{
  "id": 18117,
  "title": "SAX directories with different numbers of files",
  "url": "/competitions/second-annual-data-science-bowl/discussion/18117",
  "author_name": "afterhours",
  "post_date": "2015-12-24T21:18:37.570000",
  "votes": 0,
  "comment_count": 0,
  "views": 594,
  "content": "<p>The large majority of directories have 30 files per sax run in them. However, some have different numbers. For example, study 416 of the training set has sax directories with 25, 22, 30 and 23 files. These files seem to be contiguous from 0 (e.g., short rather than patchy).</p>\n\n<p>What are other people doing about this inconsistency? Skipping? Resizing?</p>\n\n<p>My plan had been to create a 'volume' of data based on stitching all the sax arrays into a single 3d array per sax run. However I'm not sure how to handle the different run lengths.</p>\n\n<p>Maybe stitching them together is the wrong approach, but I thought I'd just ask what others are doing.</p>",
  "messages": [
    {
      "id": 102710,
      "postDate": "2015-12-24T21:18:37.570Z",
      "content": "<p>The large majority of directories have 30 files per sax run in them. However, some have different numbers. For example, study 416 of the training set has sax directories with 25, 22, 30 and 23 files. These files seem to be contiguous from 0 (e.g., short rather than patchy).</p>\n\n<p>What are other people doing about this inconsistency? Skipping? Resizing?</p>\n\n<p>My plan had been to create a 'volume' of data based on stitching all the sax arrays into a single 3d array per sax run. However I'm not sure how to handle the different run lengths.</p>\n\n<p>Maybe stitching them together is the wrong approach, but I thought I'd just ask what others are doing.</p>",
      "rawMarkdown": "The large majority of directories have 30 files per sax run in them. However, some have different numbers. For example, study 416 of the training set has sax directories with 25, 22, 30 and 23 files. These files seem to be contiguous from 0 (e.g., short rather than patchy).\r\n\r\nWhat are other people doing about this inconsistency? Skipping? Resizing?\r\n\r\nMy plan had been to create a 'volume' of data based on stitching all the sax arrays into a single 3d array per sax run. However I'm not sure how to handle the different run lengths.\r\n\r\nMaybe stitching them together is the wrong approach, but I thought I'd just ask what others are doing."
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "102710": "The large majority of directories have 30 files per sax run in them. However, some have different numbers. For example, study 416 of the training set has sax directories with 25, 22, 30 and 23 files. These files seem to be contiguous from 0 (e.g., short rather than patchy).\r\n\r\nWhat are other people doing about this inconsistency? Skipping? Resizing?\r\n\r\nMy plan had been to create a 'volume' of data based on stitching all the sax arrays into a single 3d array per sax run. However I'm not sure how to handle the different run lengths.\r\n\r\nMaybe stitching them together is the wrong approach, but I thought I'd just ask what others are doing."
  }
}