{
  "id": 33456,
  "title": "updated labels in train set ",
  "url": "/competitions/intel-mobileodt-cervical-cancer-screening/discussion/33456",
  "author_name": "",
  "post_date": "2017-05-23T23:12:01.184711100Z",
  "votes": 10,
  "comment_count": 7,
  "views": 0,
  "content": "<p>Just noticed there was an update for labels on 5/22,  </p>\n\n<p>UPDATED 5/22: We revisited these images and found that we need to update 3 labels:\ntrain/Type_1/80 - type 3\ntrain/Type_3/968 - type 1\ntrain/Type_3/1120 - type 1</p>\n\n<p>However, there is no 80.jpg in train/Type_1, <br>\nshould it be 81.jpg? </p>\n\n<p>thanks </p>",
  "messages": [
    {
      "id": "185058",
      "postDate": "05/23/2017 23:12:01",
      "content": "<p>Just noticed there was an update for labels on 5/22,  </p>\n\n<p>UPDATED 5/22: We revisited these images and found that we need to update 3 labels:\ntrain/Type_1/80 - type 3\ntrain/Type_3/968 - type 1\ntrain/Type_3/1120 - type 1</p>\n\n<p>However, there is no 80.jpg in train/Type_1, <br>\nshould it be 81.jpg? </p>\n\n<p>thanks </p>",
      "rawMarkdown": "Just noticed there was an update for labels on 5/22,  \n\nUPDATED 5/22: We revisited these images and found that we need to update 3 labels:\ntrain/Type_1/80 - type 3\ntrain/Type_3/968 - type 1\ntrain/Type_3/1120 - type 1\n\nHowever, there is no 80.jpg in train/Type_1,   \nshould it be 81.jpg? \n\nthanks",
      "votes": null
    },
    {
      "id": "185091",
      "postDate": "05/24/2017 02:33:02",
      "content": "<p>80.jpg is in Type_2. I think it is another mistake.</p>",
      "rawMarkdown": "80.jpg is in Type_2. I think it is another mistake.",
      "votes": null
    },
    {
      "id": "185258",
      "postDate": "05/24/2017 17:44:39",
      "content": "<p>Is this true? Did they update the training data without making an announcement?</p>",
      "rawMarkdown": "Is this true? Did they update the training data without making an announcement?",
      "votes": null
    },
    {
      "id": "185843",
      "postDate": "05/26/2017 02:23:04",
      "content": "<p>There are two images named as 80.jpg (one is in train.7z, the other is in additional train set). I'm also very confused about it. </p>",
      "rawMarkdown": "There are two images named as 80.jpg (one is in train.7z, the other is in additional train set). I'm also very confused about it.",
      "votes": null
    },
    {
      "id": "185844",
      "postDate": "05/26/2017 02:25:28",
      "content": "<p>They announced at the bottom in \"Data\" page</p>",
      "rawMarkdown": "They announced at the bottom in \"Data\" page",
      "votes": null
    },
    {
      "id": "186266",
      "postDate": "05/27/2017 09:43:27",
      "content": "<p>This can hardly be considered as 'announcement' ... </p>\n\n<p>@Admins - training models, pre-processing this data is a long process. We all would appreciate a clearer communication </p>",
      "rawMarkdown": "This can hardly be considered as 'announcement' ... \n\n@Admins - training models, pre-processing this data is a long process. We all would appreciate a clearer communication",
      "votes": null
    },
    {
      "id": "186482",
      "postDate": "05/28/2017 08:26:58",
      "content": "<p>it's train/Type2/80.jpg\nsee this kernel and the comments below from Yau Ben-Or:\n<a href=\"https://www.kaggle.com/scottykwok/making-sense-out-of-some-difficult-samples\">https://www.kaggle.com/scottykwok/making-sense-out-of-some-difficult-samples</a></p>",
      "rawMarkdown": "it's train/Type2/80.jpg\nsee this kernel and the comments below from Yau Ben-Or:\nhttps://www.kaggle.com/scottykwok/making-sense-out-of-some-difficult-samples",
      "votes": null
    },
    {
      "id": "186954",
      "postDate": "05/29/2017 21:15:00",
      "content": "<p>The main questions is: \"Test part 2 has the same bad labeling or not?\" If yes, than it's better not to move these files then )</p>",
      "rawMarkdown": "The main questions is: \"Test part 2 has the same bad labeling or not?\" If yes, than it's better not to move these files then )",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 185091,
      "author_name": "chen1993",
      "author_url": "",
      "post_date": "05/24/2017 02:33:02",
      "content": "<p>80.jpg is in Type_2. I think it is another mistake.</p>",
      "votes": null,
      "replies": [
        {
          "id": 185843,
          "author_name": "hanyangkyushu",
          "author_url": "",
          "post_date": "05/26/2017 02:23:04",
          "content": "<p>There are two images named as 80.jpg (one is in train.7z, the other is in additional train set). I'm also very confused about it. </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 185258,
      "author_name": "gkericks",
      "author_url": "",
      "post_date": "05/24/2017 17:44:39",
      "content": "<p>Is this true? Did they update the training data without making an announcement?</p>",
      "votes": null,
      "replies": [
        {
          "id": 185844,
          "author_name": "hanyangkyushu",
          "author_url": "",
          "post_date": "05/26/2017 02:25:28",
          "content": "<p>They announced at the bottom in \"Data\" page</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 186266,
          "author_name": "alchemist",
          "author_url": "",
          "post_date": "05/27/2017 09:43:27",
          "content": "<p>This can hardly be considered as 'announcement' ... </p>\n\n<p>@Admins - training models, pre-processing this data is a long process. We all would appreciate a clearer communication </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 186482,
      "author_name": "steelrose",
      "author_url": "",
      "post_date": "05/28/2017 08:26:58",
      "content": "<p>it's train/Type2/80.jpg\nsee this kernel and the comments below from Yau Ben-Or:\n<a href=\"https://www.kaggle.com/scottykwok/making-sense-out-of-some-difficult-samples\">https://www.kaggle.com/scottykwok/making-sense-out-of-some-difficult-samples</a></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 186954,
      "author_name": "zfturbo",
      "author_url": "",
      "post_date": "05/29/2017 21:15:00",
      "content": "<p>The main questions is: \"Test part 2 has the same bad labeling or not?\" If yes, than it's better not to move these files then )</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "185058": "Just noticed there was an update for labels on 5/22,  \n\nUPDATED 5/22: We revisited these images and found that we need to update 3 labels:\ntrain/Type_1/80 - type 3\ntrain/Type_3/968 - type 1\ntrain/Type_3/1120 - type 1\n\nHowever, there is no 80.jpg in train/Type_1,   \nshould it be 81.jpg? \n\nthanks",
    "185091": "80.jpg is in Type_2. I think it is another mistake.",
    "185258": "Is this true? Did they update the training data without making an announcement?",
    "185843": "There are two images named as 80.jpg (one is in train.7z, the other is in additional train set). I'm also very confused about it.",
    "185844": "They announced at the bottom in \"Data\" page",
    "186266": "This can hardly be considered as 'announcement' ... \n\n@Admins - training models, pre-processing this data is a long process. We all would appreciate a clearer communication",
    "186482": "it's train/Type2/80.jpg\nsee this kernel and the comments below from Yau Ben-Or:\nhttps://www.kaggle.com/scottykwok/making-sense-out-of-some-difficult-samples",
    "186954": "The main questions is: \"Test part 2 has the same bad labeling or not?\" If yes, than it's better not to move these files then )"
  },
  "source": "meta"
}