{
  "id": 493500,
  "title": "Data Cleaning",
  "url": "/competitions/leash-BELKA/discussion/493500",
  "author_name": "",
  "post_date": "2024-04-13T18:36:50.123428100Z",
  "votes": 4,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Could the competition organizers provide some info on how the data has been cleaned/preprocessed? I'm curious if any attempt to identify partial/failed reactions or promiscuous binders were undertaken prior to binary encoding the counts data.</p>",
  "messages": [
    {
      "id": "2750584",
      "postDate": "04/13/2024 18:36:50",
      "content": "<p>Could the competition organizers provide some info on how the data has been cleaned/preprocessed? I'm curious if any attempt to identify partial/failed reactions or promiscuous binders were undertaken prior to binary encoding the counts data.</p>",
      "rawMarkdown": "Could the competition organizers provide some info on how the data has been cleaned/preprocessed? I'm curious if any attempt to identify partial/failed reactions or promiscuous binders were undertaken prior to binary encoding the counts data.",
      "votes": null
    },
    {
      "id": "2750855",
      "postDate": "04/13/2024 21:27:44",
      "content": "<p>partial/failed reactions:<br>\nWe, unfortunately, do not have yield data for the library that makes up the training set.</p>\n<p>promiscuous binders:<br>\nIn addition to the screens that went into the dataset, we screened in several control conditions. Our binding labeling compares the replicates of measurements to the replicates of control conditions to determine binding.</p>",
      "rawMarkdown": "partial/failed reactions:\nWe, unfortunately, do not have yield data for the library that makes up the training set.\n\npromiscuous binders:\nIn addition to the screens that went into the dataset, we screened in several control conditions. Our binding labeling compares the replicates of measurements to the replicates of control conditions to determine binding.",
      "votes": null
    },
    {
      "id": "2750887",
      "postDate": "04/13/2024 22:00:01",
      "content": "<p>Thanks <a href=\"https://www.kaggle.com/andrewdblevins\" target=\"_blank\">@andrewdblevins</a> !</p>",
      "rawMarkdown": "Thanks @andrewdblevins !",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2750855,
      "author_name": "andrewdblevins",
      "author_url": "",
      "post_date": "04/13/2024 21:27:44",
      "content": "<p>partial/failed reactions:<br>\nWe, unfortunately, do not have yield data for the library that makes up the training set.</p>\n<p>promiscuous binders:<br>\nIn addition to the screens that went into the dataset, we screened in several control conditions. Our binding labeling compares the replicates of measurements to the replicates of control conditions to determine binding.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2750887,
          "author_name": "chemdatafarmer",
          "author_url": "",
          "post_date": "04/13/2024 22:00:01",
          "content": "<p>Thanks <a href=\"https://www.kaggle.com/andrewdblevins\" target=\"_blank\">@andrewdblevins</a> !</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2750584": "Could the competition organizers provide some info on how the data has been cleaned/preprocessed? I'm curious if any attempt to identify partial/failed reactions or promiscuous binders were undertaken prior to binary encoding the counts data.",
    "2750855": "partial/failed reactions:\nWe, unfortunately, do not have yield data for the library that makes up the training set.\n\npromiscuous binders:\nIn addition to the screens that went into the dataset, we screened in several control conditions. Our binding labeling compares the replicates of measurements to the replicates of control conditions to determine binding.",
    "2750887": "Thanks @andrewdblevins !"
  },
  "source": "meta"
}