{
  "id": 15628,
  "title": "Uneven lengths of events?",
  "url": "/competitions/grasp-and-lift-eeg-detection/discussion/15628",
  "author_name": "",
  "post_date": "2015-07-29T20:58:26.857Z",
  "votes": 1,
  "comment_count": 2,
  "views": 431,
  "content": "<p>Is it just me or there is something wrong with the subj7_series8_events.csv? I was getting some errors in my scripts because there seems to be much more ones in the second column than in the fourth.</p>\n\n<p>You can check it yourself with awk:</p>\n\n<blockquote>\n  <p>awk -F, 'BEGIN{ cta = 0; ctb = 0 } { if ($4 &gt; 0) { cta += 1 }; if ($2 &gt; 0) { ctb += 1 } } END{ print cta, ctb }' subj7_series8_events.csv</p>\n</blockquote>\n\n<p>I'm getting the following output:</p>\n\n<blockquote>\n  <p>4528 5101</p>\n</blockquote>\n\n<p>Which is not only strange because the numbers are so different, but also they aren't divisible by 150. Did anybody had such problems too? Have I missed something in the description?</p>",
  "messages": [
    {
      "id": "87484",
      "postDate": "07/29/2015 20:58:26",
      "content": "<p>Is it just me or there is something wrong with the subj7_series8_events.csv? I was getting some errors in my scripts because there seems to be much more ones in the second column than in the fourth.</p>\n\n<p>You can check it yourself with awk:</p>\n\n<blockquote>\n  <p>awk -F, 'BEGIN{ cta = 0; ctb = 0 } { if ($4 &gt; 0) { cta += 1 }; if ($2 &gt; 0) { ctb += 1 } } END{ print cta, ctb }' subj7_series8_events.csv</p>\n</blockquote>\n\n<p>I'm getting the following output:</p>\n\n<blockquote>\n  <p>4528 5101</p>\n</blockquote>\n\n<p>Which is not only strange because the numbers are so different, but also they aren't divisible by 150. Did anybody had such problems too? Have I missed something in the description?</p>",
      "rawMarkdown": "Is it just me or there is something wrong with the subj7_series8_events.csv? I was getting some errors in my scripts because there seems to be much more ones in the second column than in the fourth.\r\n\r\nYou can check it yourself with awk:\r\n\r\n> awk -F, 'BEGIN{ cta = 0; ctb = 0 } { if ($4 > 0) { cta += 1 }; if ($2 > 0) { ctb += 1 } } END{ print cta, ctb }' subj7_series8_events.csv\r\n\r\nI'm getting the following output:\r\n\r\n> 4528 5101\r\n\r\nWhich is not only strange because the numbers are so different, but also they aren't divisible by 150. Did anybody had such problems too? Have I missed something in the description?",
      "votes": null
    },
    {
      "id": "88558",
      "postDate": "08/05/2015 17:52:28",
      "content": "<p>I had the same problem,  the number of samples of BothStartLoadPhase is not divisible by 150.\nIn the topic Announcement: Error in dataset @mibel is reporting the same problem with subject7. I hope it will be fixed.</p>",
      "rawMarkdown": "I had the same problem,  the number of samples of BothStartLoadPhase is not divisible by 150.\r\nIn the topic Announcement: Error in dataset @mibel is reporting the same problem with subject7. I hope it will be fixed.",
      "votes": null
    },
    {
      "id": "89092",
      "postDate": "08/11/2015 10:01:49",
      "content": "<p>There are a few of wrongly placed BothStartLoadPhase events in the dataset, mainly for Subject 7. \nThis happened because the start of the load phase is particularly difficult to detect. \nIn 7 cases the misplaced events overlap with the same event in the next trial, which reduces the total number of ones in that column (can be seen in subj7_series7 and subj7_series8). \n(As a side note: Your script seems to also count the first comment row as a one. The numbers should really be 4527 and 5100, which means the second number is actually divisible by 150. )</p>\n\n<p>The error only affects the second event (BothStartLoadPhase) and only a very small fraction of the trials, so we decided not to do anything about it. </p>",
      "rawMarkdown": "There are a few of wrongly placed BothStartLoadPhase events in the dataset, mainly for Subject 7. \r\nThis happened because the start of the load phase is particularly difficult to detect. \r\nIn 7 cases the misplaced events overlap with the same event in the next trial, which reduces the total number of ones in that column (can be seen in subj7_series7 and subj7_series8). \r\n(As a side note: Your script seems to also count the first comment row as a one. The numbers should really be 4527 and 5100, which means the second number is actually divisible by 150. )\r\n\r\nThe error only affects the second event (BothStartLoadPhase) and only a very small fraction of the trials, so we decided not to do anything about it.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 88558,
      "author_name": "alexandrasmoloiu",
      "author_url": "",
      "post_date": "08/05/2015 17:52:28",
      "content": "<p>I had the same problem,  the number of samples of BothStartLoadPhase is not divisible by 150.\nIn the topic Announcement: Error in dataset @mibel is reporting the same problem with subject7. I hope it will be fixed.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 89092,
      "author_name": "qwlouse",
      "author_url": "",
      "post_date": "08/11/2015 10:01:49",
      "content": "<p>There are a few of wrongly placed BothStartLoadPhase events in the dataset, mainly for Subject 7. \nThis happened because the start of the load phase is particularly difficult to detect. \nIn 7 cases the misplaced events overlap with the same event in the next trial, which reduces the total number of ones in that column (can be seen in subj7_series7 and subj7_series8). \n(As a side note: Your script seems to also count the first comment row as a one. The numbers should really be 4527 and 5100, which means the second number is actually divisible by 150. )</p>\n\n<p>The error only affects the second event (BothStartLoadPhase) and only a very small fraction of the trials, so we decided not to do anything about it. </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "87484": "Is it just me or there is something wrong with the subj7_series8_events.csv? I was getting some errors in my scripts because there seems to be much more ones in the second column than in the fourth.\r\n\r\nYou can check it yourself with awk:\r\n\r\n> awk -F, 'BEGIN{ cta = 0; ctb = 0 } { if ($4 > 0) { cta += 1 }; if ($2 > 0) { ctb += 1 } } END{ print cta, ctb }' subj7_series8_events.csv\r\n\r\nI'm getting the following output:\r\n\r\n> 4528 5101\r\n\r\nWhich is not only strange because the numbers are so different, but also they aren't divisible by 150. Did anybody had such problems too? Have I missed something in the description?",
    "88558": "I had the same problem,  the number of samples of BothStartLoadPhase is not divisible by 150.\r\nIn the topic Announcement: Error in dataset @mibel is reporting the same problem with subject7. I hope it will be fixed.",
    "89092": "There are a few of wrongly placed BothStartLoadPhase events in the dataset, mainly for Subject 7. \r\nThis happened because the start of the load phase is particularly difficult to detect. \r\nIn 7 cases the misplaced events overlap with the same event in the next trial, which reduces the total number of ones in that column (can be seen in subj7_series7 and subj7_series8). \r\n(As a side note: Your script seems to also count the first comment row as a one. The numbers should really be 4527 and 5100, which means the second number is actually divisible by 150. )\r\n\r\nThe error only affects the second event (BothStartLoadPhase) and only a very small fraction of the trials, so we decided not to do anything about it."
  },
  "source": "meta"
}