{
  "id": 600582,
  "title": "Why 512 Features for 256 Electrodes?",
  "url": "/competitions/brain-to-text-25/discussion/600582",
  "author_name": "",
  "post_date": "2025-08-23T17:42:21.233677900Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Thank you for the dataset and the exciting competition! The docs mention 256 electrodes, but data[trial_id][\"input_features\"][()].T has 512 columns. Does this mean there are two features per electrode (e.g., different preprocessing like threshold and sbp)?</p>\n<p>I also saw in Paul Kamau’s notebook the following mapping:</p>\n<p>FEATURE_MAP = {<br>\n    \"ventral_6v_thresh\": list(range(0, 64)),<br>\n    \"area_4_thresh\": list(range(64, 128)),<br>\n    \"area_55b_thresh\": list(range(128, 192)),<br>\n    \"dorsal_6v_thresh\": list(range(192, 256)),<br>\n    \"ventral_6v_sbp\": list(range(256, 320)),<br>\n    \"area_4_sbp\": list(range(320, 384)),<br>\n    \"area_55b_sbp\": list(range(384, 448)),<br>\n    \"dorsal_6v_sbp\": list(range(448, 512)),<br>\n}</p>\n<p>Could you please confirm if this is the correct electrode-to-feature mapping/order?<br>\nThank you!</p>",
  "messages": [
    {
      "id": "3273973",
      "postDate": "08/23/2025 17:42:21",
      "content": "<p>Thank you for the dataset and the exciting competition! The docs mention 256 electrodes, but data[trial_id][\"input_features\"][()].T has 512 columns. Does this mean there are two features per electrode (e.g., different preprocessing like threshold and sbp)?</p>\n<p>I also saw in Paul Kamau’s notebook the following mapping:</p>\n<p>FEATURE_MAP = {<br>\n    \"ventral_6v_thresh\": list(range(0, 64)),<br>\n    \"area_4_thresh\": list(range(64, 128)),<br>\n    \"area_55b_thresh\": list(range(128, 192)),<br>\n    \"dorsal_6v_thresh\": list(range(192, 256)),<br>\n    \"ventral_6v_sbp\": list(range(256, 320)),<br>\n    \"area_4_sbp\": list(range(320, 384)),<br>\n    \"area_55b_sbp\": list(range(384, 448)),<br>\n    \"dorsal_6v_sbp\": list(range(448, 512)),<br>\n}</p>\n<p>Could you please confirm if this is the correct electrode-to-feature mapping/order?<br>\nThank you!</p>",
      "rawMarkdown": "Thank you for the dataset and the exciting competition! The docs mention 256 electrodes, but data[trial_id][\"input_features\"][()].T has 512 columns. Does this mean there are two features per electrode (e.g., different preprocessing like threshold and sbp)?\n\nI also saw in Paul Kamau’s notebook the following mapping:\n\nFEATURE_MAP = {\n    \"ventral_6v_thresh\": list(range(0, 64)),\n    \"area_4_thresh\": list(range(64, 128)),\n    \"area_55b_thresh\": list(range(128, 192)),\n    \"dorsal_6v_thresh\": list(range(192, 256)),\n    \"ventral_6v_sbp\": list(range(256, 320)),\n    \"area_4_sbp\": list(range(320, 384)),\n    \"area_55b_sbp\": list(range(384, 448)),\n    \"dorsal_6v_sbp\": list(range(448, 512)),\n}\n\nCould you please confirm if this is the correct electrode-to-feature mapping/order?\nThank you!",
      "votes": null
    },
    {
      "id": "3274430",
      "postDate": "08/24/2025 18:40:46",
      "content": "<p>Yes, each electrode has two features: threshold crossings (-4.5 RMS) and spike band power. The \"FEATURE_MAP\" order that you listed is correct. FYI, this information is also listed on the data tab of this competition: <a href=\"https://www.kaggle.com/competitions/brain-to-text-25/data\" target=\"_blank\">https://www.kaggle.com/competitions/brain-to-text-25/data</a></p>",
      "rawMarkdown": "Yes, each electrode has two features: threshold crossings (-4.5 RMS) and spike band power. The \"FEATURE_MAP\" order that you listed is correct. FYI, this information is also listed on the data tab of this competition: https://www.kaggle.com/competitions/brain-to-text-25/data",
      "votes": null
    },
    {
      "id": "3278924",
      "postDate": "08/30/2025 17:58:58",
      "content": "<p>Thanks so much for confirming, Nick! That clears it up. I must have overlooked that detail on the data tab.</p>",
      "rawMarkdown": "Thanks so much for confirming, Nick! That clears it up. I must have overlooked that detail on the data tab.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3274430,
      "author_name": "notnickc",
      "author_url": "",
      "post_date": "08/24/2025 18:40:46",
      "content": "<p>Yes, each electrode has two features: threshold crossings (-4.5 RMS) and spike band power. The \"FEATURE_MAP\" order that you listed is correct. FYI, this information is also listed on the data tab of this competition: <a href=\"https://www.kaggle.com/competitions/brain-to-text-25/data\" target=\"_blank\">https://www.kaggle.com/competitions/brain-to-text-25/data</a></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 3278924,
      "author_name": "eiix000",
      "author_url": "",
      "post_date": "08/30/2025 17:58:58",
      "content": "<p>Thanks so much for confirming, Nick! That clears it up. I must have overlooked that detail on the data tab.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3273973": "Thank you for the dataset and the exciting competition! The docs mention 256 electrodes, but data[trial_id][\"input_features\"][()].T has 512 columns. Does this mean there are two features per electrode (e.g., different preprocessing like threshold and sbp)?\n\nI also saw in Paul Kamau’s notebook the following mapping:\n\nFEATURE_MAP = {\n    \"ventral_6v_thresh\": list(range(0, 64)),\n    \"area_4_thresh\": list(range(64, 128)),\n    \"area_55b_thresh\": list(range(128, 192)),\n    \"dorsal_6v_thresh\": list(range(192, 256)),\n    \"ventral_6v_sbp\": list(range(256, 320)),\n    \"area_4_sbp\": list(range(320, 384)),\n    \"area_55b_sbp\": list(range(384, 448)),\n    \"dorsal_6v_sbp\": list(range(448, 512)),\n}\n\nCould you please confirm if this is the correct electrode-to-feature mapping/order?\nThank you!",
    "3274430": "Yes, each electrode has two features: threshold crossings (-4.5 RMS) and spike band power. The \"FEATURE_MAP\" order that you listed is correct. FYI, this information is also listed on the data tab of this competition: https://www.kaggle.com/competitions/brain-to-text-25/data",
    "3278924": "Thanks so much for confirming, Nick! That clears it up. I must have overlooked that detail on the data tab."
  },
  "source": "meta"
}