
      "postDate": "08/26/2020 12:17:02",
      "content": "<p>Thanks mate! I appreciate the hardwork</p>",
      "rawMarkdown": "Thanks mate! I appreciate the hardwork",
      "votes": null
    },
    {
      "id": "986722",
      "postDate": "08/26/2020 18:18:59",
      "content": "<p>I guess everything works fine now?</p>",
      "rawMarkdown": "I guess everything works fine now?",
      "votes": null
    },
    {
      "id": "986960",
      "postDate": "08/26/2020 22:05:38",
      "content": "<p>Yes. Worked fine for me except that you have to comment out the random seeds in the DELF train.py file</p>",
      "rawMarkdown": "Yes. Worked fine for me except that you have to comment out the random seeds in the DELF train.py file",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 966137,
      "author_name": "khailashsanthakumar",
      "author_url": "",
      "post_date": "08/11/2020 07:11:20",
      "content": "<p>The tf records are updated.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 975252,
      "author_name": "coolcoder22",
      "author_url": "",
      "post_date": "08/18/2020 08:06:08",
      "content": "<p><a href=\"https://www.kaggle.com/khailashsanthakumar\" target=\"_blank\">@khailashsanthakumar</a> can you share code how you created tfrecord ?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 982987,
      "author_name": "chankhavu",
      "author_url": "",
      "post_date": "08/23/2020 23:17:27",
      "content": "<p>Thanks. This would be great for local code debugging!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 986363,
      "author_name": "namantuli",
      "author_url": "",
      "post_date": "08/26/2020 12:17:02",
      "content": "<p>Thanks mate! I appreciate the hardwork</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 959368,
      "author_name": "cdeotte",
      "author_url": "",
      "post_date": "08/05/2020 14:36:38",
      "content": "<p>Thanks khailash</p>",
      "votes": null,
      "replies": [
        {
          "id": 959661,
          "author_name": "khailashsanthakumar",
          "author_url": "",
          "post_date": "08/05/2020 19:43:11",
          "content": "<p>There has been a small error in the dataset. I promise you that it will fixed tomorrow. Big fan!</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 965647,
          "author_name": "yassinealouini",
          "author_url": "",
          "post_date": "08/10/2020 18:34:47",
          "content": "<p>Cheers! Has the error been fixed? And out-of-curiosity, what was it?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 966138,
          "author_name": "khailashsanthakumar",
          "author_url": "",
          "post_date": "08/11/2020 07:12:12",
          "content": "<p>i didn't convert the output to categorical</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 986722,
          "author_name": "yassinealouini",
          "author_url": "",
          "post_date": "08/26/2020 18:18:59",
          "content": "<p>I guess everything works fine now?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 986960,
          "author_name": "namantuli",
          "author_url": "",
          "post_date": "08/26/2020 22:05:38",
          "content": "<p>Yes. Worked fine for me except that you have to comment out the random seeds in the DELF train.py file</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "957306": "If somebody needs tf records(256x256) for this competition use the below links. its in batches of 3. I also reduced to the dataset to classes with 150+ images. so, there is 890 classes in total\nhttps://www.kaggle.com/khailashsanthakumar/landmarktfv31\nhttps://www.kaggle.com/khailashsanthakumar/landmarktfv32\nhttps://www.kaggle.com/khailashsanthakumar/landmarktfv33\nhttps://www.kaggle.com/khailashsanthakumar/landmarktfrecordtest\nIf its useful please upvote the discussion and dataset",
    "959368": "Thanks khailash",
    "959661": "There has been a small error in the dataset. I promise you that it will fixed tomorrow. Big fan!",
    "965647": "Cheers! Has the error been fixed? And out-of-curiosity, what was it?",
    "966137": "The tf records are updated.",
    "966138": "i didn't convert the output to categorical",
    "975252": "khailashsanthakumar can you share code how you created tfrecord ?",
    "982987": "Thanks. This would be great for local code debugging!",
    "986363": "Thanks mate! I appreciate the hardwork",
    "986722": "I guess everything works fine now?",
    "986960": "Yes. Worked fine for me except that you have to comment out the random seeds in the DELF train.py file"
  },
  "source": "meta"
}