{
  "id": 468459,
  "title": "EEG parquets not in train.csv",
  "url": "/competitions/hms-harmful-brain-activity-classification/discussion/468459",
  "author_name": "",
  "post_date": "2024-01-16T18:24:38.690294400Z",
  "votes": 4,
  "comment_count": 2,
  "views": 0,
  "content": "<p>There are 211 parquet files in the <code>train_eegs</code> directory that are not listed in <code>train.csv</code>.<br>\n<a href=\"https://www.kaggle.com/code/soshitanabe/notebook0e27efe57f\" target=\"_blank\">https://www.kaggle.com/code/soshitanabe/notebook0e27efe57f</a></p>\n<p>If these files have already been annotated but are missing from <code>train.csv</code>, the csv should be revised and updated accordingly. If these files have not been annotated (<strong>unlabled</strong>), is it possible to disclose which spectrogram_id corresponds to these eeg_ids?</p>",
  "messages": [
    {
      "id": "2604905",
      "postDate": "01/16/2024 18:24:38",
      "content": "<p>There are 211 parquet files in the <code>train_eegs</code> directory that are not listed in <code>train.csv</code>.<br>\n<a href=\"https://www.kaggle.com/code/soshitanabe/notebook0e27efe57f\" target=\"_blank\">https://www.kaggle.com/code/soshitanabe/notebook0e27efe57f</a></p>\n<p>If these files have already been annotated but are missing from <code>train.csv</code>, the csv should be revised and updated accordingly. If these files have not been annotated (<strong>unlabled</strong>), is it possible to disclose which spectrogram_id corresponds to these eeg_ids?</p>",
      "rawMarkdown": "There are 211 parquet files in the `train_eegs` directory that are not listed in `train.csv`.\nhttps://www.kaggle.com/code/soshitanabe/notebook0e27efe57f\n\nIf these files have already been annotated but are missing from `train.csv`, the csv should be revised and updated accordingly. If these files have not been annotated (**unlabled**), is it possible to disclose which spectrogram_id corresponds to these eeg_ids?",
      "votes": null
    },
    {
      "id": "2604921",
      "postDate": "01/16/2024 18:32:58",
      "content": "<p><a href=\"https://www.kaggle.com/soshitanabe\" target=\"_blank\">@soshitanabe</a> go through this discussion <a href=\"https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/467058\" target=\"_blank\">is eeg_ids missing in train.csv but available in train_eegs folder? [Solved]</a></p>\n<blockquote>\n  <p><em>from Kaggle team response</em> Those are files without paired spectrograms where I missed some cleanup. by <a href=\"https://www.kaggle.com/sohier\" target=\"_blank\">@sohier</a></p>\n</blockquote>\n<hr>\n<blockquote>\n  <p><em>option1</em> We can ignore these egg sequences, when using with spectrograms of Kaggle provided</p>\n</blockquote>\n<hr>\n<blockquote>\n  <p><em>option2</em> We can use in training using following <a href=\"https://www.kaggle.com/code/cdeotte/how-to-make-spectrogram-from-eeg\" target=\"_blank\">Notebook - How To Make Spectrogram from EEG</a> with pseudolabels</p>\n</blockquote>",
      "rawMarkdown": "soshitanabe go through this discussion [is eeg_ids missing in train.csv but available in train_eegs folder? [Solved]](https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/467058)\n\n> *from Kaggle team response* Those are files without paired spectrograms where I missed some cleanup. by @sohier\n\n----\n> *option1* We can ignore these egg sequences, when using with spectrograms of Kaggle provided\n\n----\n> *option2* We can use in training using following [Notebook - How To Make Spectrogram from EEG](https://www.kaggle.com/code/cdeotte/how-to-make-spectrogram-from-eeg) with pseudolabels",
      "votes": null
    },
    {
      "id": "2604943",
      "postDate": "01/16/2024 18:52:11",
      "content": "<p>Thank you for quick reply!</p>",
      "rawMarkdown": "Thank you for quick reply!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2604921,
      "author_name": "seshurajup",
      "author_url": "",
      "post_date": "01/16/2024 18:32:58",
      "content": "<p><a href=\"https://www.kaggle.com/soshitanabe\" target=\"_blank\">@soshitanabe</a> go through this discussion <a href=\"https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/467058\" target=\"_blank\">is eeg_ids missing in train.csv but available in train_eegs folder? [Solved]</a></p>\n<blockquote>\n  <p><em>from Kaggle team response</em> Those are files without paired spectrograms where I missed some cleanup. by <a href=\"https://www.kaggle.com/sohier\" target=\"_blank\">@sohier</a></p>\n</blockquote>\n<hr>\n<blockquote>\n  <p><em>option1</em> We can ignore these egg sequences, when using with spectrograms of Kaggle provided</p>\n</blockquote>\n<hr>\n<blockquote>\n  <p><em>option2</em> We can use in training using following <a href=\"https://www.kaggle.com/code/cdeotte/how-to-make-spectrogram-from-eeg\" target=\"_blank\">Notebook - How To Make Spectrogram from EEG</a> with pseudolabels</p>\n</blockquote>",
      "votes": null,
      "replies": [
        {
          "id": 2604943,
          "author_name": "soshitanabe",
          "author_url": "",
          "post_date": "01/16/2024 18:52:11",
          "content": "<p>Thank you for quick reply!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2604905": "There are 211 parquet files in the `train_eegs` directory that are not listed in `train.csv`.\nhttps://www.kaggle.com/code/soshitanabe/notebook0e27efe57f\n\nIf these files have already been annotated but are missing from `train.csv`, the csv should be revised and updated accordingly. If these files have not been annotated (**unlabled**), is it possible to disclose which spectrogram_id corresponds to these eeg_ids?",
    "2604921": "soshitanabe go through this discussion [is eeg_ids missing in train.csv but available in train_eegs folder? [Solved]](https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/467058)\n\n> *from Kaggle team response* Those are files without paired spectrograms where I missed some cleanup. by @sohier\n\n----\n> *option1* We can ignore these egg sequences, when using with spectrograms of Kaggle provided\n\n----\n> *option2* We can use in training using following [Notebook - How To Make Spectrogram from EEG](https://www.kaggle.com/code/cdeotte/how-to-make-spectrogram-from-eeg) with pseudolabels",
    "2604943": "Thank you for quick reply!"
  },
  "source": "meta"
}