{
  "id": 109701,
  "title": "Deleting corrupted data",
  "url": "/competitions/rsna-intracranial-hemorrhage-detection/discussion/109701",
  "author_name": "Meyk",
  "post_date": "2019-09-21T13:30:59.643000",
  "votes": 8,
  "comment_count": 5,
  "views": 0,
  "content": "<p>Hi guys, </p>\n\n<p>there is a problem with my train data set ... I have downloaded it several times without any progress (in case of corrupted images). </p>\n\n<p>Each time I am running a script it fails at the corrupted image (frustrating) with an error:</p>\n\n<blockquote>\n  <p>\"ValueError: The length of the pixel data in the dataset (153710 bytes) doesn't match the expected length (524288 bytes). The dataset may be corrupted or there may be an issue with the pixel data handler.\"</p>\n</blockquote>\n\n<p>Do you have any idea how to get rid of all corrupted images at once? </p>",
  "messages": [
    {
      "id": 631157,
      "postDate": "2019-09-21T13:30:59.643Z",
      "content": "<p>Hi guys, </p>\n\n<p>there is a problem with my train data set ... I have downloaded it several times without any progress (in case of corrupted images). </p>\n\n<p>Each time I am running a script it fails at the corrupted image (frustrating) with an error:</p>\n\n<blockquote>\n  <p>\"ValueError: The length of the pixel data in the dataset (153710 bytes) doesn't match the expected length (524288 bytes). The dataset may be corrupted or there may be an issue with the pixel data handler.\"</p>\n</blockquote>\n\n<p>Do you have any idea how to get rid of all corrupted images at once? </p>",
      "rawMarkdown": "Hi guys, \n\nthere is a problem with my train data set ... I have downloaded it several times without any progress (in case of corrupted images). \n\nEach time I am running a script it fails at the corrupted image (frustrating) with an error:\n\n&gt; \"ValueError: The length of the pixel data in the dataset (153710 bytes) doesn't match the expected length (524288 bytes). The dataset may be corrupted or there may be an issue with the pixel data handler.\"\n\nDo you have any idea how to get rid of all corrupted images at once? \n\n",
      "votes": 8
    },
    {
      "id": 631936,
      "postDate": "2019-09-23T01:18:55.663Z",
      "content": "<p><code>\nfor img in images:\n    try:\n         load/image process code\n    except ValueError:\n        continue\n</code></p>",
      "rawMarkdown": "```\nfor img in images:\n    try:\n         load/image process code\n    except ValueError:\n        continue\n```",
      "votes": 4
    },
    {
      "id": 631259,
      "postDate": "2019-09-21T17:37:46.113Z",
      "content": "<p>there is only one corrupted image. you can just delete it from the csv file</p>",
      "rawMarkdown": "there is only one corrupted image. you can just delete it from the csv file",
      "votes": 4,
      "replies": [
        {
          "id": 640177,
          "postDate": "2019-10-03T23:17:34.287Z",
          "content": "<p>Do you know which image it is?</p>",
          "rawMarkdown": "Do you know which image it is?",
          "votes": 2
        },
        {
          "id": 640181,
          "postDate": "2019-10-03T23:32:05.827Z",
          "content": "<p><code>ID_6431af929.dcm</code></p>",
          "rawMarkdown": "`ID_6431af929.dcm`",
          "votes": 10
        },
        {
          "id": 640597,
          "postDate": "2019-10-04T05:52:40.157Z",
          "content": "<p>Thank you very much for mentioning the particular image. </p>",
          "rawMarkdown": "Thank you very much for mentioning the particular image. "
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 631936,
      "author_name": "Bo Peng",
      "author_url": "",
      "post_date": "2019-09-23T01:18:55.663000",
      "content": "<p><code>\nfor img in images:\n    try:\n         load/image process code\n    except ValueError:\n        continue\n</code></p>",
      "votes": 4,
      "replies": []
    },
    {
      "id": 631259,
      "author_name": "DatNT",
      "author_url": "",
      "post_date": "2019-09-21T17:37:46.113000",
      "content": "<p>there is only one corrupted image. you can just delete it from the csv file</p>",
      "votes": 4,
      "replies": [
        {
          "id": 640177,
          "author_name": "William Green",
          "author_url": "",
          "post_date": "2019-10-03T23:17:34.287000",
          "content": "<p>Do you know which image it is?</p>",
          "votes": 2,
          "replies": []
        },
        {
          "id": 640181,
          "author_name": "DrHB",
          "author_url": "",
          "post_date": "2019-10-03T23:32:05.827000",
          "content": "<p><code>ID_6431af929.dcm</code></p>",
          "votes": 10,
          "replies": []
        },
        {
          "id": 640597,
          "author_name": "Nandita Bhattacharya",
          "author_url": "",
          "post_date": "2019-10-04T05:52:40.157000",
          "content": "<p>Thank you very much for mentioning the particular image. </p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "631157": "Hi guys, \n\nthere is a problem with my train data set ... I have downloaded it several times without any progress (in case of corrupted images). \n\nEach time I am running a script it fails at the corrupted image (frustrating) with an error:\n\n&gt; \"ValueError: The length of the pixel data in the dataset (153710 bytes) doesn't match the expected length (524288 bytes). The dataset may be corrupted or there may be an issue with the pixel data handler.\"\n\nDo you have any idea how to get rid of all corrupted images at once? \n\n",
    "631936": "```\nfor img in images:\n    try:\n         load/image process code\n    except ValueError:\n        continue\n```",
    "631259": "there is only one corrupted image. you can just delete it from the csv file"
  }
}