{
  "id": 551904,
  "title": "Synchronisation between left and right arm acceleration data and downsampling ",
  "url": "/competitions/2ia-2024-hackathon/discussion/551904",
  "author_name": "Nicolas Sutton-Charani",
  "post_date": "2024-12-16T11:28:31.406000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>By now you should have noticed that the measurement times are not exactly the same in the left and right arms acceleration data. The first thing you need to do is thus to build a synchronised acceleration dataset. To do so, one solution is to estimate the value of an arm data on the other arm times data by linear interpolation for example (e.g. 'approx' R function).</p>\n<p>Once this is done, I suggest you to downsample your dataset in order to be able to make fast computations, but you have to keep in mind that down sampling necessary involves information loss so, once you manage to get prediction from your downsample dataset, recompute your prediction on your complete dataset (without downsampling) to get the best prediction you can achieve with your method.</p>",
  "messages": [
    {
      "id": 3073341,
      "postDate": "2024-12-16T11:28:31.407Z",
      "content": "<p>By now you should have noticed that the measurement times are not exactly the same in the left and right arms acceleration data. The first thing you need to do is thus to build a synchronised acceleration dataset. To do so, one solution is to estimate the value of an arm data on the other arm times data by linear interpolation for example (e.g. 'approx' R function).</p>\n<p>Once this is done, I suggest you to downsample your dataset in order to be able to make fast computations, but you have to keep in mind that down sampling necessary involves information loss so, once you manage to get prediction from your downsample dataset, recompute your prediction on your complete dataset (without downsampling) to get the best prediction you can achieve with your method.</p>",
      "rawMarkdown": "By now you should have noticed that the measurement times are not exactly the same in the left and right arms acceleration data. The first thing you need to do is thus to build a synchronised acceleration dataset. To do so, one solution is to estimate the value of an arm data on the other arm times data by linear interpolation for example (e.g. 'approx' R function).\n\nOnce this is done, I suggest you to downsample your dataset in order to be able to make fast computations, but you have to keep in mind that down sampling necessary involves information loss so, once you manage to get prediction from your downsample dataset, recompute your prediction on your complete dataset (without downsampling) to get the best prediction you can achieve with your method."
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "3073341": "By now you should have noticed that the measurement times are not exactly the same in the left and right arms acceleration data. The first thing you need to do is thus to build a synchronised acceleration dataset. To do so, one solution is to estimate the value of an arm data on the other arm times data by linear interpolation for example (e.g. 'approx' R function).\n\nOnce this is done, I suggest you to downsample your dataset in order to be able to make fast computations, but you have to keep in mind that down sampling necessary involves information loss so, once you manage to get prediction from your downsample dataset, recompute your prediction on your complete dataset (without downsampling) to get the best prediction you can achieve with your method."
  }
}