{
  "id": 394531,
  "title": "Updated \"notype\" Data",
  "url": "/competitions/tlvmc-parkinsons-freezing-gait-prediction/discussion/394531",
  "author_name": "",
  "post_date": "2023-03-13T20:29:12.027925100Z",
  "votes": 18,
  "comment_count": 5,
  "views": 0,
  "content": "<p>Hi everyone,</p>\n<p>The <code>Event</code> column for series in the <code>train/notype</code> folder was missing event annotations. I've updated the dataset to include the correct labels.</p>\n<p>Thanks!</p>",
  "messages": [
    {
      "id": "2180466",
      "postDate": "03/13/2023 20:29:12",
      "content": "<p>Hi everyone,</p>\n<p>The <code>Event</code> column for series in the <code>train/notype</code> folder was missing event annotations. I've updated the dataset to include the correct labels.</p>\n<p>Thanks!</p>",
      "rawMarkdown": "Hi everyone,\n\nThe `Event` column for series in the `train/notype` folder was missing event annotations. I've updated the dataset to include the correct labels.\n\nThanks!",
      "votes": null
    },
    {
      "id": "2185836",
      "postDate": "03/17/2023 10:35:49",
      "content": "<p>Hello, I'm confused about the usage of tasks.csv. (e.g. the relationship between task column and 'StartHesitation,Turn,Walking')<br>\nCould you add some explanations?<br>\nBTW, 'Description of the task' not found.<br>\nThanks!</p>",
      "rawMarkdown": "Hello, I'm confused about the usage of tasks.csv. (e.g. the relationship between task column and 'StartHesitation,Turn,Walking')\nCould you add some explanations?\nBTW, 'Description of the task' not found.\nThanks!",
      "votes": null
    },
    {
      "id": "2185918",
      "postDate": "03/17/2023 12:09:21",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/wsdfjnn\" target=\"_blank\">@wsdfjnn</a> ,</p>\n<p>The <code>tasks.csv</code> file contains metadata about the experimental protocol for the DeFOG dataset. You can find more on the <a href=\"https://www.kaggle.com/competitions/tlvmc-parkinsons-freezing-gait-prediction/overview/additional-data-documentation\" target=\"_blank\">Additional Data Documentation Page</a>.</p>\n<p>Briefly, for the DeFOG dataset, events were only annotated while the subject was performing certain tasks (indicated by <code>task</code> equal to <code>true</code> in the data series). The type of task is described in <code>tasks.csv</code>.</p>\n<p>The description shouldn't be present. I'll remove it from the documentation. Thanks!</p>",
      "rawMarkdown": "Hi @wsdfjnn ,\n\nThe `tasks.csv` file contains metadata about the experimental protocol for the DeFOG dataset. You can find more on the [Additional Data Documentation Page](https://www.kaggle.com/competitions/tlvmc-parkinsons-freezing-gait-prediction/overview/additional-data-documentation).\n\nBriefly, for the DeFOG dataset, events were only annotated while the subject was performing certain tasks (indicated by `task` equal to `true` in the data series). The type of task is described in `tasks.csv`.\n\nThe description shouldn't be present. I'll remove it from the documentation. Thanks!",
      "votes": null
    },
    {
      "id": "2201461",
      "postDate": "03/29/2023 10:19:45",
      "content": "<p>Thanks! <br>\nMay I ask about the evaluation? How the mAP is calculated in this competion? Is it calculated by each event like image object detection or just calculated row by row like nomal tabular competion with an average_precision_score in sklearn?</p>",
      "rawMarkdown": "Thanks! \nMay I ask about the evaluation? How the mAP is calculated in this competion? Is it calculated by each event like image object detection or just calculated row by row like nomal tabular competion with an average_precision_score in sklearn?",
      "votes": null
    },
    {
      "id": "2201567",
      "postDate": "03/29/2023 12:11:50",
      "content": "<p>It's per class, equivalent to <code>average_precision_score</code> with <code>average='macro'</code>.</p>",
      "rawMarkdown": "It's per class, equivalent to `average_precision_score` with `average='macro'`.",
      "votes": null
    },
    {
      "id": "2201643",
      "postDate": "03/29/2023 13:15:28",
      "content": "<p>Got it. Thanks!</p>",
      "rawMarkdown": "Got it. Thanks!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2185836,
      "author_name": "wsdfjnn",
      "author_url": "",
      "post_date": "03/17/2023 10:35:49",
      "content": "<p>Hello, I'm confused about the usage of tasks.csv. (e.g. the relationship between task column and 'StartHesitation,Turn,Walking')<br>\nCould you add some explanations?<br>\nBTW, 'Description of the task' not found.<br>\nThanks!</p>",
      "votes": null,
      "replies": [
        {
          "id": 2185918,
          "author_name": "ryanholbrook",
          "author_url": "",
          "post_date": "03/17/2023 12:09:21",
          "content": "<p>Hi <a href=\"https://www.kaggle.com/wsdfjnn\" target=\"_blank\">@wsdfjnn</a> ,</p>\n<p>The <code>tasks.csv</code> file contains metadata about the experimental protocol for the DeFOG dataset. You can find more on the <a href=\"https://www.kaggle.com/competitions/tlvmc-parkinsons-freezing-gait-prediction/overview/additional-data-documentation\" target=\"_blank\">Additional Data Documentation Page</a>.</p>\n<p>Briefly, for the DeFOG dataset, events were only annotated while the subject was performing certain tasks (indicated by <code>task</code> equal to <code>true</code> in the data series). The type of task is described in <code>tasks.csv</code>.</p>\n<p>The description shouldn't be present. I'll remove it from the documentation. Thanks!</p>",
          "votes": null,
          "replies": [
            {
              "id": 2201461,
              "author_name": "wsdfjnn",
              "author_url": "",
              "post_date": "03/29/2023 10:19:45",
              "content": "<p>Thanks! <br>\nMay I ask about the evaluation? How the mAP is calculated in this competion? Is it calculated by each event like image object detection or just calculated row by row like nomal tabular competion with an average_precision_score in sklearn?</p>",
              "votes": null,
              "replies": [
                {
                  "id": 2201567,
                  "author_name": "ryanholbrook",
                  "author_url": "",
                  "post_date": "03/29/2023 12:11:50",
                  "content": "<p>It's per class, equivalent to <code>average_precision_score</code> with <code>average='macro'</code>.</p>",
                  "votes": null,
                  "replies": [
                    {
                      "id": 2201643,
                      "author_name": "wsdfjnn",
                      "author_url": "",
                      "post_date": "03/29/2023 13:15:28",
                      "content": "<p>Got it. Thanks!</p>",
                      "votes": null,
                      "replies": []
                    }
                  ]
                }
              ]
            }
          ]
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2180466": "Hi everyone,\n\nThe `Event` column for series in the `train/notype` folder was missing event annotations. I've updated the dataset to include the correct labels.\n\nThanks!",
    "2185836": "Hello, I'm confused about the usage of tasks.csv. (e.g. the relationship between task column and 'StartHesitation,Turn,Walking')\nCould you add some explanations?\nBTW, 'Description of the task' not found.\nThanks!",
    "2185918": "Hi @wsdfjnn ,\n\nThe `tasks.csv` file contains metadata about the experimental protocol for the DeFOG dataset. You can find more on the [Additional Data Documentation Page](https://www.kaggle.com/competitions/tlvmc-parkinsons-freezing-gait-prediction/overview/additional-data-documentation).\n\nBriefly, for the DeFOG dataset, events were only annotated while the subject was performing certain tasks (indicated by `task` equal to `true` in the data series). The type of task is described in `tasks.csv`.\n\nThe description shouldn't be present. I'll remove it from the documentation. Thanks!",
    "2201461": "Thanks! \nMay I ask about the evaluation? How the mAP is calculated in this competion? Is it calculated by each event like image object detection or just calculated row by row like nomal tabular competion with an average_precision_score in sklearn?",
    "2201567": "It's per class, equivalent to `average_precision_score` with `average='macro'`.",
    "2201643": "Got it. Thanks!"
  },
  "source": "meta"
}